Systems and methods for diagnosing likelihood of gastrointestinal disorders

Abstract

Systems and methods are provided herein for detecting flatus gas and / or diagnosing or predicting a gut disorder. Devices are provided that may be sized for positioning on or near a user's underwear. The devices may comprise a processor, a power source, a gas sensor positioned at an opening in the device housing and configured to generate an electrical signal indicative of the presence of a flatus; and a communication module to transmit data to a remote computing device. The processor may be programmed to: read the electrical signal output by the gas sensor; record flatus gas data based upon the signal; associate the flatus gas data with time data; and output the timed flatus gas data via the communication module. Methods are also provided for evaluating gut activity using such a device, which may include specific food challenge studies, evaluating dietary or medicinal intervention, and / or continuous flatus gas monitoring.

Description

SYSTEMS AND METHODS FOR DIAGNOSING LIKELIHOOD OF GASTROINTESTINAL DISORDERSCross Reference to Related Applications

[0001] This application claims priority to U.S. Provisional Patent Application no. 63 / 601,174, filed on November 20, 2023, the entire content of which is incorporated herein by reference.Statement of Government Support

[0002] None.Technical Field

[0003] The various methods, systems, devices, and processes described herein relate generally to the field of gastrointestinal tract (GI tract) sensing. In specific embodiments, the disclosure herein describes devices and methods for sensing attributes of gas from a patient’s GI tract such as flatulence to diagnose likelihood of GI disorders and / or effect of various GI interventions.Background

[0004] It has been estimated that 40% of adults suffer from gastrointestinal disorders. The most common cause of these disorders is the presence of Small Intestine Bacterial Overgrowth (SIBO) and varying degrees of malabsorption of specific carbohydrates, such as lactose or fructose intolerance, to name a few. One common symptom among these disorders is the excessive production of gas, which leads to bloating, abdominal pain, and increased flatulence. However, these symptoms are not specific to a particular disorder and cannot be clinically measured for diagnosis.

[0005] SIBO occurs when the microbiome in the small intestine excessively grows, causing an overproduction of gas, particularly hydrogen, as a result of bacterial fermentation(Dukowicz etal., 2007). It has been estimated that approximately 36 million American adults suffer from SIBO (Porter et al., 2017). The main cause of SIBO is multifactorial, but it has been established that certain types of foods, such as poorly absorbed carbohydrates called FODMAPs (fermentable, oligo-, di- and mono-saccharides and polyhydric alcohols), trigger the proliferation of microbes since they are poorly absorbed by the gut and thus available as a food source for microbial fermentation(Spiller, 2017). One of the main consequences of SIBO is excessive hydrogen production, which has been hypothesized to be the major cause of symptoms like bloating, abdominal pain, constipation, diarrhea, and flatulence. These symptoms can range from mild discomfort to severe pain, significantly affecting the quality of life. The main cause of these symptoms is summarized in Figure 1. In brief, the fate of sugars and nutrients that reach the small intestine is either gut absorption or acting as nutrients for the microbiome inhabiting the small intestine. In essence, there is competition between the body and the microbiome for the same food. Under normal conditions, there is a healthy balance between the gut and the intestinal flora, and thus the amount of fermented gas does not produce any discomfort for the individual. The anaerobic environment that characterizes the gut promotes sugar fermentation by the bacteria, resulting in the production of gas mainly composed of hydrogen (H2) and methane (CH4), with lower contributions from other gases such as hydrogen sulfide (H2S) and carbon dioxide (CO2)(Kirk, 1949; Suarez et al., 1998; Tomlin et al., 1991). However, if there is an overpopulation of bacteria, the balance is disrupted, and fermentation increases to abnormal levels, producing an excess of gas that raises the pressure on the gut walls, causing bloating and pain. This gas is eventually expelled from the body as flatus (colloquially known as farts), but this time the number, frequency, and volume of flatus are greater due to the excess of carbohydrate fermentation (Figure 1). Similarly, if an individual presents a malabsorptiondisorder, the same healthy homeostasis is disrupted because the poor absorption of nutrients means more unabsorbed carbohydrates are available for bacterial fermentation. In both scenarios - whether due to excessive bacteria or impaired absorption - the end result is increased bacterial fermentation leading to excess gas production and flatus. This explains why people experience a lot of flatulence after consuming certain foods such as beans or nuts or some sugar-free candies that contain fiber and carbohydrates that cannot be or are poorly absorbed by the gut. Many people who experience SIBO are not aware of their condition(Rao and Bhagatwala, 2019) and leave the problem untreated, living with constant discomfort. It has been suggested that SIBO can increase the risk factors for the pathogenesis of pancreatic carcinoma and cholangiocarcinoma (Ma et al., 2019).

