Sensing system and method for providing optimized exercise guidance to healthy users and athletes using continuously monitored analyte data

Abstract

Certain aspects of the present disclosure provide a monitoring system comprising: a continuous analyte sensor configured to generate a first set of analyte measurements associated with an analyte level of a patient; and a sensor electronics module coupled to the continuous analyte sensor and configured to receive and process the first set of analyte measurements.

Description

[0001] Cross-references to related applications This application claims priority and interest in U.S. Provisional Application No. 63 / 496,343, filed April 14, 2023, which is assigned to the assignee of this application and is expressly incorporated herein by reference in its entirety, as fully set forth below and used for all applicable purposes. Background Technology

[0002] Exercise and an active lifestyle are important for managing weight, reducing disease risk, improving fitness, and strengthening bones and muscles. However, simply maintaining an active lifestyle may not help an individual achieve their health or fitness goals, such as improvements in metabolic fitness, peak performance, and weight loss, if the exercise sessions are not optimized for the individual. On the other hand, consistent exercise across a series of individualized exercise sessions can improve metabolic fitness, peak performance, and weight loss, which can increase a user's lifespan and quality of life.

[0003] A host's metabolic fitness can be measured by the body's ability to metabolize lactate in response to exercise and / or by the body's (e.g., liver, kidneys, and skeletal muscles) ability to do so at rest, during, or after exercise. The more metabolically flexible a host's liver and muscles are, the better they can metabolize blood lactate at normal rates at rest and after exercise. This directly contributes to maintaining a healthy weight, improving peak performance, and improving liver health and function. Regular exercise sessions can improve the metabolic flexibility of a host's liver, kidneys, skeletal muscles, and bones.

[0004] Lactate and other analytes can be used to track metabolic fitness over time. Various methods are used to measure and analyze lactate, including central laboratory methods, near-patient blood gas analysis, and analysis using portable point-of-care (POC) handheld devices. Central laboratory methods involve transporting a patient's blood sample to the laboratory via a porter or endotracheal system. Unfortunately, central laboratory methods often suffer from excessively long times between blood collection and when clinicians are aware of the test results, potentially delaying clinical decisions. Therefore, lactate levels are typically tracked in users with specific disease states (e.g., liver disease, heart failure, etc.) or those receiving treatment for acute conditions or at risk of developing acute conditions (e.g., sepsis, heart attack, etc.).

[0005] However, with advancements in POC technology, near-patient benchtop blood gas analyzers have become available for lactate testing. However, these devices are not portable, and their availability is typically limited to a single dedicated unit, such as the emergency room (ED) or intensive care unit (ICU). Furthermore, when samples are drawn outside these primary units, sample turnaround time for test results can be affected by delays in transporting them to the ED or ICU.

[0006] For this reason, small handheld devices, much like blood glucose meters, are already available for lactate measurement and analysis. Currently, users can carry self-monitoring lactate monitors, which typically require pricking their finger to measure their lactate levels. However, given the inconvenience associated with the traditional finger-pricking method, users may find it difficult to perform lactate measurements, especially during or immediately after exercise. Therefore, this lack of user willingness to use the devices may prevent users from utilizing lactate measurements to achieve optimal results from their workouts.

[0007] Specifically, while handheld devices providing semi-continuous lactate measurement can effectively optimize workout phases to achieve user goals, users may abandon using the device due to the inconvenience of finger punctures, and / or may be unable to reach the handheld device for workout optimization. Therefore, users may not be able to monitor their fitness response to increases or decreases during workouts. While monitoring the effectiveness of workout phases is not critical, monitoring a user's metabolic response to exercise can help inform their overall health and fitness over time and can facilitate beneficial adjustments to future workout phases, thereby improving the user's overall health. Attached Figure Description

[0008] To gain a more detailed understanding of the features of this disclosure, reference can be made to some aspects illustrated in the accompanying drawings for a more specific description of the aspects briefly summarized above. However, it should be noted that the drawings illustrate only some typical aspects of this disclosure and should therefore not be considered as limiting its scope, as the description may acknowledge other equally valid aspects.

[0009] Figure 1 Examples of a sample treatment management system used in conjunction with embodiments of this disclosure are illustrated.

[0010] Figure 2 The diagram is a conceptual illustration of an example continuous analyte monitoring system according to certain embodiments of the present disclosure, the continuous analyte monitoring system including an example continuous analyte sensor having sensor electronics.

[0011] Figure 3 Examples of certain embodiments of the present disclosure are provided. Figure 1 The treatment management system uses example inputs and example metrics calculated based on these inputs.

[0012] Figure 4 This is a flowchart depicting an example method for classifying users as healthy users, athletes, or metabolically impaired users according to certain embodiments of this disclosure.

[0013] Figures 5A to 5C Example methods for optimizing exercise phases and providing feedback to healthy users according to certain embodiments of this disclosure are described.

[0014] Figures 6A to 6C Example methods for optimizing training phases and providing feedback to athletes according to certain embodiments of this disclosure are described.

[0015] Figure 7 A flowchart illustrating an example method for providing treatment management support using a continuous analyte monitoring system configured to continuously measure at least lactate levels, according to certain embodiments of this disclosure.

[0016] Figure 8 This is a flowchart depicting a method for training a machine learning model to provide exercise for a user, according to certain embodiments of this disclosure.

[0017] Figure 9 It describes a configuration for performing certain embodiments according to this disclosure. Figures 4 to 7 A block diagram of the computing device in operation.

[0018] Figures 10A to 10B Exemplary enzyme domain configurations for a continuous multianalyte sensor according to certain embodiments of the present disclosure are depicted.

[0019] Figures 10C to 10D Exemplary enzyme domain configurations for a continuous multianalyte sensor according to certain embodiments of the present disclosure are depicted.

[0020] Figure 10E Exemplary enzyme domain configurations for a continuous multianalyte sensor according to certain embodiments of the present disclosure are depicted.

[0021] Figures 11A to 11B An alternative view of an exemplary dual-electrode enzyme domain configuration for a continuous multianalyte sensor according to certain embodiments of the present disclosure is depicted.

[0022] Figures 11C to 11D An alternative view of an exemplary dual-electrode enzyme domain configuration for a continuous multianalyte sensor according to certain embodiments of the present disclosure is depicted.

[0023] Figure 11E An exemplary dual-electrode configuration for a continuous multianalyte sensor according to certain embodiments of the present disclosure is depicted.

[0024] Figure 12A Exemplary enzyme domain configurations for a continuous multianalyte sensor according to certain embodiments of the present disclosure are depicted.

[0025] Figures 12B to 12C Alternative exemplary enzyme domain configurations for a continuous multianalyte sensor according to certain embodiments of the present disclosure are depicted.

[0026] Figure 13 Exemplary enzyme domain configurations for a continuous multianalyte sensor according to certain embodiments of the present disclosure are depicted.

[0027] Figures 14A to 14D Alternative views of exemplary dual-electrode enzyme domain configurations G1-G4 for a continuous multianalyte sensor according to certain embodiments of the present disclosure are depicted.

[0028] Figure 15A An exemplary lactic acid sensor according to certain embodiments of the present disclosure is depicted.

[0029] Figure 15B Depicting certain embodiments according to this disclosure Figure 13 A cross-sectional view of the electroactive region of an example lactic acid sensor.

[0030] Figures 16A to 16C Exemplary embodiments of a continuous analyte sensor system, implemented as a wearable lactic acid sensor according to certain embodiments of the present disclosure, are described.

[0031] Figures 17A to 17G A one-sided coplanar analyzer sensor assembly according to certain embodiments of the present disclosure is depicted.

[0032] For ease of understanding, the same reference numerals are used where possible to denote the same elements common to the figures. Elements disclosed in one aspect are intended to be usefully used in other aspects without specific description. Detailed Implementation

[0033] Effective exercise routines can be associated with improved metabolic fitness. Exercise routines and physical activity can be effectively used to improve overall metabolic fitness, which can be measured by reducing body lactate concentration and / or improving the body's ability to clear lactate during exercise in response to various exercise intensities. However, a single type of exercise is not effective for all users, and determining the effect of a particular type of exercise performed by a user is often impractical. For example, non-analyte parameters (e.g., heart rate, respiratory rate) are insufficient to determine a user's metabolic response to exercise.

[0034] Furthermore, users can have a variety of different exercise goals, whether they are metabolically underachieving users aiming to improve metabolic fitness, healthy users with weight loss goals or goals to improve metabolic fitness and / or athletic performance over time, or athletes aiming to achieve peak performance and / or peak metabolic fitness. Therefore, certain types of exercise may be more effective for specific users depending on their physiological / fitness status and goals. However, users may not be aware of the most suitable type of exercise for their corresponding physiological status and / or goals, and / or other exercise parameters (e.g., duration, frequency, intensity, etc.) for more effectively achieving such goals.

[0035] As an example, metabolically underperforming individuals can benefit from Zone 2 training (“Zone 2”) – a type of aerobic exercise – to achieve improved metabolic fitness. For instance, Zone 2 exercises may include walking, rowing, swimming, or light jogging. Typically, Zone 2 is the maximum exercise rate at which lactate production and clearance are in balance during an exercise phase and can indicate a user’s metabolic fitness and liver function. As described herein, an exercise phase is a period of elevated physiological activity associated with physical activity, which may include the time during exercise, the time after exercise when a user’s lactate levels remain elevated, or the time after exercise when a user experiences an increased metabolic rate (e.g., 24–48 hours). Specifically, regular Zone 2 training can help improve fitness, which can be measured by, for example, reducing baseline lactate over time, improving lactate clearance rate (e.g., a larger negative value as lactate returns to baseline levels), increasing fat loss, reducing resting heart rate, reducing blood pressure, improving insulin resistance, improving muscle mitochondrial function, increasing baseline metabolic rate at rest, and increasing endurance.

[0036] In another example, healthy users can benefit from high-intensity interval training (HIIT), or anaerobic exercise, to promote a long-term increase in energy expenditure. For example, HIIT workouts may include high-intensity plyometric exercises, or sprint running, followed by low-intensity rest periods. HIIT workouts help improve metabolic fitness by promoting energy expenditure over time, improving oxygen intake, lowering resting heart rate, lowering blood pressure, and improving resting metabolic rate.

[0037] In another example, athletes may benefit from resistance training, or activities involving exercising muscles or muscle groups against external resistance (e.g., weights or resistance bands), to increase muscle mass and thus improve athletic performance. Alternatively, healthy individuals may benefit from resistance training to increase long-term energy expenditure and fat oxidation. Similar to HIIT workouts, resistance training achieves increased long-term energy expenditure and fat oxidation because the body expends energy to repair muscles in the days following the resistance training phase. Therefore, healthy individuals can benefit from resistance training through improved long-term energy expenditure and fat oxidation.

[0038] In yet another example, athletes can benefit from primary Zone 2 training alongside some HIIT and / or resistance training (e.g., 80% Zone 2 training with 20% HIIT, or 70% Zone 2 training with 30% HIIT) to increase endurance and improve peak performance. For instance, consistent Zone 2 training can allow athletes to increase the intensity required to perform Zone 2 training in future training phases, thereby improving peak performance.

[0039] As discussed in this article, particularly for healthy users and athletes, regular physical activity is crucial for helping these users optimize their workout phases, improve metabolic fitness, and help achieve performance goals. However, improving metabolic fitness through exercise presents many challenges for users, as a confluence of factors can affect their metabolic fitness and response to exercise, thus impacting the accuracy of the exercise guidance provided.

[0040] Therefore, the embodiments described herein provide systems and methods for providing optimized exercise predictions and / or recommendations to improve a user's overall health. Optimized exercise predictions and / or recommendations are achieved by classifying users based on their physiological state and goals, and optimizing the exercise phase for the user based on such classification. Optimizing the exercise phase may include providing individualized exercise recommendations and / or (e.g., by sending data including instructions to such exercise machines) automatically controlling the exercise machine to achieve an optimized set of exercise parameters and / or providing real-time guidance to the user to achieve such parameters, in an effort to improve the user's fitness and overall health based on analyte parameters and / or non-analyte parameters measured for the user, as well as other user information (e.g., demographic information, physiological information, etc.). These optimized exercise predictions and / or recommendations can be followed by the user to achieve improvements in various aspects of the user's physiological state, including the user's lactate and / or glucose levels. Optimized exercise predictions and / or recommendations can also be determined in real time.

[0041] In some implementations, the system can identify earlier predictions and / or recommendations over time and can continuously refine future predictions and / or recommendations based at least in part on these earlier predictions and / or recommendations. This refinement of predictions and / or recommendations can be performed in real time. Refinement over time can further improve the accuracy of such predictions and / or recommendations, which in turn can further improve the health of users who adjust their diet, exercise, and other aspects of their lives in response to such predictions and / or recommendations.

[0042] Furthermore, by maximizing the accuracy of the user's initial classification based on user profiles, user input, or trial training phases, and subsequently associating that classification with one or more optimized training predictions and / or recommendations for the user, future adjustments to such recommendations can be minimized. This reduction in future adjustments, in turn, minimizes the computational and / or network load requirements of the hardware computing devices responsible for determining and presenting such recommendations. Given the large number of users requesting predictions and / or recommendations, this significantly reduces the network and / or computational requirements of the associated hardware systems, thereby improving the performance of such systems. It should also be noted that, considering the high frequency of analyte sampling, the large amount of physiological and target data involved, and the complexity of the computations being performed, manually monitoring users' analyte and / or non-analyte data and adjusting training predictions and / or recommendations in real time based on such manual monitoring is not feasible.

[0043] Furthermore, as described above, existing methods for measuring lactate may not be able to optimize a user's workout phase in real time, or may be detrimental to real-time optimization of a user's workout phase, at least in some cases. Therefore, the system described herein includes a continuous analyte monitoring system configured to measure lactate levels continuously, thereby providing real-time lactate measurements that can be used for real-time optimization of a user's workout phase. Note that, as used herein, real-time may also include near-real-time measurement and / or optimization recommendations (e.g., to account for the biological delay in the arrival of lactate levels in the blood and / or interstitial fluid). Additionally, optimization of the workout phase and workout-related feedback can be provided retrospectively, as further described herein.

[0044] As used herein, the term "continuous" analyte monitoring refers to monitoring one or more analytes in a fully continuous, semi-continuous, or periodic manner, resulting in a data stream of analyte values ​​over time. This data stream of analyte values ​​over time allows the use of the algorithms described herein to derive meaningful data and insights for user segmentation, optimization of exercise phases, and provision of exercise phase-related feedback. In other words, sporadic data points (e.g., preferably spaced several months apart) generated by single-point-in-time measurements collected from patients' visits to their healthcare professionals every few months cannot form the basis for any meaningful data or insights to be derived. Therefore, without the continuous analyte monitoring system of the embodiment described herein, it is entirely impossible to continuously segment users and optimize one or more exercise phases over time, and impossible to continuously provide exercise phase-related feedback as described herein.

[0045] Furthermore, the data stream of analyte values ​​collected over time using the continuous analyte monitoring system proposed herein includes real-time analyte values, which allows for the real-time extraction of meaningful data and insights using the systems and algorithms described herein. The extracted real-time data and insights, in turn, allow for the provision of real-time classification and exercise optimization feedback for users, as well as real-time feedback related to exercise phases. Real-time analyte values, as defined herein, refer to analyte values ​​that become available and operable within seconds or minutes of generation due to at least one sensor electronics module of the continuous analyte monitoring system (1) converting sensor current (i.e., analog electrical signal) generated by the continuous analyte sensor into sensor count values, (2) calibrating the count values ​​using the calibration techniques described herein to generate at least lactate and / or other analyte concentration values ​​to account for the sensitivity of the continuous analyte sensor, and (3) transmitting the measured lactate and / or other analyte concentration data (including lactate and / or other analyte concentration values) to a display device via a wireless connection.

[0046] For example, at least one sensor electronics module may be configured to sample analog electrical signals at a specific sampling period (or rate) (such as every 1 second (1 Hz), 5 seconds, 10 seconds, 30 seconds, 1 minute, 3 minutes, 5 minutes, etc.) and transmit the measured concentration data of lactic acid and / or other analytes to a display device at a specific transmission period (or rate), which may be the same as (or longer than) the sampling period, such as every 1 minute (0.016 Hz), 5 minutes, 10 minutes, etc.

[0047] Therefore, the continuous real-time analyte data generated by the continuous analyte monitoring system described herein allows the treatment management system of this paper to classify users and optimize exercise phases, as well as provide real-time feedback on exercise phases—something technically impossible to perform using existing or conventional techniques or systems. Furthermore, due to the real-time nature of this data, it is also impossible for humans to continuously process the real-time data stream of analyte values ​​over time using the algorithms and systems described herein to derive meaningful data and insights for user classification, optimization of exercise phases, and provision of feedback on exercise phases. In other words, deriving meaningful data and insights from a continuously generated, processed, calibrated, and analyzed real-time data stream using the algorithms and systems described herein is not a task that can be performed mentally. For example, performing real-time and continuous analysis of… Figures 4 to 7 The algorithm described (which would involve using a continuous stream of real-time data generated by a continuous analysis monitoring system of users and / or a significantly large amount of population data (e.g., hundreds or thousands of data points for each of thousands or millions of users in a user population)) is not a task that can be performed mentally, especially in real time.

[0048] Furthermore, some embodiments of this paper relate to technical solutions to technical problems associated with analyte sensor systems. Specifically, each analyte sensor system manufactured by a sensor manufacturer may behave slightly differently. Therefore, inconsistencies may exist between the sensors and the measurements they produce once in use. Consequently, some embodiments of this paper relate to determining the performance of the analyte sensor system during the manufacturing calibration process (in vitro), which includes quantifying certain sensor operating parameters such as calibration slope (also known as calibration sensitivity), calibration baseline, etc.

[0049] Typically, calibration sensitivity refers to the amount of current generated by the analyte sensor of an analyte sensor system when immersed in a predetermined amount of the analyte to be measured. The current can be expressed in picoamperes (pA) or counts. The amount of analyte measured can be expressed as a concentration level in milligrams per deciliter (mg / dL), and calibration sensitivity can be expressed in pA / (mg / dL) or counts / (mg / dL). The calibration baseline refers to the amount of current generated by the analyte sensor when no analyte is detected, and can also be expressed in pA or counts.

[0050] Calibration sensitivity, calibration baseline, and other information related to the sensitivity profile of the analyte sensor system can be programmed into the sensor electronics module of the analyte sensor system during the manufacturing process. This information is then used to convert the analyte sensor electrical signal into a measured analyte concentration level. For example, the calibration slope (calibration sensitivity) can be used to predict the initial in vivo sensitivity (M0) and the final in vivo sensitivity (M... f These are programmed into the sensor electronics module and used to convert the analyte sensor electrical signal into a measured analyte concentration level.

[0051] In some implementations, during in vivo use, the sensor electronics module of the analyte sensor system samples the analog electrical signal generated by the analyte sensor to generate an analyte sensor count value, and then, based on the analyte sensor count value, the initial in vivo sensitivity (M0) and the final in vivo sensitivity (M... f This is used to determine the measured analyte concentration level. For example, it can be based on the initial in vivo sensitivity (M0) and the final in vivo sensitivity (M... f The sensitivity function M(t) is used to determine the measured analyte concentration level. The sensitivity function M(t) can be expressed in several different ways, such as in a way that is independent of the elapsed time (t) of in vivo administration. i Simple correction factor, sensitivity and time (t) i The linear relationship between ) and sensitivity and time (t) i The exponential relationship between them, etc. Equation 1 presents a method for calculating the exponential relationship based on time t. iThe technique of determining the measured analyte concentration level (ACL) by the analyte sensor count (count) at the analyte location: ACL = count / M(t) i Equation 1 The calibration baseline (baseline) can also be used based on time t. i The analyte sensor count (count) at the analyte is used to determine the measured analyte concentration level (ACL), and Equation 2 provides a technique: ACL = (count - baseline) / M(t) i Equation 2 Includes example continuous analyte sensors for predicting exercise parameters to optimize exercise, and example treatment tubes. Theory System Figure 1 An example therapeutic management system 100 is illustrated, which is used to optimize exercise parameters to improve the fitness of healthy users and / or athletes 102 (referred to herein solely as users and collectively as users). As discussed herein, improved fitness may refer to improving a user's metabolic fitness, achieving weight loss, improving peak performance, improving resting metabolic rate, promoting a long-term increase in energy expenditure, lowering blood glucose levels, raising a user's lactate threshold, etc.

[0052] In some implementations, the user can be a healthy user or an athlete. A healthy user may have goals to improve fitness by, for example, maintaining and / or improving metabolic fitness, lowering blood glucose levels, or losing weight. An athlete may have goals to improve fitness by, for example, improving peak performance, increasing lactate threshold, or improving metabolic fitness over time. As discussed herein, metabolic fitness generally refers to a user's ability to clear lactate, and in some cases, this ability may be based on the effectiveness of the user's liver, kidneys, and skeletal muscle in clearing lactate.

[0053] The lactate ion is the conjugate base of lactate. During normal metabolism and exercise, lactate is produced from pyruvate (e.g., glucose is broken down into pyruvate) by lactate dehydrogenase. At rest, approximately up to 70% of lactate is metabolized by the liver, and the remainder can be metabolized by the kidneys or skeletal muscle during exercise. Therefore, lactate can be continuously monitored to assess parameters such as lactate clearance rate (which also indicates lactate half-life), lactate levels, lactate production rate, and lactate baseline for real-time determination of metabolic fitness.

[0054] In some implementations, the treatment management system 100 includes a continuous analyte monitoring system 104, a display device 107 executing an application 106, a treatment management engine 114, a user database 110, a historical record database 112, a training server system 140, and the treatment management engine 114, each of which is described in more detail below.

[0055] As used herein, the term "analyte" is used in its general sense and includes, but is not limited to, substances or chemical components in analyzable biological fluids (e.g., blood, interstitial fluid, cerebrospinal fluid, lymph, or urine). Analytes can include naturally occurring substances, artificial substances, metabolites, and / or reaction products. Analytes used for measurement by devices and methods can include, but are not limited to, potassium, glucose, endogenous insulin, prothrombin; acylcarnitine; adenine phosphoribosyltransferase; adenosine deaminase; albumin; alpha-fetoprotein; amino acid profiles (arginine (Krebs cycle), histidine / uric acid, homocysteine, phenylalanine / tyrosine, tryptophan); androstenedione; antipyrine; arabinitol enantiomers; arginase; benzoyl succinate (cocaine); biotinase; biopterin; C-reactive protein; carnitine; carnosine; CD4; and ceruloplasmin. ; chenodeoxycholic acid; chloroquine; cholesterol; cholinesterase; conjugated 1-β-hydroxycholic acid; cortisol; creatine kinase; creatine kinase MM isoenzyme; cyclosporine A; C-peptide, d-penicillamine; deethylchloroquine; dehydroepiandrosterone sulfate; DNA (acetyltransferase polymorphism, alcohol dehydrogenase, α1-antitrypsin, glucose-6-phosphate dehydrogenase, hemoglobin A, hemoglobin S, hemoglobin C, hemoglobin D, hemoglobin E, hemoglobin F, D-punjab, hepatitis B virus, HCMV, HIV-1, HTLV-1, MCAD, R NA, PKU, Plasmodium vivax, 21-deoxycortisol); debutylhalopantide; dihydropteridine reductase; diphtheria / tetanus antitoxin; erythrocyte arginase; erythrocyte protoporphyrin; esterase D; fatty acid / acylglycine; free β-human chorionic gonadotropin; free erythrocyte porphyrin; free thyroxine (FT4); free triiodothyronine (FT3); fumarate acetylacetase; galactose / gal-1-phosphate; galactose-1-phosphate uridine dihydrogenase; gentamicin; glucose-6-phosphate dehydrogenase; glutathione; glutathione Peroxidase; Glycinecholic acid; Glycosylated hemoglobin; Halopanatin; Hemoglobin variants; Hexosamine A; Human erythrocyte carbonic anhydrase I; 17-α-hydroxyprogesterone; Hypoxanthine phosphoribosyltransferase; Immunoreactive trypsin; Lactate; Lead; Lipoproteins ((a), B / A-1, β); Lysozyme; Mefloquine; Netilmicin; Phenobarbital; Phenytoin; Phytanic acid / norphytanic acid; Progesterone; Prolactin; Prolyase; Purine nucleoside phosphorylase; Quinine; Reverse triiodothyronine (rT3); Selenium; Serum pancreatic lipase; Sisomicin; Somatostatin C;Recognize one or more specific antibodies against the following, which may include: adenovirus, antinuclear antibody, anti-ζ antibody, arbovirus, Aujeszky's disease virus, dengue virus, Dracunculus medinensis, Echinococcus granulosus, Entamoeba histolytica, enterovirus, Giardia duodenalis, Helicobacter pylori, hepatitis B virus, herpes simplex virus, HIV-1, IgE (atopic disease), influenza virus, Leishmania donovani, leptospira, measles / mumps / rubella, Mycobacterium leprae, Mycoplasma pneumoniae. pneumoniae, myoglobin, Onchocerca volvulus, parainfluenza virus, Plasmodium falciparum, poliovirus, Pseudomonas aeruginosa, respiratory syncytial virus, rickettsia (scrub typhus), Schistosoma mansoni, Toxoplasma gondii, Trepenoma pallidium, Trypanosoma cruzi / rangeli, vesicular stomatis virus, Wuchereria bancrofti, yellow fever virus Hepatitis B virus (HIV-1); succinylacetone; sulfadoxine; theophylline; thyroid-stimulating hormone (TSH); thyroxine (T4); thyroxine-binding globulin; trace elements; transferrin; UDP-galactose-4-epimerase; urea; uroporphyrinogen I synthase; vitamin A; leukocytes; and zinc protoporphyrin.

[0056] In some specific implementations, naturally occurring salts, sugars, proteins, fats, vitamins, and hormones (e.g., insulin) in blood or interstitial fluid may also constitute analytes. Ions are charged atoms or compounds that may include, for example, sodium, potassium, calcium, chloride, nitrogen, or bicarbonate. Analytes may be naturally present in biological fluids, such as metabolites, hormones, antigens, antibodies, etc. Alternatively, analytes may be introduced into the body or be exogenous, such as contrast agents for imaging, radioactive isotopes, chemical reagents, synthetic blood based on fluorocarbons, or drugs or drug compositions, including but not limited to insulin; glucagon, ethanol; cannabis (cannabis, tetrahydrocannabinol, hemp); inhalants (nitrous oxide, amyl nitrite, butyl nitrite, chlorinated hydrocarbons, hydrocarbons); cocaine (cracked cocaine); stimulants (amphetamine, methamphetamine, methylphenidate, Cylert, Preludin, Didrex, PreState, Voranil, Sand). Rex, Plegine; sedatives (barbiturates, methylquinolone, tranquilizers such as diazepam, nitrazepam, metronidazole, tranxene, methylphenidate, tranxene); hallucinogens (phencyclidine, lysergic acid, mescaline, piodine, psilocybin); narcotics (heroin, codeine, morphine, opium, meperidine, percocet, percodan, tussionex, fentanyl, dalofop, analgesic, antidiarrheal); specialty drugs (fentanyl, meperidine, amphetamine, methamphetamine and analogues of phencyclidine, e.g., ecstasy); anabolic steroids; and nicotine. Metabolites of drugs and drug compositions are also envisioned analytes. It can also analyze analytes produced in the body, such as neurochemicals and other chemicals, such as ascorbic acid, uric acid, dopamine, norepinephrine, 3-methoxytyramine (3MT), 3,4-dihydroxyphenylacetic acid (DOPAC), homovanillic acid (HVA), 5-hydroxytryptamine (5HT) and 5-hydroxyindoleacetic acid (FHIAA), as well as intermediates in the citric acid cycle.

[0057] While the analytes that can be measured and analyzed by the apparatus and methods described herein include lactic acid, glucose and / or ketones, other analytes listed above may also be considered in some cases.

[0058] In some embodiments, the continuous analyte monitoring system 104 is configured to continuously measure one or more analytes and transmit the analyte measurements to a display device 107 for use by application 106. In some embodiments, the continuous analyte monitoring system 104 transmits the analyte measurements to the display device 107 via a wireless connection (e.g., Bluetooth). In some embodiments, the display device 107 is a smartphone. However, in some other embodiments, the display device 107 may alternatively be any other type of computing device, such as a laptop computer, smartwatch, fitness tracker, bicycle computer, tablet computer, or any other computing device capable of executing application 106. In some embodiments, the continuous analyte monitoring system 104 may be further configured to transmit the analyte measurements directly to exercise machine 108 via a wireless connection (e.g., Bluetooth). In some other embodiments, exercise machine 108 may receive the analyte measurements provided by the continuous analyte monitoring system 104 via the display device 107. The continuous analyte monitoring system 104 may be referenced... Figure 2 To describe in more detail.

[0059] The exercise machine 108 can be a treadmill, elliptical trainer, stationary bike, StairMaster, indoor rowing machine, smart cable machine, or any other type of exercise machine that can execute software applications.

[0060] Application 106 is a mobile health application configured to receive and analyze analyte measurements from analyte monitoring system 104. For example, application 106 stores information about the user (including the user's analyte measurements) in the user's user profile 118 for processing and analysis, and for use by treatment management engine 114 to provide treatment management support recommendations or guidance to the user. Application 111 is a software application running on exercise machine 108, and refers to a set of instructions for at least partially controlling the operation of exercise machine 108. Application 111 may perform some or all of the functions of application 106. For example, application 111 may be configured to receive and analyze analyte measurements from analyte monitoring system 104, store information about the user (including the user's analyte measurements) in the user's user profile 118, and / or perform some or all of the operations of treatment management engine 114.

[0061] Note that any references to suggestions, instructions, or recommendations provided by the treatment management engine 114 to the user may optionally be automatically provided to the application 111 of the exercise machine 108 to automatically implement the suggestions, instructions, or recommendations. Reference Figures 5A to 7The output of any of the described steps can be automatically sent to the application 111 of the exercise machine 108 to automatically adjust exercise parameters (e.g., speed, incline, resistance, repetitions, weight, etc.) based on corresponding suggestions, instructions, or recommendations. For example, as described herein, the treatment management engine 114 can instruct the user to increase the intensity (e.g., speed, incline, etc.) to reach a desired lactate threshold. In this example, the treatment management engine 114 can communicate directly with the application 111 to automatically adjust the exercise parameters of the exercise machine 108 to the desired increased intensity, rather than instructing the user.

[0062] The treatment management engine 114 refers to a set of software instructions having one or more software modules, including a data analysis module (DAM) 116. In some embodiments, the treatment management engine 114 executes entirely on one or more computing devices in a private or public cloud. In such embodiments, application 106 and / or application 111 communicate with the treatment management engine 114 via a network (e.g., the Internet). In some other embodiments, the treatment management engine 114 executes partly on one or more local devices, such as display device 107 and / or exercise machine 108, and partly on one or more computing devices in a private or public cloud. In some other embodiments, the treatment management engine 114 executes entirely on devices such as display device 107, exercise machine 108, and / or continuous analyte monitoring system 104 (e.g., Figure 2 The sensor electronics module 204 executes on one or more local devices. As discussed in more detail herein, the treatment management engine 114 may provide treatment management support recommendations to the user via application 106 and / or application 111. The treatment management engine 114 provides treatment management support recommendations based on information included in the user profile 118.

[0063] User profile 118 may include information about the user collected from application 106. For example, application 106 provides an input set 128 that includes analyte measurements associated with one or more analytes received from continuous analyte monitoring system 104, which are stored in user profile 118. In some embodiments, the input 128 provided by application 106 may also include data other than analyte measurements. For example, application 106 may obtain additional input 128 through manual user input, one or more other non-analyte sensors or devices, other applications executed on display device 107, etc. Non-analyte sensors and devices include, but are not limited to, one or more of the following: insulin pumps, respiratory sensors, sensors or devices provided by display device 107 (e.g., accelerometers, cameras, GPS, heart rate monitors, electrocardiograms (ECGs), etc.) or other user accessories (e.g., smartwatches, continuous positive airway pressure (CPAP) machines, or fitness trackers), or any other sensor or device that provides relevant information about the user (e.g., sensors on exercise equipment). References below. Figure 3 The input 128 of the user profile 118 provided by application 106 is described in more detail.

[0064] The DAM 116 of the treatment management engine 114 is configured to process the input set 128 to determine one or more metrics 130. See below for reference. Figure 3 The indicator 130, discussed in more detail, can at least in some cases broadly indicate one or more of the user's health or status, such as the user's physiological state, trends associated with the user's health or status, etc. In some implementations, indicator 130 can then be used by the treatment management engine 114 as input to provide guidance to the user. As shown, indicator 130 is also stored in the user profile 118.

[0065] User profile 118 also includes demographic information 120, disease information 122, and / or medication information 124. In some embodiments, such information may be provided by user input or obtained from certain data repositories (e.g., electronic medical records, etc.). In some embodiments, demographic information 120 may include one or more of the user's age, height, weight, BMI (body mass index), ethnicity, sex, etc. In some embodiments, disease information 122 may include information about one or more diseases of the user, including information about the user's metabolic fitness, liver health, diabetes, kidney disease, and / or any condition or disease related to metabolic fitness. In some embodiments, disease information 122 may also include the length of time since diagnosis, the degree of disease control, adherence to disease management treatment, other types of diagnoses (e.g., heart disease, obesity), etc. In some embodiments, disease information 122 may include hospitalization and / or surgical history. In some implementations, disease information 122 may include other measures of health (e.g., heart rate, stress, sleep, etc.) or fitness (e.g., cardiovascular endurance, metabolic status, gait information, muscle strength and / or strength, other measures of muscle endurance and fitness).

[0066] In some implementations, the user profile 118 may also include information about the environment from user input and / or non-analyte sensors, such as, but not limited to, estimated or actual external temperature, humidity levels, atmospheric pressure, and / or exercise altitude. Furthermore, if it is predicted that the user will exercise outdoors, or if the user has already selected to exercise outdoors, further information about potential precipitation (e.g., rain or snow) and information about wind speed, wind chill factor, and solar index may be used to estimate and predict the user's metabolic activity and ideal exercise program and / or to make recommendations regarding the user's metabolic activity and ideal exercise program.

[0067] In some implementations, drug information 124 may include information about the amount and type of drug taken by the user.

[0068] In some implementations, drug information may include information about the intake of one or more drugs known to affect exercise performance, metabolic fitness, damage the liver (e.g., affect lactate clearance), and / or cause liver toxicity. One or more drugs known to damage the liver and / or cause hepatotoxicity may include: antibiotics such as amoxicillin / clavulanate, clindamycin, erythromycin, nitrofurantoin, rifampin, sulfonamides, tetracyclines, trimethoprim / sulfamethoxazole, and drugs used to treat tuberculosis (isoniazid and pyrazinamide); anticonvulsants such as tarbamazepine, thenobarbital, phenytoin, and valproate; antidepressants such as bupropion, fluoxetine, mirtazapine, paroxetine, sertraline, trazodone, and tricyclic antidepressants such as amitriptyline; antifungals such as ketoconazole and terbinafine; antihypertensive drugs (e.g., drugs used to treat high blood pressure, or sometimes drugs used to treat kidney or heart disease), such as captopril, enalapril, irbesartan, lisinopril, losartan, and verapamil; antipsychotics such as phenothiazines (e.g., chlorpromazine) and risperidone; and cardiac drugs. Drugs such as amiodarone and clopidogrel; beta-blockers; hormone regulators such as anabolic steroids, contraceptives (oral contraceptives), and estrogens; analgesics such as acetaminophen and nonsteroidal anti-inflammatory drugs (NSAIDs); and other drugs such as acarbose (e.g., for the treatment of diabetes), allopurinol (e.g., for the treatment of gout), antiretroviral therapy (ART) drugs (e.g., for the treatment of human immunodeficiency virus (HIV) infection), baclofen (e.g., a muscle relaxant), cyproheptadine (e.g., an antihistamine), azathioprine (e.g., for the prevention of organ transplant rejection), methotrexate (e.g., for the treatment of cancer), omeprazole (e.g., for the treatment of gastroesophageal reflux), PD-1 / PD-L1 inhibitors (e.g., anticancer drugs), statins (e.g., for the treatment of high cholesterol levels), and many types of chemotherapy, including immune checkpoint inhibitors.

[0069] In some implementations, drug information may include information about the intake of one or more drugs known to improve liver function. One or more drugs known to improve liver function may include S-adenosylmethionine, avatrombopag, dehydroemetin, entecavir, gliclazide, and pirentavir, lamivudine, metadoxine, methionine, sofosbuvir, velpatasvir, voxiprevir, telbivudine, tenofovir, trienta, ursodeoxycholic acid, etc.

[0070] In some implementations, the user profile 118 is dynamic because at least a portion of the information stored in the user profile 118 can be modified or updated over time and / or new information can be added to the user profile 118 via the treatment management engine 114 and / or application 106. Therefore, the information in the user profile 118 stored in the user database 110 provides an up-to-date repository of information related to that user.

[0071] In some implementations, user database 110 refers to a storage server operating, for example, in a public or private cloud. User database 110 can be implemented as any type of data storage, such as a relational database, a non-relational database, a key-value data storage, a file system including a hierarchical file system, etc. In some exemplary implementations, user database 110 is distributed. For example, user database 110 may include multiple distributed persistent storage devices. Furthermore, user database 110 can be replicated, allowing the storage devices to be geographically dispersed.

[0072] User database 110 includes user profiles 118 associated with multiple users, including those who have interacted with or have interacted with application 106 on their own devices. The user profiles stored in user database 110 are accessible not only by application 106 but also by treatment management engine 114 and / or exercise machine 108. Application 106, treatment management engine 114, and / or exercise machine 108 can access the user profiles in user database 110 via one or more networks (not shown) (such as one or more wireless networks). As described above, treatment management engine 114, and more specifically, its data analysis module (DAM) 116, can retrieve input 128 from user profiles 118 stored in user database 110 and calculate one or more metrics 130, which can then be stored as application data 126 in user profile 118.

[0073] In some implementations, user profiles 118 stored in user database 110 may also be stored in history database 112. User profiles 118 stored in history database 112 can serve as a repository providing both current and historical information for each user of application 106. Therefore, history database 112 essentially provides all data related to each user of application 106, where timestamps are used to store the data. A timestamp associated with any piece of information stored in history database 112 can identify, for example, when that information was obtained and / or updated.

[0074] Furthermore, in some implementations, the historical data database 112 may include data from one or more users who are not users of the continuous analyte monitoring system 104 and / or application 106. For example, analyte data from users who have used the continuous analyte monitoring system 104 and application 106 for up to five years to optimize exercise phases and improve user fitness may be maintained as time-series analyte data associated with the users over a five-year period.

[0075] Furthermore, in some embodiments, the historical data database 112 may include data from one or more users who are not users of the continuous analyte monitoring system 104 and / or application 106. For example, the historical data database 112 may include information (e.g., user profiles) associated with one or more users known to be healthy, and information (e.g., user profiles) associated with one or more users known to be athletes. The data stored in the historical data database 112 may be referred to herein as population data, which may include hundreds or thousands of data points for each of thousands or millions of users in a user population. In other words, the data stored in the historical data database 112 and used in some embodiments described herein may include gigabytes, terabytes, petabytes, exabytes, etc.

[0076] Data associated with each user stored in the historical database 112 provides time-series data collected over the user's lifespan. For example, the data may include information about the user's exercise phases, including exercise parameters that help the user achieve the desired type of exercise (e.g., speed, inclination, wattage, duration of the exercise phase, frequency of the exercise phase). The data may also include physiological information (e.g., height and weight) and non-analyte sensor data (e.g., heart rate, respiratory rate, etc.). Such data can indicate the type of exercise performed by the user (e.g., HIIT or Zone 2), the user's physiological status, the user's lactate level, glucose level, insulin level, free fatty acid level, the status / condition of one or more of the user's organs, the user's habits (e.g., activity level, food intake, etc.), and the progress of fitness outcomes over time (e.g., weight loss, metabolic fitness, etc.).

[0077] Although depicted as separate databases for clarity of concept, in some implementations, user database 110 and historical data database 112 may operate as a single database. That is, historical and current data related to users of the continuous analyte monitoring system 104 and application 106, as well as historical data related to users who were not previously users of the continuous analyte monitoring system 104 and application 106 and / or application 111, may be stored in a single database. This single database may be a storage server operating in a public or private cloud.

[0078] Figure 1 An exercise machine 108 is also illustrated, comprising a controller 109 configured to automatically control exercise intensity by adjusting various exercise parameters in response to a trigger. For example, a trigger could be user input provided to a software application (“application”) 111 via a user interface 113 of the exercise machine 108. In this example, in response to user input, the application 111 causes the controller 109 to adjust the exercise intensity by adjusting one or more parameters, such as speed, inclination, resistance, and other parameters known to those skilled in the art. As known to those skilled in the art, the controller can adjust such exercise parameters by sending commands to various electromechanical components in the exercise machine 108, such as actuators, motors, etc. The application 111 refers to a set of software instructions configured to cause the controller 109 to control the operation of the exercise machine 108 (and other operations as described above).

[0079] In another example, the trigger corresponds to an exercise intensity instruction received from a person training the user (e.g., a personal trainer). The personal trainer can pre-define the exercise phase by type (zone 2, resistance training, or HIIT), and the exercise machine 108 can receive instructions corresponding to the pre-determined exercise phase via application 111. The controller 109 can then automatically control and maintain the exercise intensity according to the prescribed exercise type to achieve the corresponding exercise zone (zone 2, resistance training, or HIIT).

[0080] In another example, a trigger may correspond to an exercise intensity instruction received by application 111 from another device (such as display device 107) to guide controller 109 to adjust the exercise intensity. For example, by executing application 106, display device 107 may generate exercise intensity instructions that can be transmitted via a wired or wireless connection to exercise machine 108 to control the operation of exercise machine 108. In yet another example, according to the embodiments described herein, application 111 may be configured to automatically (e.g., in the absence of user input and / or instructions from external devices) generate optimized exercise intensity. For example, application 111 may include instructions that partially or fully perform the functions of treatment management engine 114 to generate optimized exercise intensity instructions (including exercise intensity parameters) that controller 109 may use to adjust the exercise intensity of a user's workout on exercise machine 108.

[0081] Exercise machine 108 can be any kind of resistance machine or circular path machine. Examples of exercise machine 108 may include treadmills, elliptical trainers, glider machines, climbing machines, stair climbers, stationary bikes, etc.

[0082] As previously mentioned, the treatment management system 100 is configured to provide exercise optimization for healthy users and athletes using a continuous analyte monitoring system 104 that includes at least a continuous lactate sensor. In some embodiments, to achieve such optimization, the treatment management engine 114 is configured to (1) provide real-time and / or non-real-time exercise treatment management support (e.g., guidance) to the user and / or others (including, but not limited to, fitness instructors, family members of the user, caregivers of the user, etc.), and / or (2) provide real-time instructions to the exercise machine (e.g., exercise machine 108) for automatically adjusting the exercise intensity (including various exercise parameters) for the user to use the exercise machine for exercise. In some embodiments, the treatment management support includes optimal exercise guidance to improve or optimize metabolic fitness and / or support optimal exercise performance.

[0083] Guidance can be provided to users, coaches, family members, caregivers, etc., in the form of recorded speech or noise or synthesized speech or noise from hardware (e.g., assistive devices, wearable watches, wearable fitness trackers, and / or smartphone applications) communicating with the treatment management system 100. This guidance can instruct users to increase, decrease, or maintain their exercise intensity. Additionally, auditory signals can be in the form of rhythmic melodies or music, which can change the tempo or speed of the beat when instructing the user to accelerate or slow down their exercise. Furthermore, different rhythmic music can be selected for periods instructing the user to slow down, for example, to match the tempo of a slower song, compared to periods instructing the user to accelerate, for example, to match the tempo of a faster song. Similarly, guidance can be provided in the form of physical stimuli, such as, but not limited to, vibration, band contraction, electrical signals or impacts, temperature changes on wearable skin devices, light intensity or color-based changes, visual graphic images and / or avatar changes on displays and / or visually visible screens, light projected onto the eyes, or wearable visual devices. In addition, guidance can be delivered to users in real time during workouts or retrospectively using replay indicators, allowing users to visualize and understand how they should optimize their workouts in future phases to maximize their performance goals based on previous workouts.

[0084] Specifically, the treatment management engine 114 can be used to collect information associated with users in user profiles 118 stored in the user database 110, to perform analysis on that information, thereby classifying users as healthy users, athletes, or metabolically challenged users. Based on the classification, the treatment management engine 114 can then optimize the user's workouts to help the user achieve their goals. For example, if a user is classified as a healthy user, the treatment management engine 114 can optimize the user's workouts to maintain that user's metabolic fitness or to lose weight. As described above, optimizing workouts may include providing the user with optimized workout guidance and / or automating the operation of the workout machine by controlling the machine's speed, incline, resistance, etc., and the duration of the workout. The treatment management engine 114 can access user profiles 118 via one or more networks (not shown) to perform such analyses. In some embodiments, in addition to real-time workout guidance, workout optimization may also include suggesting the number of workout sessions per week, including the duration and intensity of each workout session. Workout sessions can be planned according to weekly or monthly time periods for optimal results.

[0085] In some implementations, the treatment management engine 114 may utilize one or more trained machine learning models capable of performing analysis on information that the treatment management engine 114 has collected / received from the user profile 118. Figure 1 In the illustrated implementation, the treatment management engine 114 may utilize a trained machine learning model provided by the training server system 140. Although depicted as a separate server for clarity of concept, in some implementations, the training server system 140 and the treatment management engine 114 may operate as a single server or system. That is, the model may be trained and used by a single server, or it may be trained by one or more servers and deployed for use on one or more other servers or systems. In some implementations, the model may be trained on one or more virtual machines (VMs) that run at least partially on one or more physical services in relational database format and / or non-relational database format.

[0086] Training server system 140 is configured to train a machine learning model using training data, which may include data (e.g., from user profiles) associated with one or more users (e.g., users or non-users of continuous analyte monitoring system 104 and / or application 106), who may be healthy users, athletes, and / or users with metabolic disorders (e.g., users with metabolic disorders, liver disease, etc.). The training data may be stored in a historical database 112 and may be accessed by training server system 140 via one or more networks (not shown) to train the machine learning model.

[0087] Training data refers to a dataset that has already been characterized and labeled. For example, a dataset may include multiple data records, each containing information corresponding to different user profiles stored in a user database 110, where each data record is characterized and labeled. In machine learning and pattern recognition, a feature is a single measurable attribute or characteristic. Typically, features that best characterize patterns in the data are selected to create predictive machine learning models. Data labeling is the process of adding one or more meaningful and informative labels to provide context to the data for use by machine learning models.

[0088] As an illustrative example, each relevant characteristic of a user reflected in the corresponding data record can be a feature used to train a machine learning model. Such features may include demographic information (e.g., age, sex, ethnicity, etc.), analyte information (e.g., normal lactate range (e.g., lactate baseline, lactate threshold, lactate clearance rate, and / or lactate change rate during and after exercise, etc.), exercise-related information (e.g., start and end times associated with exercise, exercise duration, exercise type and / or intensity (e.g., speed, incline, resistance, etc.), non-analyte sensor information (e.g., heart rate, temperature, etc.), and / or any other information related to classifying the user and / or optimizing the exercise for the user. Furthermore, the data records are labeled with information corresponding to the model being trained to predict. In one example, if the model is trained to classify users as healthy users, athletes, or metabolically unfit users, then the data in the training dataset... Records are labeled using this classification. In another example, if the model is trained to output optimized training parameters, data records in the training dataset are labeled with one or more such parameters. Note that in one example, such a model could be a multiple-input single-output (MISO) model configured to predict only one optimized training parameter (e.g., velocity). In this case, additional MISO models can be trained, each predicting one of the other training parameters (e.g., tilt, resistance, etc.). In another example, the model could be a multiple-input multiple-output (MIMO) model configured to predict multiple optimized training parameters (e.g., velocity, tilt, resistance, etc.).

[0089] The training server system 140 then trains the model using the characterized and labeled training data. Specifically, the features of each data record can be used as input to the machine learning model, and the generated output can be compared with the label associated with the corresponding data record. The model can calculate a loss based on the difference between the generated output and the provided label. This loss is then used to modify the model's internal parameters or weights. By iteratively processing each data record corresponding to each historical user, the model can be iteratively refined to generate accurate predictions of user classification, optimized training parameters, etc.

[0090] like Figure 1 As illustrated, the training server system 140 deploys these trained models to the treatment management engine 114 for use during runtime. For example, the treatment management engine 114 may obtain user profiles 118 associated with users and stored in the user database 110, use information from user profiles 118 as input to the trained models, and output predictions indicating whether a user is classified as a healthy user, athlete, or otherwise, and / or suggested exercise parameters for HIIT, resistance training, or Zone 2 exercises (e.g., such as...). Figure 1 (As shown in output 144). Output 144 generated by the treatment management engine 114 can also indicate improvements in the user's fitness and / or metabolic health over time. Output 144 may be provided (e.g., via application 106) to the user, to the user's caregiver (e.g., parent, relative, guardian, physical therapist, fitness coach, nurse, etc.), to the user's doctor or healthcare provider, or to any other individual interested in the user's health for the purpose of improving the user's health (e.g., in some cases, by implementing recommended treatments). Output 144 generated by the treatment management engine 114 is stored in the user database 110 and is used to train or retrain the trained model.

[0091] In some implementations, output 144 generated by the treatment management engine 114 may be stored in a user profile 118. Output 144 may indicate user classification (e.g., healthy user, athlete, or metabolically inept), current or future fitness level, optimized exercise parameters, etc. Output 144 stored in user profile 118 may be continuously updated by the treatment management engine 114. Thus, for example, previous fitness predictions initially stored as output 144 in user profile 118 in user database 110 and then passed to historical database 112 can provide an indication of the progress or improvement of a user's fitness over time, and provide an indication of the effectiveness of different exercise recommendations and / or treatments recommended to the user to improve fitness.

[0092] In some implementations, the training server system 140 may use the user's own historical data to train a personalized model for the user, which provides therapeutic management support and insights regarding the user's exercise goals, exercise optimization, and user fitness. For example, in some implementations, a model trained based on population data may be used to provide the user with optimized exercise parameters. However, after collecting personalized information (e.g., analyte sensor information, non-analyte sensor information, exercise type, and / or parameters, etc.) associated with one or more exercise phases performed by the user, the personalized information may be used to further personalize the model. For example, information obtained from one or more exercise phases performed by the user (e.g., including information obtained during one or more exercise phases and information obtained after one or more exercise phases reflecting the impact of such one or more phases) may be used to optimize exercise parameters for future exercise phases.

[0093] Furthermore, a user's historical data can be used as a baseline to indicate improvement, stagnation, or deterioration in user fitness. For example, improvement, stagnation, or deterioration in a user's metabolic fitness can be based on the user's lactate production and / or the user's ability to clear lactate after a workout. As an illustrative example, data from a user two weeks ago can be used as a baseline, which can be compared to the user's current data to identify whether the user's metabolic fitness has improved. In some implementations, the model may further be able to predict or anticipate a user's metabolic fitness or its future improvement / deterioration based on the user's recent data patterns (e.g., workout data, food intake data, etc.). Improvement, stagnation, or deterioration in user fitness can be communicated to the user, coach, caregiver, etc., in a visual or auditory format from one or more devices (e.g., display device 107, wearable watch, wearable fitness tracker, and / or smartphone app).

[0094] In some implementations, the trainable model is designed to provide recommendations for food, lifestyle, and other types of therapeutic management support to help users improve their fitness based on their historical data, including how different types of food and / or activities have affected their past fitness. In some implementations, the trainable model is designed to predict the root causes of certain improvements, stagnation, or deterioration in a user's fitness. For example, application 106 may display a user interface with charts showing the user's metabolic fitness using trend lines and, for example, retrospectively indicating how the user's metabolic fitness fluctuated at certain points in time.

[0095] Figure 2 Figure 200 is a conceptual illustration of an example continuous analyte monitoring system 104, including an example continuous analyte sensor with sensor electronics, according to certain aspects of this disclosure. For example, according to certain aspects of this disclosure, system 104 may be configured to continuously monitor one or more analytes belonging to a user.

[0096] The continuous analyte monitoring system 104 in the illustrated embodiment includes a sensor electronics module 204 and one or more continuous analyte sensors 202 associated with the sensor electronics module 204 (collectively referred to herein as continuous analyte sensors 202). The sensor electronics module 204 may be associated with one or more of display devices 210, 220, 230, and 240 and exercise machines (e.g., Figure 1 The sensor electronics module 204 can also wirelessly communicate with one or more medical devices (such as medical device 208, which is referred to herein as medical device 208 and collectively as medical device 208) and / or one or more other nonanalyte sensors 206 (which is referred to herein as nonanalyte sensor 206 and collectively as nonanalyte sensor 206) (e.g., directly or indirectly).

[0097] In some embodiments, the continuous analyte sensor 202 may include one or more sensors for detecting and / or measuring analytes. The continuous analyte sensor 202 may be a multi-analyte sensor configured to continuously measure two or more analytes or a single-analyte sensor configured to continuously measure a single analyte, such as a non-invasive device, subcutaneous device, percutaneous device, transdermal device, and / or intravascular device. In some embodiments, the continuous analyte sensor 202 may be configured to continuously measure a user's analyte levels using one or more techniques, such as enzymatic techniques, chemical techniques, physical techniques, electrochemical techniques, spectrophotometric techniques, polarization techniques, calorimetric techniques, iontophoresis techniques, radiation techniques, immunochemistry techniques, etc. As used herein, the term "continuous" may mean fully continuous, semi-continuous, periodic, etc. In some aspects, the continuous analyte sensor 202 provides a data stream indicating the concentration of one or more analytes in a user's body. The data stream may include raw data signals, which are then converted into a calibrated and / or filtered data stream for providing the user with estimated analyte values.

[0098] In some embodiments, the continuous analyte sensor 202 may be a multianalyte sensor configured to continuously measure multiple analytes in the user's body. For example, in some embodiments, the continuous multianalyte sensor 202 may be a single sensor configured to measure lactate, glucose, ketones (e.g., 3-β-hydroxybutyrate, acetoacetate, acetone, etc.), glycerol, and / or free fatty acids in the user's body.

[0099] In some embodiments, one or more multianalyte sensors may be used in combination with one or more single analyte sensors. As an illustrative example, a multianalyte sensor may be configured to continuously measure lactate and glucose, and in some cases may be used in combination with an analyte sensor configured to measure only ketones or only potassium. Information from each of the multianalyte and single analyte sensors can be combined to provide treatment management support using the methods described herein. In other embodiments, additional non-contact and / or periodic or semi-continuous but time-limited measurements of physiological information may be integrated into the system, such as non-contact heart rate monitoring including weighing scale information or from a sensor pad beneath the user sitting in a chair or bed, estimation of height, weight, or other parameters without physical contact via an infrared camera detecting the user's temperature and / or blood flow patterns, and / or via a vision camera with machine vision.

[0100] In some embodiments, the continuous analyte sensor 202 may include a percutaneous lead having a proximal portion coupled to the sensor electronics module 204 and a distal portion having several electrodes, such as a measuring electrode and a reference electrode. The measuring (or operating) electrode may be coated, covered, treated, embedded, etc., with one or more chemical molecules that react with a specific analyte, and the reference electrode may provide a reference voltage. The measuring electrode may generate an analog electrical signal that is delivered along a conductor extending from the measuring electrode to the proximal portion of the percutaneous lead coupled to the sensor electronics module 204. After the continuous analyte monitoring system 104 has been applied to the patient's epidermis, the continuous analyte sensor 202 penetrates the epidermis, and the distal portion extends into the dermis and / or subcutaneous tissue beneath the epidermis. Other configurations of the continuous analyte sensor 202 may also be used, such as a multi-analyte sensor comprising multiple measuring electrodes, each measuring electrode generating an analog electrical signal representing the concentration level of a specific analyte.

[0101] Typically, a single-analyte sensor generates an analog electrical signal proportional to the concentration level of a specific analyte. Similarly, each multi-analyte sensor generates multiple analog electrical signals, each proportional to the concentration level of a specific analyte. As an illustrative example, the continuous analyte sensor 202 may include a single-analyte sensor configured to measure lactate concentration levels and another single-analyte sensor configured to measure a patient's glucose concentration levels. As another illustrative example, the continuous analyte sensor 202 may include a single-analyte sensor configured to measure lactate concentration levels and one or more multi-analyte sensors configured to measure glucose concentration levels, ketone concentration levels, creatinine concentration levels, etc. As yet another illustrative example, the continuous analyte sensor 202 may include a multi-analyte sensor configured to measure lactate concentration levels, glucose concentration levels, ketone concentration levels, creatinine concentration levels, etc. Therefore, the continuous analyte sensor 202 is configured to generate at least one analog electrical signal proportional to the concentration level of a specific analyte, and the sensor electronics module 204 is configured to convert the analog electrical signal into an analyte sensor count value, calibrate the analyte sensor count value based on the sensitivity curve of the continuous analyte sensor 202 to generate a measured analyte concentration level, and transmit the measured analyte concentration level data (including the measured analyte concentration level) to a display device, such as display devices 210, 220, and / or 230, via a wireless connection. For example, the sensor electronics module 204 may be configured to sample the analog electrical signal at a specific sampling period (or rate) (such as every 1 second (1 Hz), 5 seconds, 10 seconds, 30 seconds, 1 minute, 3 minutes, 5 minutes, etc.), and transmit the measured analyte concentration data to the display device at a specific transmission period (or rate), which may be the same as (or longer than) the sampling period, such as every 1 minute (0.016 Hz), 5 minutes, 10 minutes, 30 minutes, etc. at the end of the wear cycle. Depending on the sampling period and the transmission period, the measured analyte concentration data transmitted to the display device includes at least one measured analyte concentration level with associated time stamps, serial numbers, etc.