[0006] Currently, the most common method for diagnosing SIBO and malabsorption is hydrogen breath testing. Unfortunately, this method has poor sensitivity and accuracy, and it is also timeconsuming and expensive. Breath-based testing suffers from several flaws, including that the concentration of gases of interest in breath is relatively low (making sensing difficult) and that breath testing is an active / disruptive test that cannot be conducted for lengthy durations of time. The low sensitivity, specificity and accuracy is due to low (0-100 ppm) concentrations of microbial-produced gases in breath(Nichols et al., 2021; Sachdev and Pimentel, 2013). Breath tests and symptom assessments are often subjective and may not provide an accurate picture of the ongoing disease. Moreover, the test is time-consuming, often takes place at the doctor’s healthcare clinic, only analyzes the malabsorption of one carbohydrate at a time, and does not capture the day-to-day symptoms. Doctors often end up relying on their patient’s medical history and symptoms, mostly diagnosing the condition after a long elimination process of other possible diseases. This can cause an additional burden to the health care system and to the patient.Furthermore, even when STBO is correctly diagnosed, monitoring the condition can be challenging. While breath testing can technically be used to follow-up on interventions, its poor sensitivity, inconvenience, and time-consuming nature mean it is rarely used in practice to assess treatment efficacy when SIBO treatment requires the use of antibiotics, such as metronidazole or levofloxacin, or rifaximin (Pimentel, 2009).

[0007] Consequently, there is a significant demand for innovative technologies that can address the diagnostic gap in gastrointestinal disorders.Summary

[0008] The present disclosure provides devices and methods that overcome the aforementioned drawbacks by providing various systems, devices, techniques and methods (including diagnostic methods, training methods, efficacy-determining methods, diagnostic methods, food impact determination methods, monitoring methods, and others) for detecting gastrointestinal gas produced during gut microbial fermentation, thereby enabling assessment of gut microbial activity.

[0009] In accordance with one aspect of the present disclosure, a device is provided comprising: a housing having a profile allowing it to be affixed to a user’s underwear in proximity to a perineal region of the user while the underwear is worn; a processor positioned within the housing; a gas sensor positioned within the housing and comprising at least one sensing element, wherein the at least one sensing element has a composition that interacts with at least one gas component of flatus emissions from the user to generate an electrical signal indicative of a presence or a concentration of the gas component; a communication module positioned within the housing and connected to the processor to communicate data from the processor to a remote computing device; and a memory positioned within the housing and having stored thereon a setof instructions which, when executed by the processor, cause the processor to: read the electrical signal generated by the gas sensor; record flatus gas data based upon the electrical signal; associate the flatus gas data with time data corresponding to a time the electrical signal was generated, to generate timed flatus gas data; and output the timed flatus gas data via the communication module.

[0010] In an alternative aspect, such a device may further comprise a power source positioned within the housing and connected to provide power to the processor, communication module, and memory throughout a study implemented via the device having a duration of at least twelve hours.

[0011] In another alternative aspect, such power source may comprise an electrolytic dual layer capacitor (EDLC) connected to provide rechargeable power to the device.

[0012] In another alternative aspect, such a device may have a gas sensor that is positioned in in alignment with an opening in the housing near the user’s perineal region, the opening configured to allow gas penetration to the at least one sensing element.

[0013] In another alternative aspect, such a device may also comprise a temperature sensor and a motion sensor, both disposed within the housing and connected to provide signals to the processor; and the set of instructions may further cause the processor to generate an indication of whether the device is being worn by the user based upon the temperature sensor signals and the motion sensor signals.

[0014] In another alternative aspect, the gas sensor of such a device may be an electrochemical sensor comprising a reference electrode and a counter electrode, and wherein the at least one sensing element is a working electrode comprising a material that exhibits a change in electrical current due to redox reactions when exposed to hydrogen-based gases.

[0015] In another alternative aspect, such a device may further comprise a quasi-solid electrolyte disposed in contact with the working electrode, reference electrode, and counter electrode.

[0016] In another alternative aspect, the quasi-solid electrolyte may comprise an absorbent substrate impregnated with a hygroscopic substance.