[0102] In some embodiments, the continuous analyte sensor 202 may have a thermocouple incorporated within or alongside the percutaneous lead to provide an analog temperature signal to the sensor electronics module 204, which can be used to correct the analog electrical signal or measured analyte data for temperature. In other embodiments, the thermocouple may be incorporated into the sensor electronics module 204 above the adhesive pad, or alternatively, the thermocouple may contact the patient's epidermis through an opening in the adhesive pad.

[0103] In some implementations, the sensor electronics module 204 includes, in particular, a processor 233, a storage element or memory 234, a wireless transmitter / receiver (transceiver) 236, one or more antennas coupled to the wireless transceiver 236, analog electrical signal processing circuitry, analog-to-digital (A / D) signal processing circuitry, digital signal processing circuitry, a power supply (such as a potentiostat) for the continuous analyte sensor 202, etc.

[0104] Processor 233 may be a general-purpose or special-purpose microprocessor, application-specific integrated circuit (ASIC), field-programmable gate array (FPGA), etc., which executes instructions to perform control, calculation, input / output, and other functions of sensor electronics module 204. Processor 233 may include a single integrated circuit, such as a microprocessor device, or multiple integrated circuit devices and / or circuit boards that work together to achieve appropriate functions. In some embodiments, processor 233, memory 234, wireless transceiver 236, A / D signal processing circuitry, and digital signal processing circuitry may be combined into a system-on-a-chip (SoC).

[0105] Typically, processor 233 can be configured to sample analog electrical signals at regular intervals (such as sampling periods) using A / D signal processing circuitry to generate analyte sensor counts based on the analog electrical signals generated by continuous analyte sensor 202, calibrate the analyte sensor counts based on the sensitivity curve of continuous analyte sensor 202 to generate measured analyte concentration levels, and generate measured analyte data from the measured analyte concentration levels, generating sensor data packets that particularly include the measured analyte concentration level data. Processor 233 can store the measured analyte concentration level data in memory 234 and generate sensor data packets at regular intervals (such as transmission periods) for transmission by wireless transceiver 236 to display devices, such as display devices 210, 220, 230, and / or 240. Processor 233 can also add additional data to the sensor data packets, such as supplementary sensor information including sensor identifiers, sensor status, temperature corresponding to the measured analyte data, etc. The sensor data packets are then wirelessly transmitted to the display devices via a wireless connection. In some embodiments, the wireless connection is Bluetooth or Bluetooth Low Energy (BLE) connection. In such implementations, sensor data packets are transmitted to the display device in the form of Bluetooth or BLE data packets.

[0106] In various embodiments, memory 234 may include volatile and non-volatile media. For example, memory 234 may include random access memory (RAM), dynamic RAM (DRAM), static RAM (SRAM), read-only memory (ROM), flash memory, cache memory, and / or any other type of non-transitory computer-readable media. Memory 234 may store one or more analyte sensor system applications, modules, instruction sets, etc., for execution by processor 233, such as instructions for generating measured analyte data from analyte sensor count values.

[0107] The memory 234 may also store certain sensor operating parameters 235, such as calibration slope (or calibration sensitivity), calibration baseline, etc. Specifically, calibration sensitivity, calibration baseline, and other information related to the sensitivity curve of the sensor electronics module 204 can be programmed into the sensor electronics module 204 during the manufacturing process and then used to convert the analyte sensor electrical signal into a measured analyte concentration level. For example, as discussed above, the calibration slope can be used to predict the initial in vivo sensitivity (M0) and the final in vivo sensitivity (M... f These are stored in memory 234 and used to convert the analyte sensor electrical signal into a measured analyte concentration level. In some embodiments, the calibration sensitivity (M... CC 246 and / or calibration baseline 247 may be stored in memory 234.

[0108] In some embodiments, sensor electronics module 204 includes electronic circuitry associated with measuring and processing continuous analyte sensor data, including look-ahead algorithms associated with processing and calibrating the sensor data. Sensor electronics module 204 may be physically connected to continuous analyte sensor 202 and may be integral with (non-releasably attached to) or releasably attached to continuous analyte sensor 202. Sensor electronics module 204 may include hardware, firmware, and / or software enabling the measurement of analyte levels via continuous analyte sensor 202. For example, sensor electronics module 204 may include a potentiostat, a power supply for powering the sensor, other components for signal processing and data storage, and a telemetry module for transmitting data from the sensor electronics module to, for example, one or more display devices. The electronics may be mounted to a printed circuit board (PCB) and can take various forms. For example, the electronics may take the form of integrated circuits (ICs), such as application-specific integrated circuits (ASICs), microcontrollers, and / or processors.

[0109] Display devices 210, 220, 230, and / or 240 are configured to display displayable sensor data, including analyte data that can be transmitted by sensor electronics module 204. Each of display devices 210, 220, 230, or 240 may include a display, such as touchscreen displays 212, 222, 232, and / or 242, for displaying sensor data to a user and / or receiving input from a user. For example, a graphical user interface (GUI) may be presented to the user for such purposes. In some embodiments, as an alternative to or complement to a touchscreen display, the display devices may include other types of user interfaces, such as a voice user interface, for communicating sensor data to the user of the display device and / or receiving user input. Display devices 210, 220, 230, and 240 may be Figure 1 The example shown is used to... Figure 1 An example of a display device 107 that displays sensor data to the user and / or receives input from the user. In some embodiments, the exercise machine 108 may also be used as a display device for communicating sensor data and / or exercise recommendations to the user of the display device and / or for receiving user input.

[0110] In some implementations, one, some, or all of the display devices are configured to display the sensor data as is, or otherwise convey (e.g., in language) the sensor data as it is transmitted from the sensor electronics module (e.g., in a custom data packet transmitted to the display device based on corresponding preferences), without any additional look-ahead processing required for the calibration and real-time display of the sensor data.

[0111] Multiple display devices may include custom display devices specifically designed to display certain types of displayable sensor data associated with analyte data received from the sensor electronics module. In some embodiments, the multiple display devices may be configured to provide warnings / alarms based on displayable sensor data. Display device 210 is an example of such a custom device. In some embodiments, one of the multiple display devices is a smartphone, such as display device 220, which uses a commercial operating system (OS) to represent the mobile phone and is configured to display a graphical representation of continuous sensor data, including, for example, current and historical data. Other display devices may include other handheld devices, such as display device 230 representing a tablet computer, display device 240 representing a smartwatch or fitness tracker, medical device 208 (e.g., an insulin infusion device or a blood glucose meter), and / or a desktop or laptop computer (not shown).

[0112] Because different display devices offer different user interfaces, the content of data packets (e.g., the amount, format, and / or type of data to be displayed, alarms, etc.) can be customized (e.g., programmed differently by the manufacturer and / or by the end user) for each specific display device. Therefore, in some embodiments, multiple different display devices can directly communicate wirelessly with sensor electronics modules (e.g., on-skin sensor electronics module 204, such as one physically connected to the continuous analyte sensor 202) during a sensor session to enable multiple different types and / or levels of display and / or functionality associated with displayable sensor data.

[0113] As mentioned, the sensor electronics module 204 can communicate with the medical device 208. In some example embodiments of this disclosure, the medical device 208 can be a passive device. For example, the medical device 208 can be an insulin pump for administering insulin to a user. For a variety of reasons, such an insulin pump may be expected to receive and track lactate, glucose, ketone, glycerol, and free fatty acid values ​​transmitted from the continuous analyte monitoring system 104, wherein the continuous analyte sensor 202 is configured to measure lactate, glucose, ketones, glycerol, and / or free fatty acids.

[0114] Furthermore, as mentioned, the sensor electronics module 204 can also communicate with other nonanalyte sensors 206. Nonanalyte sensors 206 may include, but are not limited to, altimeter sensors, accelerometer sensors, GPS sensors, temperature sensors, and respiratory rate sensors. Nonanalyte sensors 206 may also include monitors such as heart rate monitors, blood pressure monitors, pulse oximeters, calorie intake monitors, indirect calorimetry devices, continuous positive airway pressure (CPAP) machines, and drug delivery devices. One or more of these nonanalyte sensors 206 can provide data to the treatment management engine 114, which is further described below. In some aspects, the user can manually provide some of the data for use in... Figure 1 The training server system 140 and / or the treatment management engine 114 are processed.

[0115] In some implementations, the nonanalyte sensor 206 may also include sensors for measuring skin temperature, core temperature, perspiration rate, and / or sweat composition.

[0116] In some embodiments, the nonanalyte sensor 206 can be combined with any other configuration, such as, for example, with one or more continuous analyte sensors 202. As an illustrative example, a nonanalyte sensor, such as a temperature sensor, can be combined with a continuous lactate sensor 202 to form a lactate / temperature sensor for transmitting sensor data to the sensor electronics module 204 using a shared communication circuit. As another illustrative example, a nonanalyte sensor, such as a temperature sensor, can be combined with a multianalyte sensor 202 configured to measure lactate and glucose to form a lactate / glucose / temperature sensor for transmitting sensor data to the sensor electronics module 204 using a shared communication circuit.

[0117] In some implementations, a wireless access point (WAP) can be used to connect one or more of the continuous analyte monitoring system 104, multiple display devices, medical device 208, and / or non-analyte sensor 206 to each other. For example, WAP 138 can provide Wi-Fi and / or cellular connectivity between these devices. Near Field Communication (NFC) and / or Bluetooth can also be used. Figure 2 The devices depicted in Figure 200 are used together.

[0118] Figure 3 Examples of supply schemes based on some of the implementation schemes disclosed herein are provided. Figure 1 The treatment management system uses example inputs and example metrics calculated based on those inputs. Specifically, Figure 3 Provided in Figure 1 More detailed examples of the example inputs and example metrics introduced in the document.

[0119] Figure 3 The diagram illustrates an example input 128 on the left, an application 106 and a DAM 116 in the middle, and an indicator 130 on the right. In some embodiments, each indicator in the indicator 130 may correspond to one or more values, such as discrete numerical values, ranges, or qualitative values ​​(high / medium / low, stable / unstable, etc.). The application 106 receives the input 128 through one or more channels (e.g., manual user input, sensors, other applications executed on display device 107, EMR systems, etc.). As previously mentioned, in some embodiments, the input 128 may be processed by the DAM 116 to output multiple indicators, such as indicator 130. The input 128 and indicator 130 may be used by the training server system 140 and the treatment management engine 114 to train and deploy one or more machine learning models for purposes such as user classification, optimizing exercise parameters to improve the fitness of healthy users and athletes, and other functions described herein.

[0120] In some embodiments, starting with input 128, user statistics may also be provided as input, such as age, height, weight, BMI, body composition (e.g., % body fat or % muscle from computed tomography (CT) scans, magnetic resonance imaging (MRI) scans, dual-energy X-ray absorptiometry (DEXA) scans, or one or more of other information. In some embodiments, user statistics are provided via a user interface, through connection to an electronic source (such as an electronic medical record), or from the measuring device. In some embodiments, the measuring device includes one or more wireless devices, such as a Bluetooth-enabled scale and / or camera, which may communicate, for example, with display device 107 or exercise machine 108 to provide user data.

[0121] In some implementations, treatment / medication information is also provided as input. Medication information may include information about the type, dosage, and / or timing of one or more medications the user will be taking. Treatment information may include information about different lifestyle recommendations made by the user's doctor. For example, the user's doctor may recommend that the user follow specific dietary recommendations, exercise for at least 30 minutes daily, or reduce daily calorie intake by 500 to 1,000 calories to improve liver health. In some implementations, treatment / medication information may be provided through manual input by the user.

[0122] In some embodiments, analyte sensor data may also be provided as input, for example, via the continuous analyte monitoring system 104. In some embodiments, the analyte sensor data may include lactate data (e.g., a user's lactate level) measured by at least one lactate sensor (or multiple analyte sensors) in the continuous analyte monitoring system 104. In some embodiments, the analyte sensor data may include glucose data measured by at least one glucose sensor (or multiple analyte sensors) in the continuous analyte monitoring system 104. In some embodiments, the analyte sensor data may include ketone data measured by at least one ketone sensor (or multiple analyte sensors) in the continuous analyte monitoring system 104. In some embodiments, the analyte sensor data may include potassium data measured by at least one potassium sensor (or multiple analyte sensors) in the continuous analyte monitoring system 104. In some embodiments, the analyte sensor data may include creatinine data measured by at least one creatinine sensor (or multiple analyte sensors) in the continuous analyte monitoring system 104. In some embodiments, the analyte sensor data may include a determination of the user's creatinine to cysteine ​​protease inhibitor C ratio. In some implementations, the analyte sensor data may include time-point creatinine data and / or the time-point creatinine to cysteine ​​protease inhibitor C ratio.

[0123] In some implementations, the input may also be from one or more non-analyte sensors such as reference sensors. Figure 2 The described nonanalyte sensor 206 receives inputs. Inputs from such a nonanalyte sensor 206 may include information related to a user's heart rate, heart rate variability, respiratory rate, oxygen saturation, blood pressure, or body temperature (e.g., to detect illness, physical activity, etc.). In some embodiments, the electromagnetic sensor may also detect low-power radio frequency (RF) fields emitted from objects or tools that are in contact with or near the object, which can provide information about user activity or location.

[0124] In some implementations, input received from the non-analyte sensor may include input related to the user's insulin infusion. Specifically, this input may be received via a wireless connection on a smart pen, via user input, and / or from the insulin pump. Insulin infusion information may include one or more of the following: insulin volume, infusion time, etc. Other parameters, such as the duration of exogenous insulin action or the duration of exogenous insulin action, may also be received as input.

[0125] In some implementations, starting with input 128, food intake information may include information about one or more of meals, snacks, and / or beverages, such as size, content (carbohydrates, fat, protein, etc.), order of intake, and time of intake. In some implementations, users may provide food intake by manual input, by providing photos through an application configured to identify food type and quantity, and / or by scanning barcodes or menus. In various examples, meal quantities may be manually entered as one or more of calories, quantity (e.g., “three cookies”), menu items (e.g., “royal cheese”), and / or food exchange portions (1 piece of fruit, 1 dairy product). In some examples, meal information may be received through a convenient user interface provided by application 106 and / or application 111.

[0126] In some implementations, food intake information (type of food (e.g., liquid or solid, snack or meal, etc.) and / or composition of food (e.g., carbohydrates, fats, proteins, etc.)) can be automatically determined based on information provided by one or more sensors. Some example sensors may include body sound sensors (e.g., abdominal sounds can be used to detect the type of meal, such as liquid / solid food, snack / meal, etc.), radio frequency sensors, cameras, hyperspectral cameras, and / or analyte (e.g., insulin, glucose, lactate, etc.) sensors to determine the type and / or composition of food.

[0127] In some implementations, food intake information can be automatically determined using a combination of manual input (e.g., photo or barcode scan) and sensor data (e.g., analyte sensors, body sound sensors, etc.). Over time, the DAM 116 can identify patterns in food intake based on both manual input and sensor data. Therefore, if the DAM 116 recognizes a similar input, it can accurately determine the contents of the current meal based on the user's past diet.

[0128] In some implementations, food intake recorded by the user may be correlated with lactic acid ingested by the user. Lactic acid ingested may include any natural or engineered food or beverage containing lactic acid (e.g., lactic acid drinks, yogurt, or whole milk) or any natural or engineered food or beverage that produces lactic acid upon ingestion (e.g., fructose drinks). Such lactic acid intake information may be used by DAM 116 to calculate the user's lactic acid clearance rate, as described in more detail with respect to metric 130 calculated by DAM 116.

[0129] In some embodiments, exercise information is also provided as input. Exercise information can be any information about an activity that requires physical activity from the user. For example, the range of exercise information can be information related to low-intensity (e.g., walking a few steps) and high-intensity (e.g., running five miles) physical activity. In some embodiments, exercise information may include information related to HIIT, resistance training, or zone 2 training. In some embodiments, exercise information can also be provided via manual user input, which suggests that the user will begin a specific type of exercise and / or begin with certain exercise parameters. In some embodiments, exercise information may also be provided by the exercise machine 108, including the type, duration, and parameters of the exercise. In some embodiments, exercise information may be provided or determined based on information provided, for example, by non-analyte sensors 206 (e.g., temperature sensors, heart rate monitors, wearable blood pressure monitors, accelerometer sensors on wearable devices such as watches, fitness trackers, and / or patches). In some implementations, exercise information can be provided or determined based on, for example, information provided by the continuous analyte sensor system 104 (e.g., inferences can be made based on a user's lactate, glucose, potassium, and / or ketone data to indicate that the user is exercising). Exercise information provided by both analyte and non-analyte sensors can be used as input to a model trained to predict whether a user is exercising and / or to predict the type and / or parameters of such exercise.

[0130] When predicting whether a user is exercising based on analyte sensor data and non-analyte sensor data, the user can be asked to confirm whether they are exercising, the type of exercise, and / or the level of vigorous activity being used during a specific time period. This data can be used as feedback to retrain the model to understand the user's exercise patterns, and the need for confirmation questions decreases over time as the model optimizes. Other data, such as time and date, location, etc., can similarly be used as input to this model for exercise-related predictions.

[0131] In some implementations, time may also be provided as input, such as the time of day or time from a real-time clock. For example, in some implementations, the input analyte data may be timestamped to indicate to the user the date and time the analyte measurement was performed.

[0132] This can be achieved through the continuous analyte sensor system 104, the non-analyte sensor 206, and / or the user interface (such as... Figure 1 The display device 107 or the user interface of the exercise machine 108 provides user input for any of the inputs 128 mentioned above. As described above, in some embodiments, the DAM 116 determines or calculates the user's metrics 130 based on the inputs 128. Figure 3 The example list of index 130 is shown in the figure.

[0133] In some implementations, lactate levels can be determined based on sensor data, such as lactate measurements obtained from a continuous lactate sensor in the continuous analyte monitoring system 104. For example, lactate levels refer to timestamped lactate measurements or values ​​that are continuously generated and stored over time.

[0134] In some embodiments, the lactate production rate can be determined based on sensor data, such as lactate measurements obtained from a continuous lactate sensor in the continuous analyte monitoring system 104. Specifically, during normal metabolism and exercise, lactate is produced from pyruvate (e.g., glucose is broken down into pyruvate) by lactate dehydrogenase. In some embodiments, the lactate production rate can be determined by assessing an increase in lactate levels over a specified amount of time. In some embodiments, the lactate production rate can be expressed as a percentage of maximum heart rate (e.g., 85% of maximum heart rate) or a percentage of maximum oxygen uptake (e.g., 75%). In some other embodiments, the lactate production rate can be expressed as a function of accelerometer data. For example, accelerometer data can indicate the user's step rate over time (e.g., by increasing accelerometer data to show an increased step rate, and vice versa). Each of these step rates can be correlated with the user's lactate level at a specified time. Therefore, the step rate (e.g., accelerometer data) analyzed over time and its corresponding lactate level can provide information about the user's lactate production rate based on the accelerometer data. DAM 116 can measure the user's lactate production rate continuously, semi-continuously, or periodically over time and store the lactate production rate along with a timestamp in the user's profile 118. The lactate production rate can be timestamped to allow identification of a decrease or increase in the user's lactate production over time.

[0135] In some embodiments, a lactate baseline can be determined based on sensor data, such as lactate measurements obtained from a continuous lactate sensor in a continuous analyte monitoring system 104. The lactate baseline represents the normal lactate level during a period when a user typically expects lactate production to remain stable. A user's baseline lactate is typically expected to remain constant over time unless stimulated by actions such as the user consuming lactate-containing or lactate-producing foods or beverages, or by the user engaging in exercise. Additionally, a user's baseline lactate can also change based on the user's health, particularly in response to improvements or declines in liver health. Furthermore, each user's lactate baseline may differ. In some embodiments, a user's lactate baseline can be determined by calculating an average lactate level over a specified period of time during which lactate is not expected to fluctuate.

[0136] For example, a user's baseline lactate level can be determined during periods when the user is sleeping, sitting in a chair, or during other periods when the user is sedentary and does not ingest foods or medications that would lower or increase lactate levels. In some embodiments, DAM 116 may calculate the lactate baseline and timestamps continuously, semi-continuously, or periodically and store the corresponding information in the user's profile 118. In some embodiments, DAM 116 may use lactate levels measured during periods when the user is sedentary, does not ingest lactate, and there are no external conditions that would affect the lactate baseline to calculate the lactate baseline. In some other embodiments, DAM 116 may use lactate levels measured over a period of time in which the user engages in exercise and / or ingests lactate and / or there are external conditions that would affect the lactate baseline. In such embodiments, for example, DAM 116 may first identify which measured lactate values ​​will be used to calculate the baseline lactate by identifying lactate values ​​that may have been affected by external events, such as food intake, exercise, medications, or other perturbations that would interfere with the capture of the lactate baseline measurement. The DAM 116 can then exclude such measurements when calculating a user's lactate baseline. In some other examples, the DAM 116 can calculate the lactate baseline by first determining the percentage of lactate values ​​measured during that time period, representing the lowest lactate value measured. The DAM 116 can then take the average of that percentage to determine the lactate baseline level.

[0137] In some embodiments, the lactate clearance rate can be determined based on sensor data (e.g., lactate measurements obtained from a continuous lactate sensor in the continuous analyte monitoring system 104). Specifically, the user's lactate clearance rate indicates the rate at which lactate production exceeds lactate intake. The lactate clearance rate can indicate metabolic fitness (e.g., the slope of a lactate clearance curve can indicate metabolic fitness). In some embodiments, the lactate clearance rate can be determined by calculating the slope between an initial lactate value (e.g., during a period of increased lactate levels) and a lactate baseline associated with the user. In some embodiments, the lactate clearance rate can be calculated over time until the user's lactate level reaches a value relative to the user's lactate baseline (e.g., 50% or 75% of the lactate baseline). In some embodiments, the lactate clearance rate can be calculated over time until the user's lactate level reaches a value relative to a peak lactate level measured for the user at a previous time (e.g., the user's lactate level reaches 25%, 50%, and / or 75% of the user's peak lactate level).

[0138] In some implementations, the lactate clearance rate can be expressed as a function of the user's lactate half-life. Specifically, there is an inverse relationship between the lactate clearance rate and the lactate half-life. In increasingly healthy livers, as the calculated lactate half-life decreases, the slope of lactate clearance increases (e.g., the magnitude of the slope increases negatively). As liver function improves, the slope of lactate clearance further increases (e.g., the magnitude of the slope increases negatively), and the calculated lactate half-life further decreases. Conversely, in diseased livers, as the calculated lactate half-life increases, the slope of lactate clearance decreases (e.g., the magnitude of the slope increases negatively). Therefore, the lactate half-life can indicate the user's lactate clearance rate. The lactate clearance rate calculated over time can be timestamped and stored in the user's profile 118.

[0139] In some embodiments, lactate trends can be determined based on lactate levels over a certain time period. In some embodiments, lactate trends can be determined based on lactate production rates over a certain time period. In some embodiments, lactate trends can be determined based on calculated lactate clearance rates over a certain time period.

[0140] In some implementations, a lactate threshold may be determined based on sensor data (e.g., lactate measurements obtained from the continuous analyte monitoring system 104). The lactate threshold indicates a user's lactate level at which lactate production exceeds lactate clearance, potentially caused by the user engaging in high-intensity, anaerobic activity. Each user's lactate threshold may differ. In some implementations, the lactate threshold may be determined by identifying the lowest lactate level over a predetermined period of time during which lactate levels do not rise rapidly. In some implementations, the lactate threshold may be determined by identifying the highest lactate level before the user increases their work rate, causing a rapid rise in lactate levels. In some implementations, the user's lactate threshold may be determined during a period of time during which the user engages in physical activity (such as Zone 2, resistance training, or HIIT). In some implementations, DAM 116 may continuously or periodically calculate the lactate threshold and timestamps and store the corresponding information in the user's profile 118.

[0141] In some implementations, the intensity at the lactate threshold can be determined based on sensor data (e.g., lactate measurements obtained from the continuous analyte monitoring system 104) combined with exercise and / or non-analyte sensor data. The intensity at the lactate threshold can be an intensity that meets the lactate threshold (e.g., force, speed, power, etc.). For example, the lactate threshold can be the intensity just before the user's lactate production begins to exceed lactate clearance (as measured by the continuous lactate sensor of the continuous analyte monitoring system 104). In some implementations, the intensity at the lactate threshold can be determined over a period of time during which the user engages in physical activity (such as Zone 2, resistance training, or HIIT). In some implementations, DAM 116 can continuously or periodically calculate the intensity and timestamp at or just before reaching the lactate threshold and store the corresponding information in the user's profile 118.

[0142] In some embodiments, the initial lactate level at the start of a workout phase can be determined based on sensor data (e.g., lactate measurements obtained from the continuous analyte monitoring system 104) combined with exercise and / or non-analyte sensor data. The initial lactate level at the start of the workout can be the user's lactate level when the user begins the workout phase. In some embodiments, the initial lactate level can be determined based on the lactate level at the start of the workout, before the user's lactate begins to increase and / or decrease, or when the user inputs information related to the start of the workout phase. In some embodiments, the initial lactate level at the start of a workout phase can be determined over a period of time during which the user participates in a workout (such as Zone 2, resistance training, or HIIT). In some embodiments, DAM 116 can continuously or periodically calculate the initial lactate level and timestamp at the start of the workout phase and store the corresponding information in the user's profile 118.

[0143] In some implementations, glucose levels can be determined based on sensor data (e.g., blood glucose measurements obtained from a continuous lactate sensor in the continuous analyte monitoring system 104).

[0144] In some implementations, blood glucose trends can be determined based on glucose levels over a certain period of time.

[0145] In some implementations, historical data, real-time data, or a combination thereof can be used to determine insulin sensitivity, and insulin sensitivity can be based on one or more of the following inputs: food intake information, continuous analyte sensor data, non-analyte sensor data (e.g., insulin delivery information from an insulin device). Insulin sensitivity refers to the degree to which a user's cells respond to insulin. Improving a user's insulin sensitivity can help reduce the user's insulin resistance.

[0146] In some implementations, insulin in vivo can be determined using non-analyte sensor data inputs (e.g., insulin delivery information) and / or (e.g., from user data) known or learned insulin time-effect curves, which may take into account basal metabolic rate (e.g., insulin renewal to maintain bodily functions) and insulin usage driven by activity or food intake.

[0147] In some implementations, health and disease indicators may be determined, for example, based on one or more user inputs (e.g., pregnancy information or known disease information) from physiological sensors (e.g., temperature), activity sensors, or combinations thereof. In some implementations, based on the values ​​of health and disease indicators, for example, a user's state may be defined as one or more of healthy, sick, resting, or fatigued.

[0148] In some implementations, it may be based, for example, on the... Figure 1 The illustrated treatment management engine 114 provides one or more user inputs or outputs to determine disease staging indicators. For example, the development of liver disease may be collected because it can affect a user's metabolic fitness. In some implementations, example disease stages of liver disease may include an inflammatory phase (e.g., an early stage of enlargement or inflammation of the user's liver), a fibrotic phase (e.g., a stage with signs of scar tissue in the inflamed liver), a cirrhotic phase (e.g., a stage with severe signs of scar tissue in the inflamed liver), and end-stage liver disease (ESLD).

[0149] In some implementations, meal status indicators can indicate a user's state in terms of food intake. For example, meal status can indicate whether the user is in a fasting state, a pre-meal state, a meal state, a post-meal response state, or a stable state. In some implementations, meal status can also indicate the nutritional content of body loads (e.g., ingested meals, snacks, or beverages) and can be determined, for example, based on food intake information, meal timing information, and / or digestibility information, which may be related to food type, quantity, and / or order (e.g., which food / beverage was consumed first).

[0150] In some implementations, the eating habit index is based on the content and timing of a user's meals. For example, in one example, if the eating habit index is in the range of 0 to 1, the better / healthier the user eats, the higher the value of the user's eating habit index will be, up to 1. Furthermore, in another example, the more consistently a user's food intake adheres to a specific schedule, the closer their eating habit index will be to 1. In some implementations, the eating habit index may be based on the meals, snacks, or beverages a user consumes within a certain time period to indicate whether the user is consistently participating in a ketogenic diet (e.g., a low-carbohydrate, moderate-protein, and high-fat diet).