[0017] In another alternative aspect, the working electrode may be a platinum-based electrode, and the reference electrode and counter electrode comprise conductive carbon.

[0018] In another alternative aspect, the gas sensor of such a device may be a metal oxide semiconductor sensor and the at least one sensing element may comprise a metal oxide sensing layer of the metal oxide semiconductor sensor, wherein the metal oxide semiconductor sensor exhibits a change in electrical resistance in response to adsorption of hydrogen-based gases at the metal oxide sensing layer.

[0019] In another aspect, a system is provided for determining a likelihood of a gut condition. The system may comprise: a processor; a communication module; and a memory in communication with the processor. The memory may be storing software instructions that, when executed by the processor, cause the processor to: receive gas sensing data from a gas sensor, the gas sensing data comprising a time series of measurements indicative of concentration of a gas in a subject’s flatus emissions during a given time period; receive adherence data indicative of whether the subject was wearing the gas sensor when measurements of the time series were acquired; receive additional user data, indicative of a time when consumption of a food of interest occurred relative to the time series of measurements; determine at least one of: a timing of a first occurrence of a flatus emission after the food consumption of interest occurred, based on the gas sensing data; a microbiome activity index; or a total number of flatus emissions duringthe given time period after the food consumption of interest occurred; and output to a user an indicator of an effect of the food consumption on the subject.

[0020] In an alternative aspect, the software instructions of such a system may further cause the processor to, in association with receiving the additional user data, receive a time at which the subject consumed a food comprising a target carbohydrate or amino acid after having fasted.

[0021] In an alternative aspect, the software instructions of such a system may further cause the processor to output to a user an indicator of a likelihood the subject has a gut condition comprising carbohydrate malabsorption.

[0022] In an alternative aspect, the given time period of such a system may be at least three hours.

[0023] In an alternative aspect, the given time period of such a system may be at least one week; the time series of measurements may provide flatus monitoring throughout the user’s daily activities; and the microbiome activity index may be calculated over the given time period relative to all foods eaten by the user.

[0024] In an alternative aspect, the additional user data of such a system may be indicative of all foods eaten by the user during the given time period, and wherein the software instructions further cause the processor to identify a trigger food that, when eaten by the user, results in an increase in the microbiome activity index over a baseline microbiome activity index of the user by a threshold amount.

[0025] In an alternative aspect, the additional user data of such a system is determined via a user interface allowing the user to upload photos of the foods eaten for processing by a foodrecognition Al model, or enter nutritional information for the foods eaten; and wherein thesoftware instructions further cause the processor to determine ingredients of the foods eaten that may cause increased gastrointestinal activity.

[0026] In an alternative aspect, the adherence data of such a system is derived from at least one of an output of a temperature sensor disposed within a housing with the gas sensor or an output of a motion sensor disposed within the housing with the gas sensor.

[0027] In an alternative aspect, the additional user data of such a system a display screen of a mobile device may also be included, and wherein the software instructions further cause the processor to: prompt the user via the display screen to ensure the gas sensor is being worn at a location proximate to a perineal region; prompt the user to commence a dietary regimen; and prompt the user to consume the food of interest at a given time after commencement of the dietary regimen.

[0028] In an alternative aspect, the software instructions of such a system further cause the processor to monitor for peaks in a signal value of the gas sensing data indicative of a flatus emission, and to determine a confirmed flatus emission occurred when the signal value increases from a baseline level to a peak value exceeding a threshold value for nor more than a given period of time, then drops back to the baseline level.

[0029] In an alternative aspect, the microbiome activity index is indicative of a number of events in which the gas sensing data exceeded a threshold, and is calculated using an absolute value of a first derivative of the time series of measurements.

[0030] In an alternative aspect, the gas sensor of such a system may be any of the alternative devices noted above, and such system may be utilized to perform any of the methods noted below.

[0031] In an alternative aspect, the processor, communication module, and memory are part of the mobile device, the gas sensor may be any of the alternative devices noted above, and the system allows a subject to assess food impact outside of, and without involvement of, a healthcare clinic.