[0151] In some implementations, activity level indicators can indicate a user's activity level. In some implementations, activity level indicators are determined, for example, based on input from an activity sensor or other physiological sensor, such as nonanalyte sensor 206. In some implementations, activity level indicators can be calculated by DAM 116 based on one or more of inputs 128, such as exercise information, nonanalyte sensor data (e.g., accelerometer data), time, user input, etc. In some implementations, activity level can be represented as the user's step rate. Activity level indicators can be timestamped so that they can be correlated with the user's lactate levels simultaneously.

[0152] In some implementations, exercise program metrics may indicate one or more of the following: the type of activity the user participates in, the corresponding intensity of such activity, and the frequency with which the user participates in such activity. In some implementations, exercise program metrics may be calculated based on one or more of the following: analyte sensor data input and / or non-analyte sensor data input (e.g., non-analyte sensor data input from accelerometers, heart rate monitors, blood pressure monitors, respiratory rate sensors, etc.), calendar input, user input, etc.

[0153] In some embodiments, metabolic rate is an indicator that may include basal metabolic rate (e.g., energy consumed at rest) and / or active metabolism (e.g., energy consumed by activity, such as physical activity). In some examples, basal metabolic rate and active metabolism may be tracked as separate outcome indicators. In some embodiments, metabolic rate may be calculated by DAM 116 based on one or more of inputs 128 (e.g., exercise information, analyte sensor data, non-analyte sensor data, time, etc.). In some embodiments, metabolic rate may be calculated, and the metabolic rates calculated over time may be timestamped and stored in a user profile 118.

[0154] In some implementations, metabolic fitness is a measure of overall metabolic fitness based on changes in a user's lactate response over time at various exercise intensities, liver function, and / or the ability of the liver and skeletal muscle to effectively clear lactate. Metabolic fitness can be determined, for example, by considering the user's baseline lactate, initial lactate level at the start of exercise, lactate concentration at various exercise intensities, and lactate clearance rate over certain time periods, and comparing these metrics to defined (e.g., population-based) thresholds and / or ranges for baseline lactate, initial lactate level at the start of exercise, lactate concentration at various exercise intensities, and lactate clearance rate corresponding to various metabolic fitness levels (e.g., healthy users, metabolically challenged users, athletes, etc.). Furthermore, improvements or deteriorations in metabolic fitness can be determined by comparing the user's current baseline lactate, initial lactate level at the start of exercise, and lactate clearance rate over certain time periods with the user's past baseline lactate, initial lactate level at the start of exercise, and lactate clearance rate over certain time periods. In some implementations, metabolic fitness can be calculated over time and can be timestamped and stored in the user's profile 118.

[0155] In some implementations, the fat loss metric can be calculated by DAM 116 based on input 128 and more specifically on non-analyte sensor data (such as indirect calorimetry). Indirect calorimetry assesses the user's energy expenditure and predicts glucose, lipid, and protein utilization based on total energy expenditure. The fat loss metric can be predicted based on the utilization of macronutrients (e.g., protein, fat, or carbohydrates) and the duration of exercise. Furthermore, the fat loss metric can inform the user which macronutrients to consume after the exercise phase for optimal results (e.g., the user should consume protein instead of glucose or lactate after the exercise phase).

[0156] In some implementations, potassium levels and indicators can be calculated by DAM 116 based on input 128. Potassium levels can be determined from sensor data (e.g., potassium measurements obtained from continuous analyte monitoring system 104). Potassium levels can also be used to determine potassium indicators such as absolute maximum potassium levels, individualized maximum potassium levels, duration of potassium above a specific threshold, individualized zone 2 potassium ranges, and / or potassium change rates.

[0157] In some implementations, maximum potassium level refers to the maximum potassium level of a user determined to be unsafe over a given time period (e.g., hourly, weekly, daily, etc.). In some implementations, the absolute maximum potassium level may be consistent across all users (e.g., set at 5.5 mmol / L based on current medical guidelines). In some other implementations, each patient may have a different absolute maximum potassium level (e.g., an individualized maximum potassium level). In some implementations, the duration of potassium above a certain threshold may be the duration of potassium above either the absolute maximum potassium level for all users or the individualized maximum potassium level. In some implementations, the individualized Zone 2 potassium range may be a potassium range indicating the user's Zone 2 exercise intensity based on the user's historical data. In some implementations, the rate of change of potassium level refers to the rate at which one or more timestamped potassium measurements or values ​​change relative to one or more other timestamped potassium measurements or values. The rate of change of potassium level may be determined over a period of one or more seconds, one or more minutes, one or more hours, one or more days, etc.

[0158] Typically, different users have different physiological functions, so optimizing training phases may require different training parameters to help users achieve their goals. For example, training parameters that may be effective for healthy users aiming to maintain or improve fitness may not be effective for athletes aiming to increase peak performance or increase lactate threshold. Therefore, as regarding Figure 4 As described, classifying users by type before optimizing exercise for them (e.g., providing optimized exercise guidance and / or controlling the operation of exercise machines) ensures that exercise parameters will be effective for a particular user to achieve that particular user's specific goals.

[0159] Figure 4 This is a flowchart depicting an example method 400 for classifying users into healthy users, athletes, or metabolically impaired users according to certain embodiments of this disclosure. In some embodiments, workflow 400 uses at least one continuous lactate monitor to monitor the user (e.g., Figure 1 The users listed are categorized.

[0160] Workflow 400 can be executed by the treatment management system 100, which includes the treatment management engine 114. For example... Figure 4 As shown, workflow 400 begins at box 402 with the treatment management engine 114 receiving input data (i.e., input) to categorize users. Input can be received in various ways. For example, input can be received or retrieved from user profile 118, which includes demographic information 120, disease progression information 122, medication information 124, inputs 128, indicators 130, etc. Input can also be received as user input via the user interface of display device 107, user interface 113, and / or exercise machine 108.

[0161] For example, users can be asked and provided with user classification (e.g., user opinions on whether they are metabolically inept, healthy, or athletic), user goals related to health or exercise, user medical history, user demographics (age, sex, ethnicity, etc.), user physiological information (e.g., height, weight, etc.), user historical exercise data (e.g., exercise routine information, including, for example, how many times a user exercises per week, what types of exercise the user performs, and the typical duration of exercise), and metabolic adaptation. In some implementations, demographic information may be correlated with determining a user's metabolic fitness (e.g., a user's ability to clear lactate and glucose after meals or exercise, and / or liver health). For example, age can indicate metabolic fitness because metabolic fitness changes with age. In another example, adults with sarcopenia (e.g., those who have experienced age-related skeletal muscle degeneration) have lower metabolic fitness because they have less skeletal muscle capable of processing blood lactate compared to adults without sarcopenia.

[0162] In some implementations, a user's metabolic adaptations (such as AMP-activated protein kinase (AMPK) activation) can be used to determine the user's ability to consume glucose and oxidize lipids, and thus classify the user. When AMPK is activated, it regulates cellular metabolism, which stimulates glucose uptake and lipid oxidation to produce energy. The treatment management engine 114 can process historical exercise-related data and determine whether AMPK is activated by determining the total amount of exercise performed within a certain time period (e.g., one week). The total amount of exercise can be calculated using the duration and intensity of the exercise, where intensity can be a function of power, speed, incline, heart rate, lactate level, etc. The treatment management engine 114 can determine that AMPK is activated by the user when the total amount of exercise calculated in a future exercise period is equal to or greater than the total amount of exercise in one or more past exercise periods.

[0163] In some implementations, the treatment management engine 114 may determine AMPK activation based on the user's glucose levels. For example, if the user's glucose levels indicate that the user has not eaten for six hours and has completed the exercise phase, the treatment management engine 114 may determine that the user's AMPK is activated. In another example, if the user's glucose levels indicate that the user has not eaten for 24 hours, the treatment management engine 114 may determine that the user's AMPK is activated. In some implementations, if the user's glucose levels remain low for three hours (e.g., indicating that the user has not eaten), the treatment management engine 114 may instruct the user to continue fasting until at least six hours and / or complete the exercise phase to activate AMPK.

[0164] In some implementations, a user's historical exercise data may include physiological data associated with the user's previous exercise phases, including initial lactate levels at the start of exercise, changes in lactate levels, the rate of lactate change during exercise, and the user's resting baseline lactate levels. Physiological data associated with a user's previous exercise phases may also include other types of analyte data (e.g., glucose data, potassium data, etc.) and non-analyte data (e.g., heart rate data, temperature data, etc.) that can be used for user classification.

[0165] In some implementations, users can be categorized using additional analyte data and / or non-analyte data. For example, the treatment management engine 114 can monitor a user's cortisol levels, sleep data (e.g., total sleep time, sleep efficiency, sleep latency, awakenings after sleep onset, number of awakenings, apnea and hypopnea, body movement, REM sleep, non-REM sleep, snoring, polysomnography (PSG), body movement recorder, subjective sleep quality, sleep environment factors, glucose levels, respiratory rate, etc.), resting oxygen saturation, heart rate, heart rate variability, etc., to determine the user's stress level, exercise readiness, etc. The user's exercise readiness can assist the treatment management engine 114 in determining the user's category. Additionally, a user's stress level can alter their metabolic adaptation, for example, by affecting the user's insulin action and / or glucose release and metabolism.

[0166] In some implementations, a user's genetic profile information may be received as input to classify the user. For example, epigenetic indicators and / or changes in epigenetic indicators over time, such as a user's DNA methylation and / or miRNA expression, may be provided to the treatment management engine 114. Epigenetic indicators may change as the user completes more training phases and / or receives training protocols. The user's genetic information can assist the treatment management engine 114 in classifying the user.

[0167] At box 404, the treatment management engine 114 may determine, based on the received input, whether a user can be categorized as a metabolically unfit user, a healthy user, or an athlete. As an example, if certain types of input (such as historical exercise data, including lactate metrics associated with historical exercise phases) are unavailable, the treatment management engine 114 may determine that the user cannot be categorized, in which case the treatment management engine 114 proceeds to box 406. However, if the user (e.g., based on user-reported classifications, a set of goals reported by the user, or historical exercise data) is categorizable, the treatment management engine 114 may proceed to categorize the user based on the input received at box 402.

[0168] At box 405, the treatment management engine 114 categorizes the user based on the input received at box 402 (e.g., using a rule-based model or an AI / ML model). In implementations using a rule-based model, a rule base can be used, for example, to map the user's input to a specific category. For example, a rule-based model might take the input received at box 402 and categorize the user as a metabolically unhealthy user at box 410, an average healthy user at box 412, or an athlete at box 414.

[0169] Examples of rules could include classifying a user as either a healthy user or an athlete if the input indicates that the user self-identifies as either a healthy user or an athlete. Another example rule for classifying users could be based on the amount of exercise a user reports weekly (e.g., 150 minutes of moderate-intensity aerobic activity or 75 minutes of vigorous-intensity aerobic activity per week could classify a user as a healthy user, while more than 200 minutes of moderate-intensity aerobic activity or 100 minutes of vigorous-intensity aerobic activity per week could classify a user as an athlete). Another example rule could include classifying a user as either a healthy user or an athlete if the input indicates that the user does not have liver disease. Yet another example rule could include classifying a user as either a healthy user or an athlete if the input includes past exercise data showing lactate levels mapped to either a healthy user or an athlete, as described in more detail with respect to box 408.

[0170] In some implementations, the rules can be more granular, allowing the combination of rules and multiple inputs to output a classification. As an example of such a rule, a user can be classified as a healthy user if they self-identify as healthy, have no significant comorbidities, and / or their exercise data reflects lactate levels indices indicative of health. If a user is classified as healthy, the treatment management engine 114 proceeds to box 412 to base its classification on information about... Figure 5A The described implementation scheme optimizes workouts for users. Alternatively, if the user is categorized as an athlete, the treatment management engine 114 proceeds to box 414 to optimize workouts based on information about... Figure 6A The described implementation scheme optimizes workouts for users. Note that the implementation scheme described herein focuses on optimizing workouts for average healthy users and athletes.

[0171] Similarly, the rule-based model described above can be used instead to classify users as metabolically unhealthy. For example, as described with respect to box 408, users can be classified as metabolically unhealthy using a lactate index indicated by the user's historical exercise data.

[0172] In some implementations, an AI / ML (or, for simplicity, interchangeably, "ML model") model can be used instead of a rule-based model to predict user classification. For example, some of the inputs received at box 404 can be used as inputs to a model trained to classify the corresponding user. In such cases, a training dataset is used to train the model, which includes historical group-based data of many users who have already been classified as metabolically unfit users, healthy users, or athletic users. In this example, the training dataset is labeled with such classifications. In some implementations, the model's output may be accompanied by a confidence score. If the confidence score is below a certain threshold, the treatment management engine 114 can determine that the user is unclassifiable and proceed to box 406.

[0173] At box 406, if the user is not categorized based on the input at box 402, the user can be guided through a trial exercise phase to monitor lactate measurements, heart rate, exercise duration, maximum speed or force output, etc., and the user's lactate response to the exercise can be used to determine the user's categorization. For example, the treatment management engine 114 can suggest that the user complete a trial exercise phase for a specific duration (e.g., 15 minutes) at a gradually increasing intensity (e.g., increasing speed, power, heart rate, incline, etc. by specific metrics per minute). In an implementation using an exercise machine, the treatment management engine 114 can automatically instruct the exercise machine 108 to begin the trial exercise phase based on the gradually increasing intensity described above. For all users completing the trial exercise phase, the intensity can be increased at the same rate to accurately compare the user's lactate response (e.g., with that of other users). The treatment management engine 114 can then monitor the user's lactate response to the exercise.

[0174] At box 408, the treatment management engine 114 can be configured to classify users into metabolically underperforming users at box 422, average healthy users at box 424, or athletes at box 426. Users can be classified using different lactate metrics obtained from the trial training phase, including initial lactate levels at the start of training, the rate of lactate change throughout the trial training phase, the absence of lactate “valleys,” and the intensity / time at which a user reaches their lactate threshold. Other information, such as the input received at box 402, can also be used to classify users, as described below.

[0175] In some implementations, the treatment management engine 114 may use a rule-based model to classify users. As described above, such a rule-based model may include rules defined based on lactate levels obtained during the trial training phase. For example, such rules may define the range of lactate levels at the start of training, the rate of change in lactate throughout the trial training, the presence of lactate “valleys,” and the intensity / time it takes for a user to reach their lactate threshold, based on empirical studies involving population-based data. In some implementations, user classification may include providing users with scores based on their fitness within the classification. For example, in addition to classification, users classified as athletes aiming to improve endurance may also receive scores (e.g., scores between 1 and 10, with 10 being optimal) to further monitor the user’s endurance improvement over time. In another implementation, users classified as athletes aiming to improve strength or muscle recovery time may receive scores based on the user’s specific goals.

[0176] As an example of such a rule, if a user begins a trial training phase and their lactate level is between 1 mmol and 1.5 mmol at the start of the training, and the rate of change of lactate is slightly negative or slightly positive, the treatment management engine 114 may infer that the user is a healthy user or an athlete. However, the treatment management engine 114 may instruct the user to continue the training phase to monitor other lactate patterns. As the user continues to increase the intensity during the trial training, the user may continue to experience a near-zero rate of change of lactate for a certain period of time. However, at some point, as the intensity continues to increase, the rate of change of lactate may begin to increase at a high positive rate of change of lactate. A high positive rate of change of change is related to the user's lactate threshold, which is where lactate production exceeds the rate of lactate clearance. If a user has an initial lactate level of approximately 1 mmol to 1.5 mmol, and the user experiences a near-zero rate of change of lactate for a certain period of time before reaching their lactate threshold, the user may be a healthy user or an athlete. Identifying a user as a healthy user or an athlete may be based on the classification ranges discussed below.

[0177] The classification range can be based on the intensity at which a user reaches their lactate threshold or maximum lactate steady state, as well as other considerations. Therefore, the treatment management engine 114 can consider the intensity and / or time since the start of exercise at which the user reaches their lactate threshold (e.g., the rate of lactate change rapidly becomes positive after approaching zero). The treatment management engine 114 can then classify the user as a healthy user or an athlete based on the intensity and the amount of time it takes for the user to reach their lactate threshold. If, for example, the user's lactate threshold occurs between 125 watts and 250 watts of intensity, the user can be classified as a healthy user. For example, a healthy user might have a lactate threshold at an intensity of 200 watts or at a speed of 7 miles per hour on a treadmill. On the other hand, if, for example, the user's lactate threshold occurs at an intensity above 250 watts, the user can be classified as an athlete. For example, an athlete might have a lactate threshold at an intensity of 260 watts or at a speed of 9 miles per hour on a treadmill.

[0178] In addition to the rules described above based on lactate levels obtained during the trial training phase, rule-based models may also include other rules based on any information provided as part of user profile 118 (e.g., BMI, age, etc.) and / or information provided through user input as described above. For example, rule-based models may also use rules based on non-analyte sensor data (e.g., heart rate, respiratory rate, etc.) as well as the duration and intensity of the trial training phase to determine user classification.

[0179] In some implementations, instead of rule-based models or in addition to rule-based models, an ML model may be used to classify users based on lactate levels obtained during the trial exercise phase and / or other inputs received at box 402. In some implementations, in addition to lactate levels and inputs, the model may also receive environmental data as input to classify users. Environmental data may include altitude during the trial exercise phase, as higher altitudes result in reduced available oxygen, and therefore, users may produce more lactate during exercise. For example, some or all of the lactate levels obtained during the trial exercise phase, environmental data, and / or inputs received at box 404 may be used as inputs to a model trained to classify corresponding users. In such cases, a training dataset is used to train the model, which includes historical group data of many users who have already been classified at least based on their corresponding lactate levels as metabolically unfit users, healthy users, or athletic users. In this example, the training dataset is labeled with such classifications.

[0180] In some implementations, although a user may not have completed a workout phase within an extended timeframe (e.g., 1-2 weeks), the user may have past workout data stored in the historical database 112, including past trial workout phase data and past classifications. The treatment management engine 114 may consider workout consistency when classifying or reclassifying a user. For example, a user may be classified as a healthy user based on past classifications and / or a series of workout phases. However, after a period of inactivity, the treatment management engine 114 may reclassify the user to determine whether the user would still be classified as a healthy user after a period of inactivity.

[0181] Figures 5A to 5C Example methods are described for optimizing workout phases and providing healthy users with feedback on fitness, post-workout nutrition or glucose administration, the effectiveness of workout phases, etc. As an example, after a user is categorized as a healthy user, the treatment management engine 114 can be configured to guide the user through a series of workout phases to improve fitness or achieve the user's goals. For example, a user can use app 106 or 111 to indicate that their goal is to improve their fitness and request guidance for a specific workout phase (e.g., the first workout phase in a series). In this example, app 106 or 111 can then present a user interface for the user to select whether they want to perform Zone 2 training, resistance training, or HIIT for their upcoming workout phase.

[0182] Method 500 begins at box 502, where the treatment management engine 114 determines whether the user has already selected HIIT, resistance training, or a zone 2 workout type. If the user has already selected zone 2, the treatment management engine 114 proceeds directly to box 504. If the user has already selected HIIT, the treatment management engine 114 proceeds to the section about... Figure 5B The described method. If the user has already selected resistance training, the treatment management engine 114 will proceed to... Figure 5C The method described.

[0183] If the user has not yet selected or specified an exercise type, at box 522, the treatment management engine 114 can suggest, for example, Zone 2 exercises to users whose goals are fat loss and lowering blood sugar levels. Specifically, the treatment management engine 114 can consider the user's lactate level at the start of the exercise when determining whether to recommend HIIT, resistance training, or Zone 2 exercise. For example, if the user is a healthy user who typically has a lactate level of 1.5 mmol / L at the start of an exercise, but on this day, the user's lactate level is 2 mmol / L (e.g., due to alcohol intake the previous day), the treatment management engine 114 can suggest that the user complete a Zone 2 exercise phase (instead of a HIIT exercise phase) to lower their lactate level.

[0184] In some implementations, if the user is under time constraints, the treatment management engine 114 may suggest HIIT or resistance training. In some implementations, the treatment management engine 114 may suggest primary resistance training combined with some Zone 2 exercises to users aiming to build more muscle and / or increase resting metabolic rate. In some implementations, if the user's goal is to train specific muscle groups for various sports and / or competitions, the treatment management engine 114 may suggest resistance training. In some implementations, based on the user's medical history, such as that stored in the user profile 118, the treatment management engine 114 may recommend exercise phases unlikely to exacerbate the current injury. For example, if the user reports having undergone ankle surgery, the treatment management engine 114 may recommend low-impact Zone 2 exercises or upper body-focused resistance training phases. Additionally, if the user has a prescribed physical therapy regimen based on recent surgery or injury, the treatment management engine 114 may recommend that the user incorporate the physical therapy regimen into the user's recommended exercise phases.

[0185] In other implementations, the treatment management engine 114 may suggest a recovery day (e.g., not completing any exercise phase) based on the user's lactate level at the start of exercise or based on user input. For example, if the user's lactate level at the start of exercise is typically 1.5 mmol / L, but the user's current lactate level at the start of exercise is 2.5 mmol / L, and / or the user's input indicates that the user completed an intense exercise phase the previous day (e.g., based on the user's perception), the treatment management engine 114 may suggest that the user not exercise or may suggest that the user perform low-intensity exercise, such as walking, to promote muscle recovery. Especially in healthy users, high lactate levels may be due to overuse and indicate a need for recovery; therefore, the treatment management engine 114 may suggest a recovery day.

[0186] In some implementations, the treatment management engine 114 may provide a workout readiness score to the user based on the user's lactate level at the start of the workout compared to the average lactate level at the start of the workout, information about the user's workout phases over previous days, information about the user's perceived activity during previous workout phases, and so on. For example, the treatment management engine 114 may provide a low workout readiness score if the user has completed a series of high-intensity workouts, reported high perceived activity, and / or had a higher-than-average lactate level at the start of the workout. Alternatively, the treatment management engine 114 may provide a high workout readiness score if the user has not completed a workout phase over several days and the user's lactate level at or below average at the start of the workout.

[0187] In other implementations, the treatment management engine 114 may recommend HIIT to users whose goal is to improve their resting metabolic rate over time (e.g., improve long-term energy expenditure) or when the user may have limited time to complete a workout phase.

[0188] In some implementations, the treatment management engine 114 may recommend a primary Zone 2 workout phase (e.g., 75%-80% of the workout phase) alongside some HIIT or resistance training workout phases (e.g., 20%-25% of the workout phase) to improve peak performance, build muscle mass, and (e.g., improve metabolic fitness by improving mitochondrial health). In some implementations, the treatment management engine 114 may recommend HIIT, resistance training, or Zone 2 workouts, but the user may choose to reject the recommendations. For example, there may be a workout phase where the treatment management engine 114 recommends Zone 2 for fat burning based on the user's goals, but the user wants to focus on improving strength during that workout phase (e.g., through HIIT or resistance training). In this example, the user may choose to ignore the recommendations and select the desired type of workout based on short-term goals. However, if the user selects Zone 2 at box 502, or if the treatment management engine 114 suggests Zone 2 to the user at box 522, the treatment management engine 114 proceeds to box 504.

[0189] At box 504, the treatment management engine 114 can determine whether the user has completed a data-available Zone 2 training phase in the past by referring to the user's profile 118. For example, the treatment management engine 114 can examine the training data in the user profile 118 to determine whether there is data related to one or more of the user's past Zone 2 training phases.

[0190] At box 506, if the user has previously completed a Zone 2 exercise phase with available data, the treatment management engine 114 determines personalized exercise guidance for the user based on the user's own historical exercise data. For example, exercise data from the user profile 118 can be used to determine the set of exercise parameters for the user's exercise phase. The parameter set may include (e.g., exercise intensity, exercise duration, etc., at the start of the phase and throughout the phase). Exercise intensity itself can be a function of speed, resistance, inclination, etc. Providing personalized guidance based on the user's own historical exercise data can be done in several ways. For example, a rule-based model can be used in conjunction with the user's own historical exercise data to provide the set of exercise parameters for the user's exercise phase. Alternatively, one or more ML models trained based on the user's own historical data can be used to provide the set of exercise parameters for the user's exercise phase. For example, the training server system 140 can retrieve user-specific exercise data from the user profile 118 to train one or more ML models.

[0191] At box 508, if the user has not previously completed a Zone 2 exercise phase for which data is available, the treatment management engine 114 may use a population-based (non-personalized) model to determine exercise parameters for the user, at least until the user has performed one or more exercise phases and individual exercise data is available to the user. For example, the treatment management engine 114 may use a rule-based model that defines Zone 2 exercise parameters based on empirical studies involving population data. For instance, Zone 2 exercise parameters to achieve Zone 2 lactate levels may be determined using rules based on parameters that are effective for user groups similar to the user (e.g., based on demographic and / or physiological variables such as age, sex, weight, height, BMI, etc.).

[0192] Alternatively or concurrently, the treatment management engine 114 may use one or more ML models to provide users with exercise parameters to achieve Zone 2 lactate levels. For example, the ML model may be trained based on population-based training data associated with healthy users achieving Zone 2 lactate levels. The dataset includes data records, each containing exercise parameters, analyte data, non-analyte data, and / or other relevant information from the corresponding user's profile 118 for each exercise phase. As an example, each data record in the training dataset may include timestamped exercise parameters and corresponding timestamped analyte and non-analyte data for the user's exercise phase. Data records may be labeled with one or more exercise parameters. Using such a training dataset, a model can be trained to predict one or more exercise parameters to help healthy users achieve Zone 2 lactate levels.

[0193] Before using a personalized or population-based model to guide the user through the exercise phases based on the exercise parameters determined at boxes 506 and 508, the treatment management engine 114 may instruct the user to begin exercising at a warm-up intensity for a certain duration. In some embodiments, the intensity may be half of the intensity later provided to the user to reach the Zone 2 lactate range. In some embodiments, the duration of the warm-up intensity may be approximately 5 or 10 minutes.

[0194] At box 510, once the warm-up duration is complete, the treatment management engine 114 can instruct the user and / or the exercise machine to increase the intensity to the exercise parameters determined in box 506 or 508.

[0195] In embodiments where a user is completing a workout phase on an exercise machine (e.g., exercise machine 108), the treatment management engine 114 can automatically set or gradually increase the intensity (e.g., speed, incline, and / or resistance). The treatment management engine 114 can automatically set the exercise machine to the exercise parameters determined at boxes 506 and 508. In embodiments where the user is not using the exercise machine, the treatment management engine 114 can instruct the user to increase the intensity of their workout by gradually increasing various exercise parameters until an optimal Zone 2 lactate range is achieved. Increasing intensity can be achieved by changing various exercise parameters associated with the exercise machine (e.g., exercise machine 108) or by expending additional effort (in the form of speed or power) when the user is not using the exercise machine.

[0196] Typically, a healthy user may have an initial lactate level of approximately 1.5 mmol / L at the start of exercise. Then, unlike a user with poor metabolism, the rate of lactate change may be slightly negative or slightly positive, but approaches zero at approximately 2 mmol / L, representing the optimal Zone 2 lactate range. Therefore, the treatment management engine 114 is configured to continuously monitor the user's physiological parameters to determine the optimal exercise intensity that will allow the user to reach and maintain a lactate range with a rate of change of zero or near zero. For example, if the user does not exercise at sufficient intensity, the user will not reach the optimal Zone 2 lactate range. Alternatively, if the user exercises at excessively high intensity, the user will rapidly exceed the Zone 2 lactate range and continue to experience lactate increases beyond the desired range. Therefore, for the treatment management engine 114, it is crucial to continuously monitor the user's lactate levels (including the rate of lactate change) and physiological parameters (e.g., heart rate, respiratory rate, glucose levels, power, speed, strength, accelerometer data, etc.) to instruct the user to increase or decrease intensity to maintain within the optimal Zone 2 lactate range, as further described with respect to boxes 512 and 514.

[0197] At box 512, the treatment management engine 114 can determine, for example, based on the user's lactate levels, whether the user is already within or approaching the expected lactate level in Zone 2. If the user is already within or approaching the lactate level in Zone 2 (e.g., the rate of lactate change is close to zero), the treatment management engine 114 can instruct the user to maintain the current intensity.

[0198] Note that the expected Zone 2 lactate range (also referred to herein as the defined range) may be, for example, between 1 mmol and 2 mmol. In another example, the expected Zone 2 lactate range may be a range relative to the user's lactate level at the start of exercise. For example, the Zone 2 lactate range may be 0.5 mmol to 1.5 mmol higher than the user's lactate level at the start of exercise. However, the range may differ for different users; therefore, the treatment management engine 114 is configured to observe lactate levels and determine whether the user is approaching or nearing a zero rate of lactate change, indicating that the user is in or will be in the Zone 2 steady-state lactate range. To determine whether the user is approaching or nearing a zero rate of lactate change, the treatment management engine 114 may use one of a variety of rule-based or ML models to obtain the user's lactate level and / or other lactate indicators (e.g., rate of change) as well as other physiological parameters to predict whether the user is soon approaching or nearing a zero rate of change.

[0199] In some implementations, the treatment management engine 114 can determine whether a user has reached Zone 2 based on potassium levels. For example, based on the user's historical data, the treatment management engine 114 can determine an individualized potassium level range corresponding to the user's Zone 2 steady-state lactate range. During future training phases, the treatment management engine 114 can determine whether the user is within the Zone 2 lactate range based on whether the user's potassium levels are within the individualized Zone 2 potassium range.

[0200] In some implementations, as described above, determining whether a user has reached a certain metabolic state (e.g., Zone 2) is primarily accomplished by monitoring the user's lactate levels. However, in some cases, the user's lactate levels may be unavailable, and / or lactate measurements received from a continuous lactate sensor may be delayed relative to real-time lactate blood measurements. To address potential unavailability or lag, in some implementations, the treatment management engine 114 may additionally or alternatively use some of the user's physiological parameters (e.g., heart rate, respiratory rate, glucose levels, power, speed, strength, accelerometer data, etc.) as substitutes for lactate data. Such physiological parameters may be directly correlated with lactate levels and thus can be used to estimate real-time lactate levels during the exercise phase when such lactate levels are unavailable. Thus, during the first exercise phase where a continuous lactate sensor is available, lactate measurements can be obtained and correlated with the user's corresponding physiological parameters. These physiological parameters can then be used as substitutes for lactate levels indicating certain metabolic states (e.g., Zone 2, resistance training, HIIT) during future exercise phases where lactate information is unavailable.

[0201] At box 514, the treatment management engine 114 can observe lactate indicators (e.g., lactate levels and trends) and / or the user's physiological parameters, and if the user does not tend to achieve the optimal Zone 2 lactate range, the treatment management engine 114 can instruct the user to adjust the exercise intensity. For example, if the user does not exercise at a sufficient intensity to reach the Zone 2 lactate range (e.g., 1.8 mmol–2 mmol), or if the user exercises at too high an intensity that would cause the user's lactate level to reach the Zone 2 lactate range but subsequently increase beyond the desired range, it is considered that the user does not tend to reach the desired lactate range.

[0202] Various models (e.g., rule-based, machine learning, or predictive algorithms) can be used to determine optimal exercise parameters, thus establishing an ideal intensity to ensure the user's lactate level reaches the optimal Zone 2 lactate range without exceeding it. In some implementations, rule-based models can be utilized, where various rules can be defined around a set of parameters, including the user's current lactate level, current rate of lactate change, time since exercise began, exercise intensity, and other physiological parameters (e.g., heart rate, respiratory rate, glucose levels, power, speed, strength, accelerometer data, etc.). For example, a sample rule might specify that if the user's current lactate level is X, the current rate of lactate change is Y, and the time since exercise began is Z, then the intensity should be Q. Q can be the set of exercise parameters that bring the user into the Zone 2 range, such as speed, resistance level, altitude, power, etc. Rules can be made more granular and involve the entire user with other physiological and demographic indicators. For example, a rule could state that if a user is 40 years old and has heart rate A, temperature B, current lactate level X, current rate of lactate change Y, and time since the start of exercise Z, then the intensity should be Q.

[0203] In some other implementations, one or more prediction algorithms or models may be used to predict the optimal intensity to achieve the desired Zone 2 lactate range. For example, one or more models, which may be the same as or different from the personalized or population-based models described with respect to boxes 506 and 508, may be run continuously to take input sets (e.g., the user's current lactate level, current rate of lactate change, time since the start of exercise, and other inputs received at box 402) and / or other physiological parameters (e.g., heart rate, respiratory rate, glucose indices, power, speed, strength, accelerometer data, etc.) and output the optimal intensity and / or corresponding exercise parameters for achieving the optimal Zone 2 lactate range.

[0204] When instructing a user to increase or decrease intensity, the therapy management engine 114 can check whether the user is exercising outdoors. For example, non-analyte sensor data from a temperature sensor can indicate that the user is exercising outdoors, or the user's selection of, for example, outdoor running or walking as the exercise type can be used by the therapy management engine 114 to determine that the user is exercising outdoors. When the user is exercising outdoors, the therapy management engine 114 can consider air temperature, air quality, known allergens, humidity, wind speed, solar energy index, and altitude when suggesting increasing or decreasing intensity.

[0205] For example, based on sensor data from a temperature sensor, the treatment management engine 114 can notify the user that the environment is too hot and that exercising at the usual intensity is unsafe (e.g., recommending an exercise intensity as a function of temperature to reach the Zone 2 lactate range, even if the recommended intensity is lower / higher than historical data), or that the user should stop exercising.

[0206] Furthermore, the treatment management engine 114 can instruct the user to reduce intensity or stop exercising based on the correlation between potassium levels and temperature sensor data at higher outdoor temperatures. For example, even if the user perceives themselves to be at the correct intensity in high temperatures, potassium levels may increase beyond expected levels due to the heat. If the user perceives low to moderate activity but the potassium sensor indicates increased potassium levels, the user may be at risk of cardiac events (e.g., arrhythmias, cardiac arrest, etc.). Therefore, when the user's potassium levels increase beyond expected levels, the treatment management engine 114 can instruct the user to reduce intensity or stop exercising in high temperatures.

[0207] Furthermore, the treatment management engine 114 can use the correlation between lactate levels at a specific intensity level and temperature sensor data to provide the user with more accurate intensity instructions. For example, at the same intensity, lactate levels may be higher at higher temperatures. Therefore, at high temperatures, the treatment management engine 114 may instruct the user to exercise at a lower intensity compared to the intensity required under typical temperature conditions to achieve the Zone 2 lactate range. Conversely, at lower temperatures, lactate response may be lower, requiring increased intensity to reach the optimal Zone 2 lactate range. Without considering temperature sensor data, at high temperatures, the treatment management engine 114 might instruct the user to increase intensity, which could cause the user to increase lactate beyond the optimal Zone 2 lactate range, leading to heatstroke or other safety issues. At cooler temperatures, the treatment management engine 114 may not consider the higher intensity required to reach the Zone 2 lactate range when compared to typical temperatures that result in less efficient workout phases.

[0208] Furthermore, humidity levels captured from wearable sensors or from web / cloud-based measurements of local weather services can be used to further calculate the expected heat transfer coefficient corresponding to exercise intensity relative to temperature, and the expected corresponding lactate level for a given intensity. The treatment management engine 114 can correct for factors such as temperature, atmospheric pressure, local wind speed and / or humidity, and solar energy intensity or its absence during nighttime. Similar to the humidity data, other environmental factors can be captured from locally worn sensors or from expected parameters captured from cloud-based data on local weather at the exercise user's location, which can be manually entered into the system, automatically calculated based on sensor input or time measurements, or identified using GPS data from the user's location on one or more devices.

[0209] In addition to temperature and humidity, the therapy management engine 114 can also consider altitude when providing intensity instructions when the user is exercising outdoors. For example, if the user is jogging outdoors and approaching an upcoming hill, the therapy management engine 114 can instruct the user to maintain or even reduce the intensity. Failure to do so could result in the user exercising at an excessively high intensity when the hill factor is added. This intensity could cause the user's lactic acid levels to exceed the optimal Zone 2 lactic acid range. Therefore, the therapy management engine 114 can consider changes in altitude or anticipated changes in altitude based on GPS and / or map data and adjust the intensity recommendations accordingly.

[0210] Furthermore, when a user is exercising outdoors, the treatment management engine 114 can also take altitude into account when providing intensity instructions. For example, lactate levels increase when a user is at high altitudes, making it potentially more difficult to maintain a Zone 2 lactate range. However, as the user adapts to the altitude, lactate levels may decrease. In addition to providing more accurate exercise intensity based on altitude (e.g., suggesting lower intensity for achieving a Zone 2 lactate range at higher altitudes), the treatment management engine 114 can also determine when a user is likely to experience altitude sickness based on lactate levels.

[0211] Furthermore, when users are exercising outdoors, they can complete exercise phases (e.g., running, walking, cycling, etc.) on a specific route once or multiple times. In this case, the treatment management engine 114 can identify the route based on GPS data and past exercise data, and provide the user with information on where they should increase or decrease the intensity to maintain certain areas of lactate levels in Zone 2.

[0212] Furthermore, regardless of whether the user is exercising indoors or outdoors, the treatment management engine 114 can determine whether the user is exhibiting signs of cardiac stress based on cardiac indicators (e.g., via a heart rate monitor or ECG). For example, the treatment management engine 114 can monitor various cardiac indicators to detect signs of atrial flutter, tachycardia, atrial fibrillation, increased heart rate, etc. If the user experiences abnormal cardiac indicators and high lactate levels due to exercise, the treatment management engine 114 can instruct the user to reduce the intensity and / or stop exercising and seek medical care for potential cardiac complications.

[0213] After suggesting an increase or decrease in intensity at box 514, the treatment management engine 114 may return to box 512 to determine whether the user is within the Zone 2 lactate range, or whether the user's lactate level and rate of change indicate that the user tends to reach lactate levels within the Zone 2 lactate range. If not, the treatment management engine 114 returns to box 514 to adjust the intensity accordingly. If yes, the treatment management engine 114 proceeds to box 516.

[0214] At box 516, the treatment management engine 114 can instruct the user to maintain the current intensity for a specified duration (e.g., 5 minutes or 10 minutes). The specified duration can be the ideal length of a Zone 2 workout for the user's goals. The specified duration can also be based on the most effective length of workouts for the user based on past workout data or the most effective length for historical users with similar demographics to the user.

[0215] At box 518, once the duration is complete, for non-diabetic users, the treatment management engine 114 can guide the user to complete a cool-down at a lower intensity (e.g., walking) and monitor lactate levels (or physiological parameters) to ensure that the user's lactate levels have returned to baseline levels (e.g., 1.5 mmol). Active cooling allows the body to clear excess lactate generated due to exercise and optimizes post-workout fat burning.

[0216] For users with type 1 diabetes, the treatment management engine 114 may recommend that the user complete a cool-down at a lower intensity (e.g., walking) and / or ingest specific post-workout nutrition and / or administer glucose to prevent present or potential hypoglycemia. In some embodiments, the treatment management engine 114 may determine whether the user is taking insulin to manage their diabetes and may not instruct users not taking insulin to administer glucose.

[0217] For users with type 2 diabetes, in addition to recommending a cool-down, the treatment management engine 114 may also recommend avoiding insulin administration. Specifically, aerobic exercise (such as Zone 2) lowers blood glucose levels. If a user experiences hypoglycemia after the exercise phase, the treatment management engine 114 may determine whether the user's glucose level is dangerously low or expected to become dangerously low (e.g., less than 70 mg / dL). If the treatment management engine 114 determines that the user's glucose level is dangerously low or expected to become dangerously low, it may advise the user to consume post-workout nutrition (which may include protein or electrolyte drinks) while avoiding glucose intake. However, if glucose levels are low after the exercise phase, the user may be instructed to consume a small amount of food to increase their glucose levels and to continue monitoring glucose levels after the exercise phase. If the user is administering insulin, they may also be instructed to consume glucose after the Zone 2 exercise phase to avoid hypoglycemia. High-sugar meals or drinks (e.g., Gatorade) may not be recommended, as they can also cause a surge in lactate levels, which inhibits the cool-down process, reduces fat oxidation, and decreases overall exercise effectiveness.

[0218] Users with type 1 or type 2 diabetes who are not taking insulin may be instructed to perform a cool-down at 60%-80% of the intensity required to reach their lactate threshold. The treatment management engine 114 may instruct the user to perform a cool-down at 60%-80% of the intensity required to reach the user's lactate threshold based on analyte (e.g., lactate) and / or non-analyte data. In some embodiments, the treatment management engine 114 may instruct the user to complete the cool-down at a perceived low intensity and monitor the user's lactate level until the user's lactate level returns to baseline. In some embodiments, the treatment management engine 114 may optimize the user's cool-down duration based on analyte data (including glucose and / or lactate data) from the user's historical exercise phases to avoid hyperglycemic spikes. For example, the treatment management engine 114 may identify a user who experiences hyperglycemia after a 10-minute cool-down, where the user's glucose level is 100 mg / dL at the end of the exercise phase. During a subsequent exercise phase, if the user's glucose level is 100 mg / dL after the exercise phase, the treatment management engine 114 may recommend that the user complete a 20-minute cool-down.

[0219] Further in box 518, the treatment management engine 114 can provide users with feedback on the effectiveness of fitness and exercise phases, as well as recommendations for future exercise phases, dietary recommendations, liver health status, and estimated fat oxidation and weight loss.

[0220] For example, at box 518, users can be provided with feedback related to fitness (or its improvement) and the effects of the exercise phase, as well as recommendations for future exercise phases, dietary recommendations, liver health status, and estimated fat oxidation and weight loss.

[0221] For example, improvements or deteriorations in metabolic fitness can be determined by a user's ability to reach Zone 2 lactate levels at a similar intensity in future workout sessions. For instance, if a user required a specific speed or power to reach Zone 2 lactate levels in past workout sessions, the treatment management engine 114 can monitor whether that intensity is effectively used to reach Zone 2 lactate levels in future workout sessions. Over time, if a user requires greater intensity (e.g., speed or power) to reach Zone 2 lactate levels, the user may have improved metabolic fitness because their liver and skeletal muscles become more efficient at clearing lactate from the body at greater intensity.

[0222] In some embodiments, the feedback provided to the user at box 518 may include rewards and / or gamification features. As the user progresses closer to their goal (e.g., a certain number of workout phases, some improvement in metabolic fitness, etc.), the user may receive various badges, stars, and / or other rewards. In some embodiments, the feedback provided to the user at display device 107 may be configured, for example, to connect to the user's social media accounts, whereby the user can share updates and rewards with followers and friends via their social media accounts. Allowing the user to share with others via social media can provide support and encouragement to the user and encourage them to continue their workout program.

[0223] Additionally, feedback related to metabolic fitness can be based on changes in a user's lactate response to exercise over time. For example, when a user completes a workout phase, the treatment management engine 114 can provide feedback on improvements in lactate clearance rates (e.g., during cool-down) over time, indicating improved metabolic fitness. Feedback on improved metabolic fitness can also be based on lactate clearance over time relative to whether the user completed a cool-down after the workout phase, as cool-down promotes lactate clearance. The treatment management engine 114 can determine the intensity and duration of a cool-down while monitoring the user's lactate clearance rate over time. In some embodiments, the treatment management engine 114 can instruct the user to ingest a specific amount of lactate when not exercising (e.g., through an oral lactate tolerance test) to determine the lactate clearance rate and improvements in the lactate clearance rate when there is no muscle consumption of lactate during exercise. Improvements in the lactate clearance rate in response to the same amount of lactate and in the absence of exercise will indicate improvements in the user's metabolic fitness, and more specifically, improvements in the user's liver health.

[0224] In some implementations, feedback on the effectiveness of the exercise phase may include monitoring calorie burning and total energy expenditure after exercise. In some implementations, the metabolic equivalent (MET) equation or the Harris-Benedict equation may be used to determine calorie burning and total energy expenditure.

[0225] Additionally, the treatment management engine 114 can provide user-specific dietary recommendations based on the user's lactate response to Zone 2 exercise. For example, dietary recommendations may include specific macronutrients to consume or meal times relative to the exercise phase. For instance, after an exercise phase, the user may be recommended to consume protein within a specified timeframe (e.g., within 20 minutes of exercise) to build muscle mass. In another example, the user may be instructed not to consume lactate, glucose, or fructose, or to limit their intake of lactate, glucose, or fructose within a specified timeframe (e.g., 2 hours), in order to maintain higher levels of lipid oxidation.

[0226] In addition, the treatment management engine 114 can provide feedback related to the user's liver health. Liver health feedback may include providing the user with information about the diagnosis and / or improvement / worsening of liver disease. Information about the improvement / worsening of liver disease may be obtained based on monitoring liver disease progression using pre- or post-exercise lactate measurements (e.g., fasting lactate, overnight baseline lactate, post-exercise lactate clearance rate, lactate levels during exercise), other analyte levels (e.g., liver lipid content, liver enzyme status, glycogen storage status, bilirubin levels, or insulin resistance), and liver health tests (e.g., determining liver fat via MRI-PDFF, determining liver inflammation via ALT / AST levels, determining fibrosis activity via Pro-C3 testing, determining clotting time via international normalized ratio testing, or detecting liver elasticity via MRE imaging).

[0227] Additionally, the treatment management engine 114 can predict fat oxidation at rest and after exercise and provide the user with information about the predicted fat oxidation. The treatment management engine 114 can consider user input (e.g., the user's diet choices, sleep, medical history, or medications taken) and past exercise data (e.g., exercise duration) to determine the metabolic response to exercise and fat oxidation. The metabolic response can be the user's ability to clear lactic acid and glucose from the body after exercise compared to past workouts. Fat oxidation can be determined based on the proportion of carbohydrate and lipid levels to total calorie expenditure (e.g., using indirect calorimetry). The lipid level portion of total calorie expenditure can indicate the user's fat oxidation due to exercise.

[0228] In some implementations, the treatment management engine 114 may associate lower lactate levels during a later exercise phase compared to a previous exercise phase with greater fat oxidation. Because blood lactate accumulation can be negatively correlated with fat oxidation, lower lactate accumulation resulting from exercise and / or an improved lactate clearance rate will lead to greater expected fat oxidation. For example, a user may experience maximum fat oxidation between 1.8 mmol / L and 2.1 mmol / L. Therefore, if a user's baseline lactate is above 2.1 mmol / L, a decrease in lactate levels during this period can be observed, and the intensity level should be increased to bring the lactate level back to approximately 2.1 mM to achieve maximum fat oxidation. Additionally, a lower baseline lactate level after an exercise phase can indicate improved fat oxidation at rest. For example, if a user has a baseline lactate level of 3 mmol / L at the start of the first exercise phase, and their baseline lactate level increases to 2 mmol / L due to dietary and / or exercise guidance, the treatment management engine 114 can determine that the user has achieved improved fat oxidation.

[0229] The Treatment Management Engine 114 can also measure a user's metabolic rate after exercise and provide the user with information on their metabolic rate or improvements in metabolic rate. The Treatment Management Engine 114 can predict post-exercise metabolic rate using peak lactate, lactate clearance rate, and lactate change rate. The Treatment Management Engine 114 can also measure a user's metabolic rate across a range of exercise phases using indirect calorimetry, the Weir equation, or various analyte measurements (e.g., glucose, free fatty acids, β-hydroxybutyrate, and glycerol) to determine improvements in metabolic rate. Improvements in metabolic rate can reflect improved liver health, improved overall health, and weight loss over time.

[0230] The treatment management engine 114 can also be configured to interface with a continuous positive airway pressure (CPAP) machine that can be used as an indirect calorimeter. If a user uses a CPAP machine daily, utilizing the CPAP machine as an indirect calorimeter may be beneficial for collecting consistent indirect calorimetric readings. For example, the CPAP can be configured as an indirect calorimeter to measure the metabolic use of a specific fuel (e.g., fat or carbohydrate) via a carbon dioxide (CO2) sensor and a flow sensor to determine the rate of CO2 production. In a CPAP machine, a CO2 sensor and a flow sensor can be added to the CPAP mask or hose to measure oxygen intake from inhaled air and CO2 produced from exhaled air during sleep or rest. Using a CPAP machine to measure nighttime energy expenditure can be used for comparison with a user's metabolic rate and energy expenditure after exercise.

[0231] The therapy management engine 114 can also be configured to receive input from the user after the exercise phase. For example, the user may be prompted to provide input to the therapy management engine 114 via the user interface of display device 107, the user interface 113 of exercise machine 108, etc. The therapy management engine 114 may request the user's perceived activity level after the Zone 2 phase. In some embodiments, for example, the user may be given the option to select their perceived activity level during the exercise phase based on the Borg scale of perceived activity (e.g., "I feel I can continue at this pace," "I have reached my limit," or "I feel nauseous and cannot continue at this intensity"). Receiving input from the user regarding their perceived activity level can assist the therapy management engine 114 in suggesting an increase or decrease in intensity for future exercise phases. For example, if the user provides input that they can continue at that pace after exercise, the therapy management engine 114 may suggest that the user exercise at a higher intensity in future exercise phases if the user does not reach the optimal Zone 2 lactate range until 20 minutes of exercise. Higher intensity will allow the user to reach the Zone 2 lactate range more quickly, thereby maximizing the exercise effect without causing overexertion.

[0232] However, in some cases, users taking beta-blockers may have a perceived activity level higher than it actually is. For example, a user taking beta-blockers might perceive themselves as overactive and provide input on this situation, but the user's lactate measurement could signal the treatment management engine 114 that the user was within the Zone 2 lactate range during the perceived overactivity. In this example, the treatment management engine 114 is unlikely to adjust the intensity of future workout phases based on the user's perceived activity level.

[0233] In some cases, potassium levels can be used to determine when a user has experienced hyperactivity during or after a workout. For example, a user may have experienced hyperactivity if their potassium level reaches an absolute maximum, an individualized maximum, or exceeds a set duration above the maximum potassium level. In another example, a user may also have experienced hyperactivity if their potassium level shows a high rate of change (which could indicate that their potassium level will eventually exceed an absolute or individualized maximum potassium level).

[0234] At box 520, once the user has completed Zone 2 exercise, the treatment management engine 114 can utilize information about the user acquired during the exercise phase (e.g., analyte data, non-analyte data, user input, etc.) to optimize future exercise phases. For example, this information can be used in one or more rule-based models at box 506 during future exercise phases. In another example, this information can be used to personalize the ML models utilized at boxes 506 or 508 (e.g., training / retraining these ML models). Therefore, for future exercise phases, the treatment management engine 114 may be able to more effectively predict the optimal exercise parameters that will enable a particular user to reach the Zone 2 lactate range.

[0235] As described above, at least in some implementations, optimization of the exercise phase for a user can be achieved by monitoring the user's response to exercise and providing intensity guidance based on the user's lactate index obtained using a continuous analyte sensor system (e.g., continuous analyte monitoring system 104). However, in some cases, the lactate index may be unavailable and / or may be subject to a time lag (e.g., the lactate index may not indicate real-time blood lactate levels). In cases where the lactate index is unavailable and / or subject to a lag, certain physiological parameters directly related to the lactate index can be used as substitutes for the lactate index. These substitute parameters (referred to herein as physiological parameters) can serve as estimates of real-time lactate levels. Physiological parameters may include heart rate, respiratory rate, glucose, power, speed, strength, data from an accelerometer, etc. For example, during a previous exercise phase, when the user reaches the Zone 2 lactate range, the user may be at a heart rate of 120 beats per minute (as determined by post-exercise analysis taking into account the lag in lactate measurement). During future training phases, heart rate or other real-time physiological parameters can be used to estimate the user's real-time lactate level, so that when the user reaches a heart rate of 120 beats per minute, the treatment management engine 114 can assume that the user has reached the Zone 2 lactate range.

[0236] exist Figure 5B If, at box 522, the treatment management engine 114 determines that the user has already selected a HIIT workout type or has already recommended HIIT based on the user's goals, time commitments, etc., then the method can begin at box 524. At box 524, the treatment management engine 114 can determine whether the user has completed any previously available HIIT phases by referring to the user's profile 118. For example, the treatment management engine 114 can examine the workout data in the user profile 118 to determine if there is data related to one or more of the user's past HIIT phases.

[0237] At box 526, if the user has previously completed a HIIT phase for which data is available, the treatment management engine 114 determines personalized exercise guidance for the user based on the user's own historical exercise data. For example, exercise data from the user profile 118 can be used to determine the set of exercise parameters for the user's exercise phase. The parameter set may include (e.g., exercise intensity, exercise duration, etc., at the start of the phase and throughout the phase). Exercise intensity itself can be a function of speed, resistance, inclination, etc. Providing personalized guidance based on the user's own historical exercise data can be done in several ways. For example, a rule-based model can be used in conjunction with the user's own historical exercise data to provide the set of exercise parameters for the user's exercise phase. Alternatively or additionally, one or more ML models trained based on the user's own historical data can be used to provide the set of exercise parameters for the user's exercise phase. For example, the training server system 140 can retrieve user-specific exercise data from the user profile 118 to train one or more ML models.

[0238] At box 528, if the user has not previously completed a HIIT phase for which data is available, the treatment management engine 114 may use a non-personalized, population-based model to determine the exercise parameters for the user, at least until the user has performed one or more exercise phases and personal exercise data is available to the user. For example, the treatment management engine 114 may use a rule-based model that defines HIIT exercise parameters based on empirical studies involving population data. For instance, HIIT exercise parameters may be determined based on rules that are effective for user groups similar to the user (e.g., based on demographic and / or physiological variables such as age, sex, weight, height, BMI, etc.) to achieve a HIIT intensity range.

[0239] Alternatively or concurrently, the treatment management engine 114 may use one or more ML models to provide users with exercise parameters to achieve a HIIT intensity range (i.e., lactate levels in the HIIT zone). For example, the ML model may be trained based on population-based training data associated with healthy users who have achieved a HIIT intensity range. The dataset includes data records, each containing exercise parameters, analyte data, non-analyte data, and / or other relevant information from the corresponding user's profile 118 for each exercise phase. As an example, each data record in the training dataset may include timestamped analyte and non-analyte data for the user's exercise phase. Data records may be labeled with one or more exercise parameters. Using such a training dataset, a model can be trained to predict one or more exercise parameters to help the user achieve a HIIT intensity range.

[0240] Before using a personalized or population-based model to guide the user through the exercise phases based on the exercise parameters determined at boxes 526 and 528, the therapy management engine 114 may instruct the user to begin exercising at a warm-up intensity for a certain duration. In some embodiments, the intensity may be half of the intensity later provided to the user to reach the HIIT intensity range. In some embodiments, the duration of the warm-up intensity may be approximately 5 minutes or 10 minutes.

[0241] At box 530, once the warm-up duration is complete, the treatment management engine 114 can instruct the user to increase the intensity to the recommended exercise parameters determined in boxes 526 or 528.

[0242] In some implementations, a user may be completing an exercise phase on an exercise machine (e.g., exercise machine 108), and the treatment management engine 114 may automatically set or gradually increase the intensity (e.g., speed, incline, and / or resistance, etc.). The treatment management engine 114 may automatically set the exercise machine to the exercise parameters determined at boxes 526 and 528. In implementations where the user is not using the exercise machine, the treatment management engine 114 may instruct the user to increase the intensity of their exercise by gradually increasing various exercise parameters until the desired HIIT intensity range is achieved. Increasing intensity may be achieved by changing various exercise parameters associated with the exercise machine (e.g., exercise machine 108) or by expending additional effort (in the form of speed or power) when the user is not using the exercise machine.

[0243] Typically, healthy users who have not recently consumed a diet high in lactate or lactate precursor metabolites may have an initial lactate level of approximately 1.5 mmol / L at the start of exercise. Then, unlike users with poor metabolism, the rate of change in lactate may be slightly negative or slightly positive, but close to zero, as the rate of lactate production is roughly the same as the rate of lactate clearance in the body. However, given the high-intensity nature of HIIT exercise, the lactate slope may not show an initial slightly negative or slightly positive rate of change, but rather begin to increase rapidly. To ensure that users reach and do not exceed the HIIT intensity range, the treatment management engine 114 is configured to continuously monitor the user's lactate levels and physiological parameters to determine the optimal exercise intensity for reaching the HIIT intensity range. For example, if the user does not exercise at sufficient intensity, the user will not reach the optimal HIIT intensity range. Alternatively, if the user exercises at excessively high intensity, the user may not be able to maintain the exercise intensity to the optimal HIIT duration. Therefore, for the treatment management engine 114, it is crucial to continuously monitor the user's lactate levels (including the rate of lactate change) and physiological parameters instruct the user to increase or decrease the intensity to achieve the optimal HIIT intensity range, as further discussed in boxes 532 and 534.

[0244] At box 532, the treatment management engine 114 can determine whether the lactate level reflects or is expected to be close to the user's desired HIIT intensity range. For example, the treatment management engine 114 can monitor values ​​associated with the area under the curve (AUC) of lactate over time, the increase in lactate relative to the user's baseline lactate, and / or the rate of lactate increase to determine whether the user has reached or is expected to reach the HIIT intensity range (also referred to herein as the defined range). For healthy users, reaching, for example, 5 mmol–15 mmol of lactate corresponding to the AUC can indicate that the user has reached the desired HIIT intensity range.