[0032] In other aspects, certain methods are also provided. For example, a method is provided for diagnosing a gut disorder comprising: causing a subject to fast for an initial fasting period; providing a wearable flatus sensor to be worn by the subject; administering a dietary intervention to be consumed by the subject, and recording a time the dietary intervention was consumed; continuously recording flatus data from the wearable flatus sensor throughout a measurement period, the measurement period comprising at least one hour; determining flatus activity based on the flatus data, the flatus activity including at least one of: time elapsed from consumption of the dietary intervention to first flatus emission, total number of flatus emissions, frequency of flatus emissions, intensity of flatus emissions, time distribution of flatus emissions, flatus emission volume, and total volume of all flatus emissions; and based on the flatus activity and time of consumption of the dietary intervention, providing a diagnosis of the gut disorder.

[0033] In alternative methods, a dietary intervention is a carbohydrate, and the gut disorder is malabsorption of the carbohydrate.

[0034] In alternative methods, the gut disorder is at least one of small intestine bacterial overgrowth (SIBO) or irritable bowel syndrome diarrhea dominant (IBS-D), and the diagnosis is made by providing the flatus activity and information concerning the dietary intervention to a trained machine learning algorithm configured to determine a likelihood of SIBO or IBS-D based on training data of past patients having SIBO or IBS-D and past patients that did not haveSIBO or IBS-D.

[0035] In alternative methods, the dietary intervention comprises a carbohydrate challenge utilizing at least one of glucose or fructose, and the gut disorder is secondary malabsorption in individuals with at least one of: recent chemotherapy, Celiac disease, or inflammatory bowel disease.

[0036] In alternative methods, the wearable flatus sensor comprises a device according to any of the alternatives noted herein, and may further be performed (in whole or in part) through any of the alternative systems provided herein.

[0037] In another aspect, a method is provided for evaluating a gut disorder, comprising: providing a wearable flatus sensor to be worn by a subject during a study period; recording flatus data from the wearable flatus sensor during the study period; receiving an indication of a reported stress level of the subject during the study period; establishing a baseline gutmicrobiome activity index by analyzing flatus data during periods of low or no reported stress level; measuring stress-induced changes in gut-microbiome activity by comparing flatus data during periods of elevated reported stress level against the baseline gut-microbiome activity index; quantifying a magnitude of deviation from baseline during periods of elevated reported stress level; determining temporal relationships between onset of elevated stress level and changes in gut-microbiome activity, including a temporal delay between stress onset and gutmicrobiome response and a duration of elevated gut-microbiome activity following stress onset; and generating a gut-brain reactivity profile based on: the magnitude of gut-microbiome activity changes during periods of elevated reported stress level; the temporal delay between stress onset and gut-microbiome response and the duration of elevated gut-microbiome activity following stress events.

[0038] In another aspect, the wearable flatus sensor of such a method may comprise any of the alternative aspects of the devices noted herein.

[0039] In accordance with another aspect of the present disclosure, a method is provided for diagnosing a gut disorder (and / or for providing a likelihood of, or detecting indicators of, such gut disorder) comprising: causing a subject to fast for an initial fasting period (which may be of variable duration and / or optional); providing a wearable flatus sensor to be worn by the subject; administering a dietary intervention to be consumed by the subject, and recording a time the dietary intervention was consumed; continuously recording flatus data from the sensor throughout a measurement period, the measurement period comprising at least one hour; determining flatus activity based on the flatus data, the flatus activity including at least one of: time elapsed from consumption of the dietary intervention to first flatus emission, total number of flatus emissions, frequency of flatus emissions, intensity of flatus emissions, time distribution of flatus emissions, flatus emission volume, and total volume of all flatus emissions; and based on the flatus activity and time of consumption of the dietary intervention, providing a diagnosis of the gut disorder. In accordance with yet another aspect of the present disclosure, a method is provided for long-term monitoring of gut health comprising: providing a wearable flatus sensor to be worn by the subject continuously for a period of at least one week during a subject or user’s daily activities; recording flatus data and food consumption data, wherein the food consumption data may be entered through a user interface that accepts uploaded photos of foods for processing by a food-recognition Al model or manual entry of nutritional information; calculating a microbiome activity index relative to foods eaten; and identifying foods that, when consumed, result in an increase in the microbiome activity index above a threshold amount compared to the subject's baseline.

[0040] These aspects are nonlimiting. Other aspects and features of the systems and methods described herein will be provided below.Brief Description of the Drawings

[0041] The patent or application file contains at least one drawing executed in color. Copies of this patent or patent application publication with color drawing(s) will be provided by the Office upon request and payment of the necessary fee.

[0042] The foregoing features of embodiments will be more readily understood by reference to the following detailed description, taken with reference to the accompanying drawings, in which:

[0043] FIG. 1 A is an exploded view of components of an electrochemical sensor in accordance with aspects of the present disclosure.

[0044] FIG. IB is a set of top and side views of components of an electrochemical sensor in accordance with aspects of the present disclosure.

[0045] FIG. 2 is a conceptual block diagram of a sensing device in accordance with aspects of the present disclosure.

[0046] FIG. 3 is a block diagram illustrating data flow within an example system in accordance with aspects of the present disclosure.

[0047] FIG. 4 is a flow diagram illustrating an example process 400 for determining a likelihood of a gut condition, in accordance with aspects of the present disclosure.

[0048] FIG. 5 is a flow diagram illustrating an example protocol method 500, in accordance with aspects of the present disclosure.

[0049] FIGS. 6A-C illustrate various prototype designs, in accordance with aspects of the present disclosure.

[0050] FIG. 7 illustrates conceptual view of a battery holder and separator, in accordance with aspects of the present disclosure.

[0051] FIG. 8 illustrates venograms depicting normalized and observed flatulence frequency and intensity, in accordance with aspects of the present disclosure.

[0052] FIG. 9 illustrates a comparison of sensor gain levels, in accordance with aspects of the present disclosure.

[0053] FIG. 10 illustrates a sensor analysis, in accordance with aspects of the present disclosure.

[0054] FIG. 11 illustrates calibration curves, response time, interference data, and other experimental findings associated with a sensor, in accordance with aspects of the present disclosure.

[0055] FIG. 12 illustrates results corresponding to temperature and accelerometer outputs and a wearing algorithm, in accordance with aspects of the present disclosure.

[0056] FIG. 13 illustrates an analysis of a Microbiome Activity Index, in accordance with aspects of the present disclosure.Detailed Description

[0057] As used in this specification and the claims, the singular forms “a,” “an,” and “the” include plural forms unless the context clearly dictates otherwise.

[0058] As used herein, “about”, “approximately,” “substantially,” and “significantly” will be understood by persons of ordinary skill in the art and will vary to some extent on the context in which they are used. If there are uses of the term which are not clear to persons of ordinary skill in the art given the context in which it is used, “about” and “approximately” will mean up to plus or minus 10% of the particular term and “substantially” and “significantly” will mean more than plus or minus 10% of the particular term.

[0059] As used herein, the terms “include” and “including” have the same meaning as the terms “comprise” and “comprising.” The terms “comprise” and “comprising” should be interpreted as being “open” transitional terms that permit the inclusion of additional components further to those components recited in the claims. The terms “consist” and “consisting of’ should be interpreted as being “closed” transitional terms that do not permit the inclusion of additional components other than the components recited in the claims. The term “consisting essentially of’ should be interpreted to be partially closed and allowing the inclusion only of additional components that do not fundamentally alter the nature of the claimed subject matter.

[0060] The phrase “such as” should be interpreted as “for example, including.” Moreover, the use of any and all exemplary language, including but not limited to “such as”, is intended merely to better illuminate the invention and does not pose a limitation on the scope of the invention unless otherwise claimed.

[0061] Furthermore, in those instances where a convention analogous to “at least one of A, B and C, etc.” is used, in general such a construction is intended in the sense of one having ordinary skill in the art would understand the convention (e.g., “a system having at least one of A, B and C” would include but not be limited to systems that have A alone, B alone, C alone, A and B together, A and C together, B and C together, and / or A, B, and C together.). It will be further understood by those within the art that virtually any disjunctive word and / or phrase presenting two or more alternative terms, whether in the description or figures, should be understood to contemplate the possibilities of including one of the terms, either of the terms, or both terms. For example, the phrase “A or B” will be understood to include the possibilities of “A” or “B” or “A and B.

[0062] All language such as “up to,” “at least,” “greater than,” “less than,” and the like, include the number recited and refer to ranges which can subsequently be broken down into ranges and subranges. A range includes each individual member. Thus, for example, a group having 1-3 members refers to groups having 1, 2, or 3 members. Similarly, a group having 6 members refers to groups having 1, 2, 3, 4, or 6 members, and so forth.

[0063] The modal verb “may” refers to the preferred use or selection of one or more options or choices among the several described embodiments or features contained within the same. Where no options or choices are disclosed regarding a particular embodiment or feature contained in the same, the modal verb “may” refers to an affirmative act regarding how to make or use an aspect of a described embodiment or feature contained in the same, or a definitive decision to use a specific skill regarding a described embodiment or feature contained in the same. In this latter context, the modal verb “may” has the same meaning and connotation as the auxiliary verb “can.”