[0245] To determine whether a user is approaching a physiological state corresponding to the HIIT intensity range, the treatment management engine 114 can use one of a variety of rule-based models or ML models to obtain the user's lactate levels and other physiological parameters to predict whether the user is about to approach a physiological state corresponding to the desired HIIT intensity range.

[0246] At box 534, the treatment management engine 114 can observe lactate levels and, if the user is not inclined to achieve the optimal HIIT intensity range, instruct the user to increase or decrease the exercise intensity. For example, if the user does not exercise at sufficient intensity to reach the HIIT intensity range, or if the user exercises at too high an intensity that would cause the user's lactate levels to increase beyond the desired HIIT intensity range, it is considered that the user is not inclined to reach the desired lactate range.

[0247] Various models (e.g., rule-based, machine learning, or predictive algorithms) can be used to determine optimal exercise parameters, thus establishing an ideal intensity to ensure the user's lactate levels reach the desired HIIT intensity range without exceeding it. In some implementations, rule-based models can be utilized, where various rules can be defined around a set of parameters, including the user's current lactate level, current rate of lactate change, time since exercise began, exercise intensity, and other parameters. For example, an example rule might state that if the user's current lactate level is X, current rate of lactate change is Y, and time since exercise began is Z, then the intensity should be Q. Q can be the set of exercise parameters that bring the user into the HIIT intensity range, such as speed, resistance level, altitude, power, etc. Rules can become more granular and involve the entire user, along with other physiological and demographic indicators. For example, a rule might state that if the user is 40 years old and has heart rate A, temperature B, current lactate level X, current rate of lactate change Y, and time since exercise began Z, then the intensity should be Q.

[0248] In some other implementations, one or more prediction algorithms or models may be used to predict the optimal intensity to achieve the desired HIIT intensity range. For example, one or more models, which may be the same as or different from the personalized or population-based models described with respect to boxes 526 and 528, may be run continuously to obtain a set of inputs (e.g., the user's current lactate level, current rate of lactate change, time since the start of the workout, and other inputs received at box 402) and output the optimal intensity and / or corresponding workout parameters for achieving the desired HIIT intensity range.

[0249] When instructing a user to increase or decrease the intensity, the therapy management engine 114 can check whether the user is exercising outdoors. For example, non-analyte sensor data from a temperature sensor can indicate that the user is exercising outdoors, or the user's selection of, for example, outdoor running or walking as the exercise type can be used by the therapy management engine 114 to determine that the user is exercising outdoors. When the user is exercising outdoors, the therapy management engine 114 can determine air temperature, air quality, humidity, wind speed, known allergens, and altitude when suggesting increasing or decreasing the intensity.

[0250] For example, based on sensor data from a temperature sensor, the treatment management engine 114 can notify the user that the environment is too hot and that exercising at the usual intensity is unsafe (e.g., recommending an exercise intensity as a function of temperature to reach the HIIT intensity range, even if the recommended intensity is lower / higher than historical data), or that the user should stop exercising.

[0251] Furthermore, the treatment management engine 114 can instruct the user to reduce intensity or stop exercising based on the correlation between potassium levels and temperature sensor data at higher outdoor temperatures. For example, even if the user perceives themselves to be at the correct intensity in high temperatures, potassium levels may increase beyond expected levels due to the heat. If the user perceives low to moderate activity but the potassium sensor indicates increased potassium levels, the user may be at risk of cardiac events (e.g., arrhythmias, cardiac arrest, etc.). Therefore, when the user's potassium levels increase beyond expected levels, the treatment management engine 114 can instruct the user to reduce intensity or stop exercising in high temperatures.

[0252] Furthermore, the treatment management engine 114 can use the correlation between lactate levels at a specific intensity level and temperature sensor data to provide users with more accurate intensity instructions. For example, at the same intensity, lactate levels may be higher at higher temperatures. Therefore, at high temperatures, the treatment management engine 114 may instruct the user to exercise at a lower intensity compared to the intensity required under typical temperature conditions to achieve the HIIT intensity range. Conversely, at lower temperatures, lactate response may be lower, requiring increased intensity to reach the optimal HIIT intensity range. Without considering temperature sensor data, at high temperatures, the treatment management engine 114 might instruct the user to increase intensity, potentially causing the user to increase lactate levels beyond the optimal HIIT intensity range, leading to heatstroke or other safety issues. At cooler temperatures, the treatment management engine 114 may not consider the higher intensity required to reach the HIIT intensity range when compared to typical temperatures that result in less efficient workout phases.

[0253] Furthermore, humidity levels captured from wearable non-analyte sensors or from network / cloud-based measurements of local weather services can be used to further calculate the expected heat transfer coefficient corresponding to exercise intensity relative to temperature, and the expected corresponding lactate level for a given intensity. The treatment management engine 114 can correct for factors such as temperature, atmospheric pressure, local wind speed and / or humidity, and solar energy intensity or its absence during nighttime. Similar to the humidity data, other environmental factors can be captured from locally worn sensors or from expected parameters captured from cloud-based data on local weather at the exercise user's location, which can be manually entered into the system, automatically calculated based on sensor input or time measurements, or identified using GPS data from the user's location on one or more devices.

[0254] In addition to temperature and humidity, the therapy management engine 114 can also consider altitude when providing intensity instructions when the user is exercising outdoors. For example, if the user is jogging outdoors and approaching an upcoming hill, the therapy management engine 114 can instruct the user to maintain or even reduce the intensity. Failure to do so could result in the user exercising at an excessively high intensity when the hill factor is added. This intensity could cause the user's lactic acid levels to exceed the optimal HIIT intensity range. Therefore, the therapy management engine 114 can consider changes in altitude or anticipated changes in altitude based on GPS and / or map data and adjust the intensity recommendations accordingly.

[0255] Furthermore, when users are exercising outdoors, they may complete exercise phases (e.g., running, walking, cycling, etc.) on a specific route once or multiple times. In such cases, the therapy management engine 114 can identify routes based on GPS data and past exercise data, and provide certain locations where the user should increase or decrease intensity to maintain the HIIT intensity range. Additionally, exercise time can be recommended based on environmental conditions and other factors measured from the user's previous activities, using weather, time of day, and other indicators related to the outdoor environment, indicating the time of day and / or day with optimal environmental conditions indicating best results.

[0256] Furthermore, regardless of whether the user is exercising indoors or outdoors, the treatment management engine 114 can determine whether the user is exhibiting signs of cardiac stress based on cardiac indicators (e.g., via a heart rate monitor or ECG). For example, the treatment management engine 114 can monitor various cardiac indicators to detect signs of atrial flutter, tachycardia, atrial fibrillation, increased heart rate, etc. If the user experiences abnormal cardiac indicators and high lactate levels due to exercise, the treatment management engine 114 can instruct the user to reduce the intensity and / or stop exercising and seek medical care for potential cardiac complications.

[0257] After suggesting an increase or decrease in intensity at box 534, the treatment management engine 114 may return to box 532 to determine whether the user has reached or is approaching the expected HIIT intensity range. If not, the treatment management engine 114 returns to box 534 to adjust the intensity accordingly. If yes, the treatment management engine 114 proceeds to box 536.

[0258] At box 536, the treatment management engine 114 can instruct the user to maintain the current intensity for a specified duration (e.g., 5 minutes). The specified duration can be the ideal length for a HIIT phase targeting the user's goals. The specified duration can also be based on the most effective length of exercise for the user based on past exercise data or the most effective length for historical users with similar demographics to the user.

[0259] At box 538, once the duration is complete, for non-diabetic users, the treatment management engine 114 can guide the user to complete the cool-down at a lower intensity (e.g., walking) and monitor lactate levels (or physiological parameters) to ensure that the user's lactate levels have returned to baseline levels (e.g., 1.5 mmol). Active cool-down allows the body to clear excess lactate produced due to exercise and optimizes post-workout fat burning.

[0260] For users with type 1 or type 2 diabetes, the treatment management engine 114 can guide them to perform a cool-down at a lower intensity (e.g., walking) and / or consume specific post-workout nutrients to prevent current or potential hyperglycemia. Specifically, anaerobic exercise (such as HIIT) may cause a high lactate spike during exercise and a post-workout glucose spike when the user has cleared excess lactate from their body. To counteract the post-workout glucose spike, the treatment management engine 114 can monitor glucose levels during and after exercise to determine when the user is likely to experience hyperglycemia. If the user is likely to experience hyperglycemia or is currently experiencing hyperglycemia, the treatment management engine 114 can recommend that the user consume a lactic acid beverage and / or a meal containing lactate and glucose to increase glucose absorption.

[0261] It can also guide diabetic users to avoid insulin in response to high glucose levels. If a user actively administers insulin, it can lead to "insulin buildup" if the user administers insulin while the body is still clearing lactate. When the body clears the remaining lactate and begins to clear glucose, the combination of glucose clearance and insulin further promoting glucose uptake can lead to dangerous hypoglycemia.

[0262] If, after the HIIT phase, a user's glucose level is low but does not require immediate action (e.g., 70 mg / dL or 3.9 mmol / L), glucose intake or insulin administration is not recommended because the user will experience a glucose spike once their liver and skeletal muscles have cleared excess lactic acid during the active cool-down. Alternatively, if a user's glucose level is dangerously low (e.g., 54 mg / dL), they may be instructed to consume small amounts of glucose and monitor post-exercise glucose levels to avoid hyperglycemia. High-sugar meals or beverages (e.g., Gatorade) may not be recommended as they can cause a surge in lactic acid levels, inhibiting the cool-down process, reducing fat oxidation, and decreasing overall workout effectiveness.

[0263] Further at box 538, the treatment management engine 114 can provide the user with feedback on fitness (e.g., metabolic fitness) and the effects of the exercise phase, as well as recommendations for future exercise phases, dietary recommendations, liver health status, and estimated fat oxidation and weight loss.

[0264] For example, feedback associated with metabolic fitness may include changes in a user's lactate response to exercise over time. As an example, when a user completes a workout phase, the treatment management engine 114 may provide feedback on improvements in lactate clearance rates over time (e.g., during / after HIIT workouts and / or during cool-downs), indicating improved metabolic fitness. Feedback on improved metabolic fitness may also be based on lactate clearance over time relative to whether the user completes a cool-down after a workout phase, as cool-downs promote lactate clearance. The treatment management engine 114 may determine the intensity and duration of a cool-down while monitoring the user's lactate clearance rate over time. In some embodiments, the treatment management engine 114 may instruct the user to ingest a specific amount of lactate when not exercising (e.g., an oral lactate tolerance test) to determine the lactate clearance rate and improvements in the lactate clearance rate when there is no muscle consumption of lactate during exercise. Improvements in the lactate clearance rate in response to the same amount of lactate and the absence of exercise will indicate improvements in the user's metabolic fitness.

[0265] In some implementations, feedback on the effectiveness of the exercise phase may include monitoring calorie burning and total energy expenditure after exercise. In some implementations, the metabolic equivalent (MET) equation or the Harris-Benedict equation may be used to determine calorie burning and total energy expenditure. Finally, recommendations from the Treatment Management Engine 114 for future exercise phases, dietary recommendations, liver health status, and estimated fat oxidation and weight loss are discussed in detail below.

[0266] Additionally, the treatment management engine 114 can provide user-specific dietary recommendations based on the user's lactate response to HIIT. For example, dietary recommendations may include specific macronutrients to consume or suggested meal times relative to the exercise phase. For instance, after an exercise phase, the user may be recommended to consume protein over a specified time period (e.g., within 20 minutes of exercise). In this example, the user would be instructed to consume protein to build muscle mass. In another example, the user may be instructed not to consume lactate, glucose, or fructose, or to limit their intake of lactate, glucose, or fructose over a specified time period (e.g., 2 hours), in order to maintain higher levels of lipid oxidation.

[0267] In addition, the treatment management engine 114 can provide feedback related to the user's liver health. Liver health feedback may include providing the user with information about the diagnosis of liver disease and / or the improvement or worsening of liver disease. Information about the improvement or worsening of liver disease may be obtained based on monitoring the progression of liver disease using pre- or post-exercise lactate measurements (e.g., fasting lactate, overnight baseline lactate, post-exercise lactate clearance rate, lactate levels during exercise), other analyte levels (e.g., liver lipid content, liver enzyme status, glycogen storage status, bilirubin levels, or insulin resistance), and liver health tests (e.g., determining liver fat via MRI-PDFF, determining liver inflammation via ALT / AST levels, determining fibrosis activity via Pro-C3 testing, determining clotting time via international normalized ratio testing, or detecting liver elasticity via MRE imaging).

[0268] Additionally, the treatment management engine 114 can predict fat oxidation at rest and after exercise and provide the user with information about the predicted fat oxidation. The treatment management engine 114 can consider user input (e.g., the user's diet choices, sleep, medical history, or medications taken) and past exercise data (e.g., exercise duration) to determine the metabolic response to exercise and fat oxidation. The metabolic response can be the user's ability to clear lactic acid and glucose from the body after exercise compared to past workouts. Fat oxidation can be determined based on the proportion of carbohydrate and lipid levels to total calorie expenditure (e.g., using indirect calorimetry). The lipid level portion of total calorie expenditure can indicate the user's fat oxidation due to exercise.

[0269] In some implementations, the treatment management engine 114 may associate lower lactate levels over time during subsequent exercise phases with greater fat oxidation. Because blood lactate accumulation can be negatively correlated with fat oxidation, lower lactate accumulation resulting from exercise and / or an improved lactate clearance rate will lead to greater expected fat oxidation. For example, a user may experience maximum fat oxidation between 1.8 mM and 2.1 mM lactate. Therefore, if the user's baseline lactate is above 2.1 mM, a decrease in lactate levels during this period can be observed, and the intensity level should be increased to bring the lactate level back to approximately 2.1 mM to achieve maximum fat oxidation. Additionally, a lower baseline lactate level after an exercise phase can indicate improved fat oxidation at rest. For example, if a user has a baseline lactate level of 2 mM at the start of the first exercise phase, and the user's baseline lactate level increases to 1.5 mM due to dietary and / or exercise guidance, the treatment management engine 114 can determine that the user has achieved improved fat oxidation.

[0270] The Treatment Management Engine 114 can also measure a user's metabolic rate after exercise and provide the user with information on their metabolic rate or the improvement in their metabolic rate relative to a previous metabolic rate. The Treatment Management Engine 114 can use peak lactate, lactate clearance rate, and lactate change rate to predict post-exercise metabolic rate. The Treatment Management Engine 114 can also use indirect calorimetry, the Weir equation, or analyte measurements (e.g., glucose, free fatty acids, β-hydroxybutyrate, and glycerol) to measure a user's metabolic rate across a series of exercise phases to determine improvements in metabolic rate. Improvements in metabolic rate can reflect improved liver health, improved overall health, and weight loss over time. If the exercise program is continued over a period of time, the Treatment Management Engine 114 may be able to determine future weight loss. For example, the Treatment Management Engine 114 may be able to predict expected weight loss after one month of continuous exercise.

[0271] The treatment management engine 114 can also be configured to interface with a CPAP machine that can be used as an indirect calorimeter. If a user uses a CPAP machine daily, utilizing the CPAP machine as an indirect calorimeter may be beneficial for collecting consistent indirect calorimetric readings. For example, the CPAP can be configured as an indirect calorimeter to measure the metabolic use of a specific fuel (e.g., fat or carbohydrate) via a carbon dioxide (CO2) sensor and a flow sensor to determine the rate of CO2 production. In a CPAP machine, a CO2 sensor and a flow sensor can be added to the CPAP mask or hose to measure oxygen ingested from inhaled air and CO2 produced from exhaled air during sleep or rest. Using a CPAP machine to measure nighttime energy expenditure can be used to compare with the user's metabolic rate and energy expenditure after exercise, providing feedback to the user.

[0272] Additionally, at box 538, feedback related to metabolic fitness can be provided to the user. Metabolic fitness can be determined by the user's ability to reach a similar HIIT intensity range in the future. For example, if the user required a specific speed or power to reach a HIIT intensity range in past workout phases, the treatment management engine 114 can monitor whether that intensity is effectively used to reach the HIIT intensity range in the future. Over time, if the user requires greater intensity (e.g., speed or power) to reach a HIIT intensity range, the user may have improved metabolic fitness because the user's liver, kidneys, and skeletal muscles become more efficient at clearing lactic acid from the body at greater intensity.

[0273] In some embodiments, the feedback provided to the user at box 538 may include rewards and / or gamification features. As the user progresses closer to their goal (e.g., a certain number of workout phases, some improvement in metabolic fitness, etc.), the user may receive various badges, stars, and / or other rewards. In some embodiments, the feedback provided to the user at display device 107 may be configured, for example, to connect to the user's social media accounts, whereby the user can share updates and rewards with followers and friends via their social media accounts. Allowing the user to share with others via social media can provide support and encouragement to the user and encourage them to continue their workout program.

[0274] The therapy management engine 114 can also be configured to receive input from a user. For example, the user may be prompted to provide input to the therapy management engine 114 via the user interface of display device 107, the user interface 113 of exercise machine 108, etc. The therapy management engine 114 may request the user's perceived activity level after a HIIT phase or an interval during a HIIT phase. In some embodiments, for example, the user may be given the option to select their perceived activity level during the exercise phase based on a Borg scale of perceived activity (e.g., "I feel I can continue at this pace," "I have reached my limit," or "I feel nauseous and cannot continue at this intensity"). Receiving input from the user regarding their perceived activity level can assist the therapy management engine 114 in suggesting an increase or decrease in intensity during future exercise phases. For example, if the user provides input that they feel nauseous and cannot continue after reaching a HIIT intensity range early in the exercise phase and continuing for a few more minutes, the therapy management engine 114 may suggest that the user exercise at a lower intensity in the future. A lower intensity will still allow the user to reach a HIIT intensity range and provide an effective exercise, without the user feeling that they cannot continue at that intensity.

[0275] However, in some cases, users taking beta-blockers may have a higher perceived activity level than they actually are. For example, a user taking beta-blockers might perceive themselves as overactive and provide input on this situation, but the user's lactate measurement could signal the treatment management engine 114 that the user has achieved the ideal HIIT intensity range during the perceived overactivity. In this example, the treatment management engine 114 is unlikely to adjust the intensity of future workout phases based on the user's perceived activity level.

[0276] In some cases, potassium levels can be used to determine when a user has experienced excessive activity. For example, a user may have experienced excessive activity if their potassium level reaches an absolute maximum, an individual maximum, or exceeds a set duration above the maximum potassium level. In another example, a user may also have experienced excessive activity if their potassium level shows a high rate of change (which could indicate that the user's potassium level will eventually exceed the absolute or individual maximum potassium level).

[0277] At box 540, once the user has completed a HIIT phase, the treatment management engine 114 can utilize information about the user acquired during the training phase (e.g., analyte data, non-analyte data, user input, etc.) to optimize future training phases. For example, this information can be used in one or more rule-based models at box 526 during future training phases. In another example, this information can be used to personalize the ML models utilized at boxes 526 and 528 (e.g., training / retraining these ML models). Therefore, for future training phases, the treatment management engine 114 may be able to more effectively predict the optimal training parameters that will enable a particular user to reach a HIIT intensity range.

[0278] exist Figure 5C If the model determines that the user has already selected a resistance training exercise type, or if the treatment management engine 114 has already suggested resistance training based on the user's goals, medical history, time commitment, etc., at box 522, then the treatment management engine 114 may begin at box 542. At box 542, the treatment management engine 114 can determine whether the user has completed a resistance training phase with available data in the past by referring to the user's profile 118. For example, the treatment management engine 114 can examine the exercise data in the user's profile 118 to determine whether there is data related to one or more of the user's past resistance training phases.

[0279] At box 544, if the user has previously completed a resistance training phase for which data is available, the treatment management engine 114 determines personalized training guidance for the user based on the user's own historical exercise data. For example, exercise data from user profile 118 can be used to determine the set of exercise parameters for the user's training phase. In some embodiments, the set of exercise parameters for resistance training may include the type of equipment to be used (e.g., free weights, barbell weights, resistance bands, etc.), the type of exercise to be performed (e.g., squats, bicep curls, etc.), number of repetitions, number of sets, weight, time between repetitions and / or sets, cadence of repetitions, speed of repetition execution, target lactate level, etc. Providing personalized guidance based on the user's own historical exercise data can be done in several ways. For example, a rule-based model can be used in conjunction with the user's own historical exercise data to provide the set of exercise parameters for the user's training phase. Alternatively or additionally, one or more ML models trained based on the user's own historical data can be used to provide the set of exercise parameters for the user's training phase. For example, training server system 140 can retrieve user-specific exercise data from user profile 118 to train one or more ML models.

[0280] At box 546, if the user has not previously completed a resistance training phase for which data is available, the treatment management engine 114 may use a non-personalized, population-based model to determine the training parameters for the user, at least until the user has performed one or more resistance training phases and personal resistance training data is available to the user. For example, the treatment management engine 114 may use a rule-based model that defines resistance training parameters determined based on empirical studies involving population data. For instance, resistance training parameters may be determined using rules based on parameters that are effective for user groups similar to the user in achieving a range of resistance training intensity. To identify user groups similar to the user, the population-based model may consider demographic variables (e.g., age, gender, etc.) and physiological variables (e.g., weight, height, BMI, body composition, etc.) to determine parameters for achieving a range of resistance training intensity.

[0281] Alternatively or concurrently, the treatment management engine 114 may use one or more ML models to provide users with exercise parameters to achieve a resistance training intensity range (i.e., lactate levels in the resistance training zone). For example, the ML model may be trained based on population-based training data associated with healthy users who have achieved the resistance training intensity range. The dataset includes data records, each containing exercise parameters, analyte data, non-analyte data, and / or other relevant information from the corresponding user's profile 118 for each exercise phase. As an example, each data record in the training dataset may include timestamped analyte and non-analyte data for the user's exercise phase. Data records may be labeled with one or more exercise parameters. Using such a training dataset, a model can be trained to predict one or more exercise parameters to help the user achieve the resistance training intensity range.

[0282] Before using a personalized or population-based model to guide a user through training phases based on the training parameters determined at boxes 544 and 546, the treatment management engine 114 may prompt the user via the user interface of the display device 107 to provide input, for example, related to any injury the user has currently suffered or recently suffered. The user may also be prompted to provide the treatment management engine 114 with a risk tolerance level (e.g., the amount of risk the user is willing to accept of recurrence or worsening of the injury due to the resistance training phase). Based on the user-reported injury and risk tolerance level, the treatment management engine 114 may provide training parameters to prevent further injury or the risk of further injury to the user based on the user's risk tolerance. In some embodiments, the treatment management engine 114 may determine ideal training parameters for the user based on historical user group data of users who have suffered the same or similar injuries. Training parameters from historical users with similar reported injuries can be used to determine the predicted injury risk for a specific training parameter (e.g., a specific exercise, muscle group, or body weight), and this risk value can be compared with the risk tolerance level provided by the user to provide the user with appropriate training parameters.

[0283] Before beginning the workout phase, the treatment management engine 114 may instruct the user to begin exercising at a warm-up intensity for a certain duration. In some embodiments, the intensity may be half of the intensity later provided to the user to reach the resistance training intensity range. In some embodiments, the duration of the warm-up intensity may be approximately 5 or 10 minutes.

[0284] At box 548, once the warm-up duration is complete, the treatment management engine 114 can instruct the user to increase the intensity to the recommended exercise parameters determined in boxes 544 or 546.

[0285] Typically, healthy users who have not recently consumed a diet high in lactate or lactate precursor metabolites may have an initial lactate level of approximately 1.5 mmol / L at the start of exercise. Then, unlike users with poor metabolism, the rate of change in lactate may be slightly negative or slightly positive, but close to zero, as the rate of lactate production is roughly the same as the rate of lactate clearance in the body. However, given the high-intensity nature of HIIT exercise, the lactate slope may not show an initial slightly negative or slightly positive rate of change, but rather begin to increase rapidly. To ensure that users reach and do not exceed the resistance training intensity range, the treatment management engine 114 is configured to continuously monitor the user's lactate indices and physiological parameters to determine the optimal exercise intensity and / or parameters for reaching the resistance training intensity range.

[0286] Furthermore, the treatment management engine 114 can determine rest periods between workout sessions within a resistance training phase based on data from non-analyte sensors and / or accelerometer data. For example, the correlation between heart rate and / or respiratory rate and lactate levels in past workout phases allows the treatment management engine 114 to determine when the user is in a rest period or to instruct the user to take rest periods to maintain desired lactate levels. Determining rest periods allows the treatment management engine 114 to predict when the user's lactate levels may become more stable and / or decrease over a short period. Additionally, determining rest periods allows the treatment management engine 114 to determine the user's real-time lactate levels for comparison with one or more target lactate levels. Further, based on the known duration and frequency of rest periods in workout phases, the treatment management engine 114 can recommend variations in the duration and frequency of rest periods in future workout phases to improve future workout efficiency.

[0287] For example, if a user's training parameters are not optimal (e.g., the user is not exercising with sufficient intensity based on sets, repetitions, or weights), the user will not reach the optimal resistance training intensity range. Alternatively, if the user exercises with parameters that are too intense, the user will reach the resistance training range more quickly, but may subsequently experience increased lactate levels beyond the optimal resistance training intensity range. Therefore, for the treatment management engine, it is crucial to continuously monitor the user's lactate levels and physiological parameters to instruct the user to increase or decrease the intensity of the parameters to reach the optimal resistance training intensity range, as discussed further with respect to boxes 550 and 552. As described herein, monitoring physiological parameters other than lactate levels can account for the potential lag in lactate levels when the user begins the resistance training phase. Furthermore, given resistance training typically involves training a single muscle group at a time (e.g., legs or arms), lactate levels may first increase locally in the trained muscles before being released into the bloodstream, causing a further lag in systemic lactate levels. Therefore, monitoring physiological parameters can assist the treatment management engine 114 in providing accurate real-time feedback to the user.

[0288] At box 550, the treatment management engine 114 can determine whether the user's lactate level reflects or is expected to approach the user's desired resistance training intensity range. For example, the treatment management engine 114 can monitor values ​​associated with the area under the lactate curve over time, the increase in lactate relative to the user's baseline lactate level, and / or the rate of lactate increase to determine whether the user has reached or is expected to reach the resistance training intensity range (also referred to herein as the defined range). Alternatively or additionally, the treatment management engine 114 can monitor physiological parameters to determine when the user's lactate level reflects or is expected to approach the user's desired resistance training intensity range. For healthy users, reaching, for example, 5 mmol–15 mmol lactate or 3 mmol–7 mmol lactate corresponding to the area under the curve or the expected area under the curve may indicate that the user has reached the desired resistance training intensity range.

[0289] To determine whether a user is approaching a physiological state corresponding to the resistance training intensity range, the treatment management engine 114 can use one of a variety of rule-based or ML models to obtain the user's lactate levels and other physiological parameters to predict whether the user is about to approach a physiological state corresponding to the desired resistance training intensity range.

[0290] At box 552, the treatment management engine 114 can observe lactate levels and, if the user is not inclined to achieve the optimal resistance training intensity range, can instruct the user to increase or decrease exercise intensity and / or adjust exercise parameters. For example, if the user does not exercise at sufficient intensity to reach the resistance training intensity range, or if the user increases the intensity too much, causing the user's lactate level to rise beyond the desired resistance training intensity range, it is considered that the user is not inclined to reach the desired lactate range.

[0291] Various models (e.g., rule-based, machine learning, or predictive algorithms) can be used to determine optimal exercise parameters, thus establishing an ideal intensity to ensure the user's lactate levels reach the desired resistance training intensity range without exceeding it. In some implementations, rule-based models can be utilized, where various rules can be defined around a set of parameters, including the user's current lactate level, current rate of lactate change, time since the start of exercise, exercise intensity, and other parameters. For example, an example rule might specify that if the user's current lactate level is X, the current rate of lactate change is Y, and the time since the start of exercise is Z, then the intensity should be Q. Q can be the set of exercise parameters that bring the user into the resistance training intensity range, such as exercise type, number of repetitions, number of sets, rest between sets, etc. Rules can become more granular and involve the entire user, including other physiological and demographic indicators. For example, a rule might state that if the user is 40 years old and has heart rate A, temperature B, current lactate level X, and current rate of lactate change Y, then the intensity should be Q.

[0292] In some other implementations, one or more prediction algorithms or models may be used to predict the optimal intensity to achieve a desired range of resistance training intensity. For example, one or more models, which may be the same as or different from the personalized or population-based models described with respect to boxes 544 and 546, may be run continuously to take a set of inputs (e.g., the user's current lactate level, current rate of lactate change, and other inputs received at box 402) and output the optimal intensity and / or corresponding training parameters for achieving the desired range of resistance training intensity.

[0293] In some implementations, as described with reference to HIIT and Zone 2, the treatment management engine 114 can determine multiple factors when instructing the user to increase or decrease the intensity. For example, the treatment management engine 114 can determine whether the user is exercising outdoors, including ambient temperature, air quality, humidity, wind speed, known allergens, and altitude.

[0294] After suggesting an increase or decrease in intensity at box 552, the treatment management engine 114 may return to box 548 to determine whether the user has reached or is approaching the expected resistance training intensity range. If not, the treatment management engine 114 returns to box 552 to adjust the intensity accordingly as described above. If yes, the treatment management engine 114 proceeds to box 554.

[0295] At box 554, the treatment management engine 114 can instruct the user to maintain the current intensity for a specified duration (e.g., 5-10 minutes). The specified duration can be the ideal length of a resistance training phase for the user's goals. The specified duration can also be based on the most effective length of training for the user based on past training data, or the most effective length for historical users with similar demographics to the user.

[0296] At box 556, once the duration is complete, for non-diabetic users, the treatment management engine 114 can guide the user to complete the cool-down at a lower intensity (e.g., walking) and monitor lactate levels (or physiological parameters) to ensure that the user's lactate levels have returned to baseline levels (e.g., 1.5 mmol). Active cool-down allows the body to clear excess lactate generated due to exercise and optimizes long-term energy expenditure after exercise.

[0297] For users with type 1 and type 2 diabetes, the Treatment Management Engine 114 can instruct them to perform a cool-down at a lower intensity (e.g., walking) and / or to ingest specific post-workout nutrients to prevent immediate or potential hyperglycemia. Similar to HIIT workouts, resistance training is an anaerobic exercise that can cause a high lactate spike during the workout and a post-workout glucose spike once the user has cleared excess lactate from their body. If, after a resistance training phase, the user's glucose is low but does not require immediate action (e.g., 70 mg / dL or 3.9 mmol / L), glucose intake or insulin administration will not be recommended, as the user will experience a glucose spike once their liver and skeletal muscles have cleared excess lactate during the active cool-down. Alternatively, if the user's glucose level is dangerously low (e.g., 54 mg / dL), they can be instructed to ingest small amounts of glucose and monitor post-workout glucose levels to avoid hyperglycemia. High-sugar meals or beverages (e.g., Gatorade) may not be recommended, as they can cause a surge in lactate levels, inhibiting the cool-down process, reducing fat oxidation, decreasing long-term energy expenditure, and diminishing overall workout effectiveness.

[0298] Further at box 556, the treatment management engine 114 can provide users with feedback on the effects of metabolic fitness and exercise phases, as well as recommendations for future exercise phases, dietary recommendations, liver health status, estimated fat oxidation, and estimated weight loss.

[0299] For example, feedback associated with metabolic fitness may include changes in a user's lactate response to exercise over time. For instance, when a user completes a workout phase, the treatment management engine 114 may provide feedback on improvements in lactate clearance rates over time, indicating improved metabolic fitness. Feedback on improved metabolic fitness may also be based on lactate clearance over time relative to whether the user completed a cool-down after the workout phase, as cool-downs promote lactate clearance. The treatment management engine 114 may determine the intensity and duration of a cool-down while monitoring the user's lactate clearance rate over time. In some embodiments, the treatment management engine 114 may instruct the user to ingest a specific amount of lactate when not exercising (e.g., through an oral lactate tolerance test) to determine the lactate clearance rate and improvements in the lactate clearance rate when there is no muscle consumption of lactate during exercise. Improvements in the lactate clearance rate in response to the same amount of lactate and in the absence of exercise will indicate improvements in the user's metabolic fitness.

[0300] In some implementations, feedback on the effectiveness of the exercise phase may include monitoring calorie burning and total energy expenditure after exercise. In some implementations, the metabolic equivalent (MET) equation or the Harris-Benedict equation may be used to determine calorie burning and total energy expenditure. Finally, recommendations from the Treatment Management Engine 114 for future exercise phases, dietary recommendations, liver health status, and estimated fat oxidation and weight loss are discussed in detail below.

[0301] Additionally, the treatment management engine 114 can provide user-specific dietary recommendations based on the user's lactate response to resistance training. For example, dietary recommendations may include specific macronutrients to consume or suggested meal times relative to the exercise phase. For instance, after an exercise phase, the user may be recommended to consume protein over a specified time period (e.g., within 20 minutes of exercise). In this example, the user would be instructed to consume protein to build muscle mass. In another example, the user may be instructed not to consume lactate, glucose, or fructose, or to restrict their intake of lactate, glucose, or fructose over a specified time period (e.g., 2 hours), in order to maintain higher levels of fat oxidation and greater long-term energy expenditure.

[0302] In addition, the treatment management engine 114 can provide feedback related to the user's liver health. Liver health feedback may include providing the user with information about the diagnosis of liver disease and / or the improvement or worsening of liver disease. Information about the improvement or worsening of liver disease may be obtained based on monitoring the progression of liver disease using pre- or post-exercise lactate measurements (e.g., fasting lactate, overnight baseline lactate, post-exercise lactate clearance rate, lactate levels during exercise), other analyte levels (e.g., liver lipid content, liver enzyme status, glycogen storage status, bilirubin levels, or insulin resistance), and liver health tests (e.g., determining liver fat via MRI-PDFF, determining liver inflammation via ALT / AST levels, determining fibrosis activity via Pro-C3 testing, determining clotting time via international normalized ratio testing, or detecting liver elasticity via MRE imaging).

[0303] Additionally, the treatment management engine 114 can predict fat oxidation at rest and after exercise and provide the user with information about the predicted fat oxidation. The treatment management engine 114 can consider user input (e.g., the user's diet choices, sleep, medical history, or medications taken) and past exercise data (e.g., exercise duration) to determine the metabolic response to exercise and fat oxidation. The metabolic response can be the user's ability to clear lactic acid and glucose from the body after exercise compared to past workouts. Fat oxidation can be determined based on the proportion of carbohydrate and lipid levels to total calorie expenditure (e.g., using indirect calorimetry). The lipid level portion of total calorie expenditure can indicate the user's fat oxidation due to exercise.

[0304] In some implementations, the treatment management engine 114 may associate lower lactate levels over time during subsequent exercise phases with greater fat oxidation. Because blood lactate accumulation can be negatively correlated with fat oxidation, lower lactate accumulation resulting from exercise and / or improved lactate clearance rates will lead to greater expected fat oxidation. For example, a user may experience a maximum fat oxidation between 1.8 mmol / L and 2.1 mmol / L. Therefore, if a user's baseline lactate level is higher than 2.1 mmol / L, a decrease in lactate levels to less than 2.1 mmol / L during exercise will result in increased fat oxidation. Additionally, lower baseline lactate levels after an exercise phase may indicate improved fat oxidation at rest. For example, if a user has a baseline lactate level of 2 mmol / L at the start of the first exercise phase, and their baseline lactate level increases to 1.5 mmol / L due to dietary and / or exercise guidance, the treatment management engine 114 may determine that the user has achieved improved fat oxidation.

[0305] The Treatment Management Engine 114 can also measure a user's metabolic rate after exercise and provide the user with information on their metabolic rate or the improvement in their metabolic rate relative to a previous metabolic rate. The Treatment Management Engine 114 can use peak lactate, lactate clearance rate, and lactate change rate to predict post-exercise metabolic rate. The Treatment Management Engine 114 can also use indirect calorimetry, the Weir equation, or analyte measurements (e.g., glucose, free fatty acids, β-hydroxybutyrate, and glycerol) to measure a user's metabolic rate across a series of exercise phases to determine improvements in metabolic rate. Improvements in metabolic rate can reflect improved liver health, improved overall health, and weight loss over time. If the exercise program is continued over a period of time, the Treatment Management Engine 114 may be able to determine future weight loss. For example, the Treatment Management Engine 114 may be able to predict expected weight loss after one month of continuous exercise.

[0306] The treatment management engine 114 can also be configured to interface with a CPAP machine that can be used as an indirect calorimeter. If a user uses a CPAP machine daily, utilizing the CPAP machine as an indirect calorimeter may be beneficial for collecting consistent indirect calorimetric readings. For example, the CPAP can be configured as an indirect calorimeter to measure the metabolic use of a specific fuel (e.g., fat or carbohydrate) via a carbon dioxide (CO2) sensor and a flow sensor to determine the rate of CO2 production. In a CPAP machine, a CO2 sensor and a flow sensor can be added to the CPAP mask or hose to measure oxygen ingested from inhaled air and CO2 produced from exhaled air during sleep or rest. Using a CPAP machine to measure nighttime energy expenditure can be used to compare with the user's metabolic rate and energy expenditure after exercise, providing feedback to the user.

[0307] Additionally, at box 556, feedback related to metabolic fitness can be provided to the user. Metabolic fitness can be determined by the user's ability to reach a resistance training intensity range at a similar intensity in the future. For example, if the user required a specific speed or power to reach a resistance training intensity range in past training phases, the treatment management engine 114 can monitor whether that intensity is effectively used to reach the resistance training intensity range in the future. Over time, if the user requires greater intensity (e.g., higher weight, more repetitions, more sets, less rest time, etc.) to reach the resistance training intensity range, the user may have improved metabolic fitness because the user's liver and skeletal muscles become more efficient at clearing lactic acid from the body at greater intensity.

[0308] In some embodiments, the feedback provided to the user at box 556 may include rewards and / or gamification features. As the user progresses closer to their goal (e.g., a certain number of workout phases, some improvement in metabolic fitness, etc.), the user may receive various badges, stars, and / or other rewards. In some embodiments, the feedback provided to the user at display device 107 may be configured, for example, to connect to the user's social media accounts, whereby the user can share updates and rewards with followers and friends via their social media accounts. Allowing the user to share with others via social media can provide support and encouragement to the user and encourage them to continue their workout program.

[0309] The therapy management engine 114 can also be configured to receive input from a user. For example, the user may be prompted to provide input to the therapy management engine 114 via a user interface such as the display device 107. The therapy management engine 114 may request the user's perceived activity level after a resistance training phase or after a specific set or workout during a resistance training phase. In some embodiments, the user may be given the option to select their perceived activity level during the workout phase (e.g., "I can't complete the recommended number of repetitions" or "I'm almost at my limit"). Receiving input from the user regarding their perceived activity level can assist the therapy management engine 114 in suggesting increases or decreases in workout parameters and / or intensity for future workout phases. For example, if the user provides input that they feel nauseous and cannot continue after reaching the resistance training intensity range early in the workout phase and continuing for a few more sets, the therapy management engine 114 may suggest that the user complete fewer sets or fewer repetitions in the future. A lower number of repetitions or sets will still allow the user to reach the resistance training intensity range and provide an effective workout phase without the user feeling unable to continue or complete the workout phase.

[0310] However, in some cases, users taking beta-blockers may have a perceived activity level higher than it actually is. For example, a user taking beta-blockers might perceive themselves as overexerting and provide input on this situation, but the user's lactate measurement could signal the treatment management engine 114 that the user has achieved the ideal resistance training intensity range during the perceived overexertion. In this example, the treatment management engine 114 is unlikely to adjust the intensity in future workout phases based on the user's perceived activity level.

[0311] In some cases, potassium levels can be used to determine when a user has experienced excessive activity. For example, a user may have experienced excessive activity if their potassium level reaches an absolute maximum, an individual maximum, or exceeds a set duration above the maximum potassium level. In another example, a user may also have experienced excessive activity if their potassium level shows a high rate of change (which could indicate that the user's potassium level will eventually exceed the absolute or individual maximum potassium level).

[0312] At box 558, once the user has completed the resistance training phase, the treatment management engine 114 can utilize information about the user acquired during the training phase (e.g., analyte data, non-analyte data, user input, etc.) to optimize future training phases. For example, this information can be used in one or more rule-based models at box 544 during future training phases. In another example, this information can be used to personalize the ML models utilized at boxes 544 and 546 (e.g., training / retraining these ML models). Therefore, for future training phases, the treatment management engine 114 may be able to more effectively predict the optimal training parameters that will allow a particular user to reach the range of resistance training intensity.

[0313] Figures 6A to 6C Example methods for optimizing training phases and providing athletes with feedback on fitness, post-workout nutrition or glucose administration, the effectiveness of training phases, etc., are described. As an example, after a user is categorized as an athlete, the treatment management engine 114 can be configured to guide the user through a series of training phases to improve fitness or achieve the user's goals. For example, the user can use app 106 or 111 to indicate that their goal is to improve peak performance, train for a specific competition or event, or increase their lactate threshold. The athlete can request guidance for a specific training phase (e.g., the first training phase in a series) to achieve their goals. In this example, app 106 or 111 can then present a user interface for the user to select whether they want to perform Zone 2 training, resistance training, or HIIT for their upcoming training phase.

[0314] Method 600 begins at box 602, where the treatment management engine 114 determines whether the user has already selected HIIT, resistance training, or a two-zone workout type. If the user has already selected two zones, the treatment management engine 114 proceeds directly to box 604. If the user has already selected HIIT, the treatment management engine 114 proceeds to the section about... Figure 6B The described method. If the user has already selected resistance training, the treatment management engine 114 will proceed to... Figure 5C The method described.

[0315] If the user has not yet selected or specified an exercise type, at box 622, the treatment management engine 114 may recommend, for example, Zone 2 to users whose goals are to improve endurance, metabolic fitness, peak performance, and fitness baseline, as well as muscle mitochondrial health and function, thereby improving race performance. In some embodiments, the treatment management engine 114 may recommend that the user complete 80% of their training in Zone 2 and 20% in the “race pace” zone, similar to the expected pace and intensity of an upcoming race. In other exemplary cases, the treatment management engine 114 may recommend HIIT to users whose goals are to improve strength and VO2 max, or when the user may have limited time to complete a training phase. In some embodiments, the treatment management engine 114 may recommend resistance training to users whose goals are to improve the strength of one or more muscle groups, particularly one or more muscle groups associated with a specific movement or event (e.g., leg muscles used for cycling). As the user's goals and fitness levels change, the treatment management engine 114 may recommend different types of exercise to the user over time.

[0316] In other implementations, the treatment management engine 114 may suggest a recovery day (e.g., not completing any training phase) based on the user's lactate level at the start of exercise or based on user input. For example, if the user's lactate level at the start of exercise is typically 1.0 mmol, but the user's current lactate level at the start of exercise is 2.0 mmol, and / or the user's input indicates that the user completed an intense training phase the previous day (e.g., based on the user's perception), then the treatment management engine 114 may suggest that the user not exercise. Particularly in athletes, high lactate levels may be due to overuse and / or overtraining. High lactate levels at the start of exercise may indicate the need for recovery; therefore, the treatment management engine 114 may suggest a recovery day.

[0317] In some implementations, the treatment management engine 114 may provide a workout readiness score to the user based on the user's lactate level at the start of the workout compared to the average lactate level at the start of the workout, information about the user's workout phases over previous days, information about the user's perceived activity during previous workout phases, and so on. For example, the treatment management engine 114 may provide a low workout readiness score if the user has completed a series of high-intensity workouts, reported high perceived activity, and / or had a higher-than-average lactate level at the start of the workout. Alternatively, the treatment management engine 114 may provide a high workout readiness score if the user has not completed a workout phase over several days and the user's lactate level at or below average at the start of the workout.

[0318] In some implementations, the treatment management engine 114 may recommend a primary Zone 2 training phase (e.g., 75%-80% of the training phase) in conjunction with some HIIT or resistance training training phases (or other training, such as weight training) (e.g., 20%-25% of the training phase) to improve peak performance and (e.g., by improving mitochondrial health) improve metabolic fitness.

[0319] In some implementations, if a user is training for a specific performance goal (e.g., running 13 miles at a 7-minute mile pace), the treatment management engine 114 may recommend a "race pace" training phase that simulates HIIT intensity but lasts for a longer period than HIIT intervals. To achieve this performance goal, the athlete user can be instructed to maintain a specific intensity level corresponding to their goal. In this example, lactate measurement can be used to inform the user of the maximum amount of time they can maintain a specific intensity level. For example, the user's goal could be to run 13 miles at a 7-minute mile pace while maintaining a lactate level of 4 millimoles.

[0320] In some implementations, the treatment management engine 114 may recommend HIIT, resistance training, or Zone 2 workouts, but the user may choose to reject the recommendations. For example, there might be a workout phase where the treatment management engine 114 recommends Zone 2 for fat burning based on the user's goals, but the user wants to focus on improving strength during that workout phase (e.g., through HIIT). In this example, the user may choose to ignore the recommendations and choose the desired type of workout based on short-term goals. However, if the user selects Zone 2 at box 602, or if the treatment management engine 114 suggests Zone 2 to the user at box 622, the treatment management engine 114 proceeds to box 604.

[0321] At box 604, the treatment management engine 114 can determine whether the user has completed a data-available Zone 2 training phase in the past by referring to the user's profile 118. For example, the treatment management engine 114 can examine the training data in the user profile 118 to determine whether there is data related to one or more of the user's past Zone 2 training phases.

[0322] At box 606, if the user has previously completed a Zone 2 exercise phase for which data is available, the treatment management engine 114 determines personalized exercise guidance for the user based on the user's own historical exercise data. For example, exercise data from the user profile 118 can be used to determine the set of exercise parameters for the user's exercise phase. The parameter set may include (e.g., exercise intensity, exercise duration, etc., at the start of the phase and throughout the phase). Exercise intensity itself can be a function of speed, resistance, inclination, etc. Providing personalized guidance based on the user's own historical exercise data can be done in several ways. For example, a rule-based model can be used in conjunction with the user's own historical exercise data to provide the set of exercise parameters for the user's exercise phase. Alternatively or additionally, one or more ML models trained based on the user's own historical data can be used to provide the set of exercise parameters for the user's exercise phase. For example, the training server system 140 can retrieve user-specific exercise data from the user profile 118 to train one or more ML models.

[0323] At box 608, if the user has not previously completed a Zone 2 training phase for which data is available, the treatment management engine 114 may use a population-based (non-personalized) model to determine the training parameters for the user, at least until the user has performed one or more training phases and individual training data is available to the user. For example, the treatment management engine 114 may use a rule-based model that defines Zone 2 training parameters based on empirical studies involving population data. For instance, Zone 2 training parameters for achieving Zone 2 lactate levels may be determined using rules based on parameters effective for a similar group of athletes (e.g., based on demographic and / or physiological variables such as age, sex, weight, height, BMI, etc.) similar to the user.

[0324] Alternatively or concurrently, the treatment management engine 114 may use one or more ML models to provide users with exercise parameters to achieve Zone 2 lactate levels. For example, the ML model may be trained based on population-based training data associated with athletes achieving Zone 2 lactate levels. The dataset includes data records, each containing exercise parameters, analyte data, non-analyte data, and / or other relevant information from the corresponding user's profile 118 for each exercise phase. As an example, each data record in the training dataset may include timestamped exercise parameters and corresponding timestamped analyte and non-analyte data for the user's exercise phase. Data records may be labeled with one or more exercise parameters. Using such a training dataset, a model can be trained to predict one or more exercise parameters to help the user achieve Zone 2 lactate levels.

[0325] Before using a personalized or population-based model to guide the user through the exercise phases based on the exercise parameters determined at boxes 606 and 608, the treatment management engine 114 may instruct the user to begin exercising at a warm-up intensity for a certain duration. In some embodiments, the intensity may be half of the intensity later provided to the user to reach the Zone 2 lactate range. In some embodiments, the duration of the warm-up intensity may be approximately 5 minutes or 10 minutes.

[0326] At box 610, once the warm-up duration is complete, the treatment management engine 114 can instruct the user and / or the exercise machine to increase the intensity to the exercise parameters determined in box 606 or 608.

[0327] In embodiments where a user is completing a workout phase on an exercise machine (e.g., exercise machine 108), the treatment management engine 114 can automatically set or gradually increase the intensity (e.g., speed, incline, and / or resistance). The treatment management engine 114 can automatically set the exercise machine to the exercise parameters determined at boxes 606 and 608. In embodiments where the user is not using the exercise machine, the treatment management engine 114 can instruct the user to increase the intensity of their workout by gradually increasing various exercise parameters until an optimal Zone 2 lactate range is achieved. Increasing intensity can be achieved by changing various exercise parameters associated with the exercise machine (e.g., exercise machine 108) or by expending additional effort (in the form of speed or power) when the user is not using the exercise machine.

[0328] Typically, athletes may have an initial lactate level of approximately 1 mmol / L at the start of exercise. Similar to healthy users, the rate of change of an athlete's lactate may be slightly negative or slightly positive, but close to zero at approximately 2.0 mmol / L, thus representing the optimal Zone 2 lactate range. Therefore, the treatment management engine 114 is configured to continuously monitor the user's physiological parameters to determine the optimal exercise intensity that will allow the user to reach and maintain a lactate range with a rate of change of zero or close to zero. For example, if the user does not exercise at sufficient intensity, the user will not reach the optimal Zone 2 lactate range. Alternatively, if the user exercises at excessively high intensity, the user will rapidly exceed the Zone 2 lactate range and continue to experience lactate increases beyond the desired range. Therefore, for the treatment management engine 114, it is crucial to continuously monitor the user's lactate levels (including the rate of change of lactate) and physiological parameters (e.g., heart rate, respiratory rate, glucose levels, power, speed, strength, accelerometer data, etc.) to instruct the user to increase or decrease intensity to maintain within the optimal Zone 2 lactate range, as further described with respect to boxes 612 and 614.

[0329] At box 612, the treatment management engine 114 can determine, for example, based on the user's lactate levels, whether the user is already within or approaching a desired lactate level in Zone 2. If the user is already within or approaching a lactate level in Zone 2 (e.g., the rate of lactate change is close to zero), the treatment management engine 114 can instruct the user to maintain the current intensity.

[0330] Note that the expected Zone 2 lactate range (also referred to herein as the defined range) may be, for example, between 1 mmol and 2 mmol. In another example, the expected Zone 2 lactate range may be a range relative to the user's lactate level at the start of exercise. For example, the Zone 2 lactate range may be 0.5 mmol to 1.5 mmol higher than the user's lactate level at the start of exercise. However, the range may differ for different users; therefore, the treatment management engine 114 is configured to observe lactate levels and determine whether the user is approaching or nearing a zero rate of lactate change, indicating that the user is in or will be in the Zone 2 lactate range. To determine whether the user is approaching or nearing a zero rate of lactate change, the treatment management engine 114 may use one of a variety of rule-based or ML models to obtain the user's lactate level and / or other lactate indicators (e.g., rate of change) as well as other physiological parameters to predict whether the user will soon approach or near a zero rate of change.

[0331] In some implementations, the treatment management engine 114 can determine whether a user has reached Zone 2 based on potassium levels. For example, based on the user's historical data, the treatment management engine 114 can determine an individualized potassium level range corresponding to the user's Zone 2 steady-state lactate range. During future training phases, the treatment management engine 114 can determine whether the user is within the Zone 2 lactate range based on whether the user's potassium levels are within the individualized Zone 2 potassium range.

[0332] In some implementations, as described above, determining whether a user has reached a certain metabolic state (e.g., Zone 2) is primarily accomplished by monitoring the user's lactate levels. However, in some cases, the user's lactate levels may be unavailable, and / or lactate measurements received from a continuous lactate sensor may be delayed relative to real-time lactate blood measurements. To address potential unavailability or lag, in some implementations, the treatment management engine 114 may additionally or alternatively use some of the user's physiological parameters (e.g., heart rate, respiratory rate, glucose levels, power, speed, strength, accelerometer data, etc.) as substitutes for lactate data. Such physiological parameters may be directly correlated with lactate levels and thus can be used to estimate real-time lactate levels during the exercise phase when such lactate levels are unavailable. Thus, during the first exercise phase where a continuous lactate sensor is available, lactate measurements can be obtained and correlated with the user's corresponding physiological parameters. These physiological parameters can then be used as substitutes for lactate levels indicating certain metabolic states (e.g., Zone 2, resistance training, HIIT) during future exercise phases where lactate information is unavailable.

[0333] At box 614, the treatment management engine 114 can observe lactate indicators (e.g., lactate levels and trends) and / or the user's physiological parameters, and if the user does not tend to achieve the optimal Zone 2 lactate range, the treatment management engine 114 can instruct the user to adjust the exercise intensity. For example, if the user does not exercise at a sufficient intensity to reach the Zone 2 lactate range (e.g., 1.8 mmol–2 mmol), or if the user exercises at too high an intensity that would cause the user's lactate level to reach the Zone 2 lactate range but subsequently increase beyond the desired range, it is considered that the user does not tend to reach the desired lactate range.

[0334] Various models (e.g., rule-based, machine learning, or predictive algorithms) can be used to determine optimal exercise parameters, thus establishing an ideal intensity to ensure the user's lactate level reaches the optimal Zone 2 lactate range without exceeding it. In some implementations, rule-based models can be utilized, where various rules can be defined around a set of parameters, including the user's current lactate level, current rate of lactate change, time since exercise began, exercise intensity, and other physiological parameters (e.g., heart rate, respiratory rate, glucose levels, power, speed, strength, accelerometer data, etc.). For example, a sample rule might specify that if the user's current lactate level is X, the current rate of lactate change is Y, and the time since exercise began is Z, then the intensity should be Q. Q can be the set of exercise parameters that bring the user into the Zone 2 range, such as speed, resistance level, altitude, power, etc. Rules can be made more granular and involve the entire user with other physiological and demographic indicators. For example, a rule could state that if a user is 40 years old and has heart rate A, temperature B, current lactate level X, current rate of lactate change Y, and time since the start of exercise Z, then the intensity should be Q.

[0335] In some other implementations, one or more prediction algorithms or models may be used to predict the optimal intensity to achieve the desired Zone 2 lactate range. For example, one or more models, which may be the same as or different from the personalized or population-based models described with respect to boxes 606 and 608, may be run continuously to take input sets (e.g., the user's current lactate level, current rate of lactate change, time since the start of exercise, and other inputs received at box 402) and / or other physiological parameters (e.g., heart rate, respiratory rate, glucose indices, power, speed, strength, accelerometer data, etc.) and output the optimal intensity and / or corresponding exercise parameters for achieving the optimal Zone 2 lactate range.

[0336] When instructing a user to increase or decrease intensity, the therapy management engine 114 can check whether the user is exercising outdoors. For example, non-analyte sensor data from a temperature sensor can indicate that the user is exercising outdoors, or the user's selection of, for example, outdoor running or walking as the exercise type can be used by the therapy management engine 114 to determine that the user is exercising outdoors. When the user is exercising outdoors, the therapy management engine 114 can consider air temperature, air quality, known allergens, humidity, wind speed, solar energy index, and altitude when suggesting increasing or decreasing intensity.

[0337] For example, based on sensor data from a temperature sensor, the treatment management engine 114 can notify the user that the environment is too hot and that exercising at the usual intensity is unsafe (e.g., recommending an exercise intensity as a function of temperature to reach the Zone 2 lactate range, even if the recommended intensity is lower / higher than historical data), or that the user should stop exercising.

[0338] Furthermore, the treatment management engine 114 can instruct the user to reduce intensity or stop exercising based on the correlation between potassium levels and temperature sensor data at higher outdoor temperatures. For example, even if the user perceives themselves to be at the correct intensity in high temperatures, potassium levels may increase beyond expected levels due to the heat. If the user perceives low to moderate activity but the potassium sensor indicates increased potassium levels, the user may be at risk of cardiac events (e.g., arrhythmias, cardiac arrest, etc.). Therefore, when the user's potassium levels increase beyond expected levels, the treatment management engine 114 can instruct the user to reduce intensity or stop exercising in high temperatures.

[0339] Furthermore, the treatment management engine 114 can use the correlation between lactate levels at a specific intensity level and temperature sensor data to provide the user with more accurate intensity instructions. For example, at the same intensity, lactate levels may be higher at higher temperatures. Therefore, at high temperatures, the treatment management engine 114 may instruct the user to exercise at a lower intensity compared to the intensity required under typical temperature conditions to achieve the Zone 2 lactate range. Conversely, at lower temperatures, lactate response may be lower, requiring increased intensity to reach the optimal Zone 2 lactate range. Without considering temperature sensor data, at high temperatures, the treatment management engine 114 might instruct the user to increase intensity, which could cause the user to increase lactate beyond the optimal Zone 2 lactate range, leading to heatstroke or other safety issues. At cooler temperatures, the treatment management engine 114 may not consider the higher intensity required to reach the Zone 2 lactate range when compared to typical temperatures that result in less efficient workout phases.

[0340] Furthermore, humidity levels captured from wearable sensors or from web / cloud-based measurements of local weather services can be used to further calculate the expected heat transfer coefficient corresponding to exercise intensity relative to temperature, and the expected corresponding lactate level for a given intensity. The treatment management engine 114 can correct for factors such as temperature, atmospheric pressure, local wind speed and / or humidity, and solar energy intensity or its absence during nighttime. Similar to the humidity data, other environmental factors can be captured from locally worn sensors or from expected parameters captured from cloud-based data on local weather at the exercise user's location, which can be manually entered into the system, automatically calculated based on sensor input or time measurements, or identified using GPS data from the user's location on one or more devices.

[0341] In addition to temperature and humidity, the therapy management engine 114 can also consider altitude when providing intensity instructions when the user is exercising outdoors. For example, if the user is jogging outdoors and approaching an upcoming hill, the therapy management engine 114 can instruct the user to maintain or even reduce the intensity. Failure to do so could result in the user exercising at an excessively high intensity when the hill factor is added. This intensity could cause the user's lactic acid levels to exceed the optimal Zone 2 lactic acid range. Therefore, the therapy management engine 114 can consider changes in altitude or anticipated changes in altitude based on GPS and / or map data and adjust the intensity recommendations accordingly.

[0342] Furthermore, when a user is exercising outdoors, the treatment management engine 114 can also take altitude into account when providing intensity instructions. For example, lactate levels increase when a user is at high altitudes, making it potentially more difficult to maintain a Zone 2 lactate range. However, as the user adapts to the altitude, lactate levels may decrease. In addition to providing more accurate exercise intensity based on altitude (e.g., suggesting lower intensity for achieving a Zone 2 lactate range at higher altitudes), the treatment management engine 114 can also determine when a user is likely to experience altitude sickness based on lactate levels.

[0343] Furthermore, when users are exercising outdoors, they can complete exercise phases (e.g., walking, running, cycling, etc.) on a specific route once or multiple times. In this case, the treatment management engine 114 can identify the route based on GPS data and past exercise data, and provide the user with information on where they should increase or decrease the intensity to maintain certain areas of lactate levels in Zone 2.

[0344] Furthermore, regardless of whether the user is exercising indoors or outdoors, the treatment management engine 114 can determine whether the user is exhibiting signs of cardiac stress based on cardiac indicators (e.g., via a heart rate monitor or ECG). For example, the treatment management engine 114 can monitor various cardiac indicators to detect signs of atrial flutter, tachycardia, atrial fibrillation, increased heart rate, etc. If the user experiences abnormal cardiac indicators and high lactate levels due to exercise, the treatment management engine 114 can instruct the user to reduce the intensity and / or stop exercising and seek medical care for potential cardiac complications.

[0345] After suggesting an increase or decrease in intensity at box 614, the treatment management engine 114 may return to box 612 to determine whether the user is within the Zone 2 lactate range, or whether the user's lactate level and rate of change indicate that the user tends to reach lactate levels within the Zone 2 lactate range. If not, the treatment management engine 114 returns to box 614 to adjust the intensity accordingly. If yes, the treatment management engine 114 proceeds to box 616.

[0346] At box 616, the treatment management engine 114 may instruct the user to maintain the current intensity for a specified duration (e.g., 5 minutes or 10 minutes). The specified duration may be the ideal length of a Zone 2 workout for the user's goals. The specified duration may also be based on the most effective length of workouts for the user based on past workout data or the most effective length for historical users with similar demographics. In some implementations, the athlete may be instructed to maintain Zone 2 for an extended period (e.g., 3-4 hours), which may deplete the user's carbohydrate stores. If the athlete is expected to deplete their carbohydrate stores during the workout phase, the treatment management engine 114 may instruct the user to ingest some nutrients (e.g., carbohydrates) to maintain Zone 2.

[0347] At box 618, once the duration is complete, for non-diabetic users, the treatment management engine 114 can guide the user to complete the cool-down at a lower intensity (e.g., walking) and monitor lactate levels (or physiological parameters) to ensure that the user's lactate levels have returned to baseline levels (e.g., 1 mmol). Active cool-down allows the body to clear excess lactate produced due to exercise and optimizes post-workout fat burning.

[0348] For users with type 1 diabetes, the treatment management engine 114 may recommend that the user complete a cool-down at a lower intensity (e.g., walking) and / or ingest specific post-workout nutrition and / or administer glucose to prevent present or potential hypoglycemia. In some embodiments, the treatment management engine 114 may determine whether the user is taking insulin to manage their diabetes and may not instruct users not taking insulin to administer glucose.

[0349] For users with type 2 diabetes, in addition to recommending a cool-down, the treatment management engine 114 may recommend avoiding insulin administration. Specifically, aerobic exercise (such as Zone 2) lowers blood glucose levels. If a user experiences hypoglycemia after the exercise phase, the treatment management engine 114 may determine whether the user's glucose level is dangerously low or expected to become dangerously low (e.g., less than 55 mg / dL). If the treatment management engine 114 determines that the user's glucose level is dangerously low or expected to become dangerously low, it may advise the user to consume post-workout nutrition (which may include protein or electrolyte drinks) while avoiding glucose intake. However, if glucose levels are low after the exercise phase, the user may be instructed to consume a small amount of food to increase their glucose levels and to continue monitoring glucose levels after the exercise phase. If the user is administering insulin, they may also be instructed to consume glucose after the Zone 2 exercise phase to avoid hypoglycemia. High-sugar meals or drinks (e.g., Gatorade) may not be recommended, as they can also cause a surge in lactate levels, which inhibits the cool-down process, reduces fat oxidation, and decreases overall exercise effectiveness.

[0350] Users with type 1 or type 2 diabetes who are not taking insulin can be instructed to perform a cool-down at 60%-80% of the intensity required to reach their lactate threshold. The treatment management engine 114 can instruct the user to perform a cool-down at 60%-80% of the intensity required to reach their lactate threshold based on analyte (e.g., lactate) and / or non-analyte data. In some embodiments, the treatment management engine 114 can optimize the user's cool-down duration based on analyte data (including glucose and / or lactate data) from the user's historical exercise phases to avoid hyperglycemic spikes. For example, the treatment management engine 114 can identify a user who experiences hyperglycemia after a 10-minute cool-down, where the user's glucose level is 100 mg / dL at the end of the exercise phase. During a subsequent exercise phase, if the user's glucose level is 100 mg / dL after the exercise phase, the treatment management engine 114 can recommend that the user complete a 20-minute cool-down.

[0351] Further at box 618, the treatment management engine 114 can provide users with feedback on the effectiveness of fitness and exercise phases, as well as recommendations for future exercise phases, dietary recommendations, liver health status, and estimated fat oxidation and weight loss.

[0352] For example, at box 618, feedback related to metabolic fitness (or its improvement) can be provided to the user. Metabolic fitness can be determined by the user's ability to reach Zone 2 lactate range at a similar intensity in future workout sessions. For example, if the user required a specific speed or power to reach Zone 2 lactate range in past workout sessions, the treatment management engine 114 can monitor whether that intensity is effective in reaching Zone 2 lactate range in future workout sessions. Over time, if the user requires greater intensity (e.g., speed or power) to reach Zone 2 lactate range, the user may have improved metabolic fitness because the user's liver and skeletal muscles become more efficient at clearing lactate from the body at greater intensity.

[0353] In some embodiments, the feedback provided to the user at box 618 may include rewards and / or gamification features. As the user progresses closer to their goal (e.g., a certain number of workout phases, some improvement in metabolic fitness, etc.), the user may receive various badges, stars, and / or other rewards. In some embodiments, the feedback provided to the user at display device 107 may be configured, for example, to connect to the user's social media accounts, whereby the user can share updates and rewards with followers and friends via their social media accounts. Allowing the user to share with others via social media can provide support and encouragement to the user and encourage them to continue their workout program.

[0354] Additionally, feedback related to metabolic fitness can be based on changes in a user's lactate response to exercise over time. For example, when a user completes a workout phase, the treatment management engine 114 can provide feedback on improvements in lactate clearance rates (e.g., during cool-down) over time, indicating improved metabolic fitness. Feedback on improved metabolic fitness can also be based on lactate clearance over time relative to whether the user completed a cool-down after the workout phase, as cool-down promotes lactate clearance. The treatment management engine 114 can determine the intensity and duration of a cool-down while monitoring the user's lactate clearance rate over time. In some embodiments, the treatment management engine 114 can instruct the user to ingest a specific amount of lactate when not exercising (e.g., an oral lactate tolerance test) to determine the lactate clearance rate and improvements in the lactate clearance rate when there is no muscle consumption of lactate during exercise. Improvements in the lactate clearance rate in response to the same amount of lactate and in the absence of exercise will indicate improvements in the user's metabolic fitness.

[0355] In some implementations, feedback on the effectiveness of the exercise phase may include monitoring calorie burning and total energy expenditure after exercise. In some implementations, the metabolic equivalent (MET) equation or the Harris-Benedict equation may be used to determine calorie burning and total energy expenditure.

[0356] Additionally, the treatment management engine 114 can provide personalized dietary recommendations based on the user's lactate response to Zone 2 exercise. For example, dietary recommendations may include specific macronutrients to consume or meal times relative to the exercise phase. For instance, after an exercise phase, the user may be recommended to consume protein over a specific time period (e.g., within 20 minutes of exercise) to build muscle mass. In another example, the user may be instructed not to consume lactate, glucose, or fructose, or to limit their intake of lactate, glucose, or fructose for a specified time period (e.g., 2 hours), in order to maintain higher levels of lipid oxidation.

[0357] In addition, the treatment management engine 114 can provide feedback related to the user's liver health. Liver health feedback may include providing the user with information about the diagnosis and / or improvement / worsening of liver disease. Information about the improvement / worsening of liver disease may be obtained based on monitoring liver disease progression using pre- or post-exercise lactate measurements (e.g., fasting lactate, overnight baseline lactate, post-exercise lactate clearance rate, lactate levels during exercise), other analyte levels (e.g., liver lipid content, liver enzyme status, glycogen storage status, bilirubin levels, or insulin resistance), and liver health tests (e.g., determining liver fat via MRI-PDFF, determining liver inflammation via ALT / AST levels, determining fibrosis activity via Pro-C3 testing, determining clotting time via international normalized ratio testing, or detecting liver elasticity via MRE imaging).

[0358] Additionally, the treatment management engine 114 can predict fat oxidation at rest and after exercise and provide the user with information about the predicted fat oxidation. The treatment management engine 114 can consider user input (e.g., the user's diet choices, sleep, medical history, or medications taken) and past exercise data (e.g., exercise duration) to determine the metabolic response to exercise and fat oxidation. The metabolic response can be the user's ability to clear lactic acid and glucose from the body after exercise compared to past workouts. Fat oxidation can be determined based on the proportion of carbohydrate and lipid levels to total calorie expenditure (e.g., using indirect calorimetry). The lipid level portion of total calorie expenditure can indicate the user's fat oxidation due to exercise.

[0359] In some implementations, the treatment management engine 114 may associate lower lactate levels during a later exercise phase with greater fat oxidation compared to a previous exercise phase. Because blood lactate accumulation can be negatively correlated with fat oxidation, lower lactate accumulation resulting from exercise and / or an improved lactate clearance rate will lead to greater expected fat oxidation. For example, a user may experience a maximum fat oxidation between 1.5 mmol / L and 2.1 mmol / L lactate. Therefore, if a user's baseline lactate level is higher than 2.1 mmol / L, a decrease in lactate levels to less than 2.1 mmol / L during exercise will result in increased fat oxidation. Additionally, a lower baseline lactate level after an exercise phase may indicate improved fat oxidation at rest. For example, if a user has a baseline lactate level of 1.5 mmol / L at the start of the first exercise phase, and their baseline lactate level increases to 1 mmol / L due to dietary and / or exercise guidance, the treatment management engine 114 may determine that the user has achieved improved fat oxidation.

[0360] The Treatment Management Engine 114 can also measure a user's metabolic rate after exercise and provide the user with information on their metabolic rate or improvements in metabolic rate. The Treatment Management Engine 114 can predict post-exercise metabolic rate using peak lactate, lactate clearance rate, and lactate change rate. The Treatment Management Engine 114 can also measure a user's metabolic rate across a range of exercise phases using indirect calorimetry, the Weir equation, or various analyte measurements (e.g., glucose, free fatty acids, β-hydroxybutyrate, and glycerol) to determine improvements in metabolic rate. Improvements in metabolic rate can reflect improved liver health, improved overall health, and weight loss over time.

[0361] The treatment management engine 114 can also be configured to interface with a continuous positive airway pressure (CPAP) machine that can be used as an indirect calorimeter. If a user uses a CPAP machine daily, utilizing the CPAP machine as an indirect calorimeter may be beneficial for collecting consistent indirect calorimetric readings. For example, the CPAP can be configured as an indirect calorimeter to measure the metabolic use of a specific fuel (e.g., fat or carbohydrate) via a carbon dioxide (CO2) sensor and a flow sensor to determine the rate of CO2 production. In a CPAP machine, a CO2 sensor and a flow sensor can be added to the CPAP mask or hose to measure oxygen intake from inhaled air and CO2 produced from exhaled air during sleep or rest. Using a CPAP machine to measure nighttime energy expenditure can be used for comparison with a user's metabolic rate and energy expenditure after exercise.

[0362] The therapy management engine 114 can also be configured to receive input from the user after a workout phase. For example, the user may be prompted to provide input to the therapy management engine 114 via the user interface of display device 107, the user interface 113 of exercise machine 108, etc. The therapy management engine 114 may request the user's perceived activity level after a Zone 2 workout phase. In some embodiments, for example, the user may be given the option to select their perceived activity level during the workout phase based on a Borg scale of perceived activity (e.g., "I feel I can continue at this pace," "I've reached my limit," or "I feel nauseous and cannot continue at this intensity"). Receiving input from the user regarding their perceived activity level can assist the therapy management engine 114 in suggesting an increase or decrease in intensity for future workout phases. For example, if the user provides input that they can continue at that pace after the workout, the therapy management engine 114 may suggest a higher intensity workout for a future workout phase if the user does not reach their optimal Zone 2 lactate range until 20 minutes into the workout. Higher intensity will allow the user to reach their Zone 2 lactate range more quickly, thereby maximizing the workout's effectiveness without causing overexertion.

[0363] However, in some cases, users taking beta-blockers may have a perceived activity level higher than it actually is. For example, a user taking beta-blockers might perceive themselves as overactive and provide input on this situation, but the user's lactate measurement could signal the treatment management engine 114 that the user was within the Zone 2 lactate range during the perceived overactivity. In this example, the treatment management engine 114 is unlikely to adjust the intensity of future workout phases based on the user's perceived activity level.

[0364] In some cases, potassium levels can be used to determine when a user has experienced hyperactivity during or after a workout. For example, a user may have experienced hyperactivity if their potassium level reaches an absolute maximum, an individualized maximum, or exceeds a set duration above the maximum potassium level. In another example, a user may also have experienced hyperactivity if their potassium level shows a high rate of change (which could indicate that their potassium level will eventually exceed an absolute or individualized maximum potassium level).

[0365] At box 620, once the user has completed a Zone 2 workout, the treatment management engine 114 can utilize information about the user acquired during the workout phase (e.g., analyte data, non-analyte data, user input, etc.) to optimize future workout phases. For example, this information can be used in one or more rule-based models at box 606 during future workout phases. In another example, this information can be used to personalize the ML models utilized at boxes 606 or 608 (e.g., training / retraining these ML models). Therefore, for future workout phases, the treatment management engine 114 may be able to more effectively predict the optimal workout parameters that will enable a particular user to reach the Zone 2 lactate range.

[0366] As described above, at least in some implementations, optimization of the exercise phase for a user can be achieved by monitoring the user's response to exercise and providing intensity guidance based on the user's lactate index obtained using a continuous analyte sensor system (e.g., continuous analyte monitoring system 104). However, in some cases, the lactate index may be unavailable and / or may be subject to a time lag (e.g., the lactate index may not indicate real-time blood lactate levels). In cases where the lactate index is unavailable and / or subject to a lag, certain physiological parameters directly related to the lactate index can be used as substitutes for the lactate index. These substitute parameters (referred to herein as physiological parameters) can serve as estimates of real-time lactate levels. Physiological parameters may include heart rate, respiratory rate, glucose, power, speed, strength, data from an accelerometer, etc. For example, during a previous exercise phase, when the user reaches the Zone 2 lactate range, the user may be at a heart rate of 120 beats per minute (as determined by post-exercise analysis taking into account the lag in lactate measurement). During future training phases, heart rate or other real-time physiological parameters can be used to estimate the user's real-time lactate level, so that when the user reaches a heart rate of 120 beats per minute, the treatment management engine 114 can assume that the user has reached the Zone 2 lactate range.

[0367] exist Figure 6BIf, at box 622, the treatment management engine 114 determines that the user has already selected a HIIT workout type or has already recommended HIIT based on the user's goals, time commitments, etc., then the method can begin at box 624. At box 624, the treatment management engine 114 can determine whether the user has completed any previously available HIIT phases by referring to the user's profile 118. For example, the treatment management engine 114 can examine the workout data in the user profile 118 to determine if there is data related to one or more of the user's past HIIT phases.

[0368] At box 626, if the user has previously completed a HIIT phase for which data is available, the treatment management engine 114 determines personalized exercise guidance for the user based on the user's own historical exercise data. For example, exercise data from the user profile 118 can be used to determine a set of exercise parameters for the user's exercise phase. This set of parameters may include (e.g., exercise intensity, exercise duration, etc., at the start of the phase and throughout the phase). Exercise intensity itself can be a function of speed, resistance, inclination, etc. Providing personalized guidance based on the user's own historical exercise data can be done in several ways. For example, a rule-based model can be used in conjunction with the user's own historical exercise data to provide a set of exercise parameters for the user's exercise phase. Alternatively or additionally, one or more ML models trained based on the user's own historical data can be used to provide a set of exercise parameters for the user's exercise phase. For example, the training server system 140 can retrieve user-specific exercise data from the user profile 118 to train one or more ML models.

[0369] At box 628, if the user has not previously completed a HIIT phase for which data is available, the treatment management engine 114 may use a non-personalized, population-based model to determine the exercise parameters for the user, at least until the user has performed one or more exercise phases and personal exercise data is available to the user. For example, the treatment management engine 114 may use a rule-based model that defines HIIT exercise parameters based on empirical studies involving population data. For instance, HIIT exercise parameters may be determined based on rules that are effective for user groups similar to the user (e.g., based on demographic and / or physiological variables such as age, sex, weight, height, BMI, etc.) to achieve a HIIT intensity range.

[0370] Alternatively or concurrently, the treatment management engine 114 may use one or more ML models to provide users with exercise parameters to achieve a HIIT intensity range (i.e., lactate levels in the HIIT zone). For example, the ML model may be trained based on population-based training data associated with athletes achieving a HIIT intensity range. The dataset includes data records, each containing exercise parameters, analyte data, non-analyte data, and / or other relevant information from the corresponding user's profile 118 for each exercise phase. As an example, each data record in the training dataset may include timestamped analyte and non-analyte data for the user's exercise phase. Data records may be labeled with one or more exercise parameters. Using such a training dataset, a model can be trained to predict one or more exercise parameters to help the user achieve a HIIT intensity range.

[0371] Before using a personalized or group-based model to guide the user through the exercise phases based on the exercise parameters determined at boxes 626 and 628, the therapy management engine 114 may instruct the user to begin exercising at a warm-up intensity for a certain duration. In some embodiments, the intensity may be half of the intensity later provided to the user to reach the HIIT intensity range. In some embodiments, the duration of the warm-up intensity may be approximately 5 minutes or 10 minutes.

[0372] At box 630, once the warm-up duration is complete, the treatment management engine 114 can instruct the user to increase the intensity to the recommended exercise parameters determined in boxes 626 or 628.

[0373] In some implementations, a user may be completing an exercise phase on an exercise machine (e.g., exercise machine 108), and the treatment management engine 114 may automatically set or gradually increase the intensity (e.g., speed, incline, and / or resistance, etc.). The treatment management engine 114 may automatically set the exercise machine to the exercise parameters determined at boxes 626 and 628. In implementations where the user is not using the exercise machine, the treatment management engine 114 may instruct the user to increase the intensity of their exercise by gradually increasing various exercise parameters until the desired HIIT intensity range is achieved. Increasing intensity may be achieved by changing various exercise parameters associated with the exercise machine (e.g., exercise machine 108) or by expending additional effort (in the form of speed or power) when the user is not using the exercise machine.

[0374] Typically, athletes may have an initial lactate level of approximately 1 mmol / L at the start of exercise. Then, during warm-up and as the user begins the HIIT phase, the rate of change in lactate may be slightly negative or slightly positive, but close to zero, because the rate of lactate production is roughly the same as the rate of lactate clearance in the body. However, given the high-intensity nature of HIIT exercise, the lactate slope may not show an initial slightly negative or slightly positive rate of change, but instead begins to increase rapidly. To ensure that the user reaches and does not exceed the HIIT intensity range, the treatment management engine 114 is configured to continuously monitor the user's lactate levels and physiological parameters to determine the optimal exercise intensity for reaching the HIIT intensity range. For example, if the user does not exercise at sufficient intensity, the user will not reach the optimal HIIT intensity range. Alternatively, if the user exercises at excessively high intensity, the user may not be able to maintain the exercise intensity to the optimal HIIT duration. Therefore, for the treatment management engine 114, it is crucial to continuously monitor the user's lactate levels (including the rate of lactate change) and physiological parameters instruct the user to increase or decrease the intensity to achieve the optimal HIIT intensity range, as further discussed in boxes 632 and 634.

[0375] At box 632, the treatment management engine 114 can determine whether the lactate level reflects or is expected to be close to the user's desired HIIT intensity range. For example, the treatment management engine 114 can monitor values ​​associated with the area under the curve (AUC) of lactate over time, the increase in lactate relative to the user's baseline lactate, and / or the rate of lactate increase to determine whether the user has reached or is expected to reach a HIIT intensity range (also referred to herein as the defined range). For athletes, reaching, for example, 12 mmol of lactate corresponding to the AUC could indicate that the user has reached the desired HIIT intensity range.

[0376] To determine whether a user is approaching a physiological state corresponding to the HIIT intensity range, the treatment management engine 114 can use one of a variety of rule-based models or ML models to obtain the user's lactate levels and other physiological parameters to predict whether the user is about to approach a physiological state corresponding to the desired HIIT intensity range.

[0377] At box 634, the treatment management engine 114 can observe lactate levels and, if the user is not inclined to achieve the optimal HIIT intensity range, instruct the user to increase or decrease the exercise intensity. For example, if the user does not exercise at sufficient intensity to reach the HIIT intensity range, or if the user exercises at too high an intensity that would cause the user's lactate levels to increase beyond the desired HIIT intensity range, it is considered that the user is not inclined to reach the desired lactate range.

[0378] Various models (e.g., rule-based, machine learning, or predictive algorithms) can be used to determine optimal exercise parameters, thus establishing an ideal intensity to ensure the user's lactate levels reach the desired HIIT intensity range without exceeding it. In some implementations, rule-based models can be utilized, where various rules can be defined around a set of parameters, including the user's current lactate level, current rate of lactate change, time since exercise began, exercise intensity, and other parameters. For example, an example rule might state that if the user's current lactate level is X, current rate of lactate change is Y, and time since exercise began is Z, then the intensity should be Q. Q can be the set of exercise parameters that bring the user into the HIIT intensity range, such as speed, resistance level, altitude, power, etc. Rules can become more granular and involve the entire user, along with other physiological and demographic indicators. For example, a rule might state that if the user is 40 years old and has heart rate A, temperature B, current lactate level X, current rate of lactate change Y, and time since exercise began Z, then the intensity should be Q.

[0379] In some other implementations, one or more prediction algorithms or models may be used to predict the optimal intensity to achieve the desired HIIT intensity range. For example, one or more models, which may be the same as or different from the personalized or population-based models described with respect to boxes 626 and 628, may be run continuously to obtain a set of inputs (e.g., the user's current lactate level, current rate of lactate change, time since the start of the workout, and other inputs received at box 402) and output the optimal intensity and / or corresponding workout parameters for achieving the desired HIIT intensity range.

[0380] When instructing a user to increase or decrease the intensity, the therapy management engine 114 can check whether the user is exercising outdoors. For example, non-analyte sensor data from a temperature sensor can indicate that the user is exercising outdoors, or the user's selection of, for example, outdoor running or walking as the exercise type can be used by the therapy management engine 114 to determine that the user is exercising outdoors. When the user is exercising outdoors, the therapy management engine 114 can determine air temperature, air quality, humidity, wind speed, known allergens, and altitude when suggesting ...

Claims

1. A monitoring system, the monitoring system comprising: A continuous analyte sensor, configured to generate a first set of analyte measurements correlated with the user's analyte level; and A sensor electronics module coupled to the continuous analyte sensor and configured to receive and process the first set of analyte measurements.

2. The monitoring system according to claim 1, wherein the continuous analyte sensor comprises: Base, The working electrode is disposed on the substrate. A reference electrode is disposed on the substrate, wherein the first set of analyte measurements generated by the continuous analyte sensor corresponds at least in part to an electromotive force based on the potential difference generated between the working electrode and the reference electrode.

3. The monitoring system according to claim 1, wherein: The continuous analyte sensor includes a continuous lactate sensor, and The first set of analyte measurements includes lactic acid measurements.

4. The monitoring system according to claim 3, further comprising: One or more memories, the one or more memories including executable instructions; One or more processors, which communicate data with the one or more memories and are configured to execute the executable instructions to: Based on a first set of analyte measurements obtained during the trial training phase or input received from the user, the user is classified as an athlete or a healthy user; and The training phase for the user is optimized based on the user's classification.

5. The monitoring system of claim 4, wherein classifying the user as an athlete or a healthy user based on the first set of analyte measurements includes detecting an initial lactate level and lactate threshold of 1.5 mmol of the first set of analyte measurements during the experimental training phase.

6. The monitoring system of claim 5, wherein classifying the user as an athlete or a healthy user based on the first set of analyte measurements includes associating the lactate threshold of the first set of analyte measurements with exercise parameters to classify the user as the athlete or the healthy user.

7. The monitoring system of claim 4, wherein classifying the user as an athlete or a healthy user is further based on nonanalyte data obtained from nonanalyte sensors during the experimental training phase, wherein the nonanalyte data includes accelerometer data, heart rate data, heart rate variability data, oxygen saturation data, blood pressure data, or body temperature data.

8. The monitoring system of claim 4, wherein the input received from the user includes self-classification information, the user's health goals, the user's exercise goals, or the user's historical exercise data.

9. The monitoring system of claim 4, wherein optimizing the exercise phase for the user includes: The training parameters for the training phase are determined based on the user's classification or the first set of analytes measured during the trial training phase. as well as Electronic signals are transmitted to the exercise machine so that the exercise machine operates based on the determined exercise parameters.

10. The monitoring system of claim 9, wherein optimizing the exercise phase further comprises: A second set of analyte measurements monitored during the exercise phase; Determine that the second set of analyte measurements is within the defined range for the said exercise phase; Determine whether to maintain the exercise parameters for the specified duration; as well as The exercise machine continues to operate based on the exercise parameters for a specified duration.

11. The monitoring system of claim 4, wherein optimizing the exercise phase further comprises: Provide the user with feedback on the effectiveness of the exercise phase.

12. The monitoring system of claim 11, wherein the effect of the exercise phase is determined based on the calorie burn and total energy consumption after the exercise phase.

13. The monitoring system of claim 10, wherein the one or more processors are further configured to: The second set of analyte measurements and the user's non-analyte data during the exercise phase are used to optimize future exercise phases.

14. The monitoring system of claim 4, wherein optimizing the exercise phase for the user includes: The training parameters for the training phase are determined based on the user's classification or the first set of analytes measured during the trial training phase. as well as The user is instructed to exercise according to the determined exercise parameters.

15. The monitoring system of claim 14, wherein instructing the user to exercise according to the determined exercise parameters includes the user manually adjusting the current exercise parameters to achieve the determined exercise parameters.

16. The monitoring system of claim 15, wherein the exercise parameters include speed, inclination, resistance, number of repetitions, or weight.

17. The monitoring system of claim 14, wherein optimizing the exercise phase further comprises: A second set of analyte measurements monitored during the exercise phase; Determine that the second set of analyte measurements is within the defined range for the said exercise phase; as well as Instruct the user to maintain the exercise parameters for a specified duration.

18. The monitoring system of claim 17, wherein optimizing the exercise phase further comprises: Provide the user with feedback on the effectiveness of the exercise phase.

19. The monitoring system of claim 18, wherein the effect of the exercise phase is determined based on calorie burning and total energy consumption after the exercise phase.

20. The monitoring system of claim 18, wherein the one or more processors are further configured to: Optimize future training phases based on the effects of the aforementioned training phase.

Images