[0064] Various embodiments, configurations, materials, devices, systems, methods, and techniques are disclosed herein for determining whether a given individual exhibits a likelihood of various gut-related disorders, through measuring gut microbial gas emissions (e.g., flatus). With respect to the devices and systems described below, certain alternative components and materials are described, none of which are intended to be limiting or required. The description of components of such devices and systems is intended to be illustrative only, and neither a minimum nor limit of the types of components that could be used in various embodiments hereof. Similarly, the methods described herein are explained with reference to optional steps and modifications, none of which are intended to be limiting or required. The methods described herein can be performed using hardware such as (or including) the devices and systems described herein butneed not be implemented through such hardware except in specific examples that identify the use of such hardware.

[0065] Embodiments disclosed herein provide for the measurement of characteristics of gut microbial gas emissions (e.g., flatus) and making determinations about a patient’s gas emissions that establish data on which a healthcare provider can assess the likelihood of certain statuses concerning the patient’s gut.

[0066] The inventors have determined that utilizing flatus as source of information about a patient’s gut activity provides certain advantages not present in existing methods of assessing gut activity. The correlation between flatus composition, intensity, number, and frequency reflects the activity of the gut microbiome. Sensing devices of the present disclosure capture and analyze information about flatus emissions using electrochemical sensors; in some embodiments just a single electrochemical sensor of various specialized designs described below.

[0067] In some embodiments, a sensing device comprising such an electrochemical sensor is disposed in or on the patient’s underwear for a period of time during which measurements of such gas emissions are made and data is acquired. Thus, certain design considerations can lend to better data acquisition. For example, the device should be contained within a housing having a size and form factor that patients find acceptable and comfortable to wear in or on underwear in a suitable location, such as within the intergluteal cleft or near the patient’s anus (provided, however, that alternative locations on or within a patient’s underwear, garment, or specific securing means such as a band, tape, or adhesive may likewise be utilized so long as near enough to emissions of flatus to allow for suitable detection). Likewise, in some embodiments a design goal might be for the device to have a power source and power consumption profile that avoids the need for bulky orexternal battery compartments or cords. Similarly, in some embodiments it may be desirable to avoid physical connects for data acquisition, such as USB cords or related wires. And, given that the device may be worn for a lengthy duration of time, some embodiments may avoid the use of consumables that need replenishing within a period of time that could interfere with a test and should be robust to the environment in which they operate (physical forces, temperature, humidity, moisture, foreign debris, etc.).

[0068] In some embodiments, it may be advantageous to reduce the number of components necessary to make a gas emission measurement and / or simplify the type or number of measurements required to make an assessment regarding the patient’s gut activity. For example, it may be desirable to have only one sensor measuring gas characteristics, and / or to measure only certain components of the patient’s flatus emissions (such as H2).

[0069] The present disclosure includes sections, below, setting forth examples of sensing devices and systems, as well as examples of methods and techniques for making assessments of gut disorder likelihood. The present disclosure also includes an Examples section, which provides further detail on some experiments performed by the inventors as well as various example embodiments and use cases.

[0070] Example Embodiments of Sensing Devices and Systems

[0071] Referring now to Fig. 1A, an exploded view is shown of components of one design of electrochemical sensor 100. The components include a PCB 102 having a set of electrodes 104, an electrolytic layer 106, and electrical leads connected to the set of electrodes 108. In some embodiments, an electrochemical sensor 100 may operate based on amperometric principles, rather than metal-oxide or chemi -resistor principles. As shown in FIG. IB, the set of electrodes104 on the electrochemical sensor 100 comprises three electrodes: a counter electrode, a reference electrode, and a working electrode. Shapes, orientations, and positioning of electrodes other than as shown in FIG. IB are contemplated, for example including electrodes on opposing sides of the substrate and two, four, or five electrodes. These electrodes operate in coordination to make measurements of gas characteristics (as opposed to being thought of as separate sensors or complementary sensors). The electrodes may be formed from various conductive compounds, and may be attached to or printed onto a substrate such as a PCB. In some examples, the one or more of the electrodes may be a platinum-based electrode. In one embodiment, carbon conductive ink was used for the counter and reference electrodes and another platinum-NafionO-carbon ink (Pt- Nafion-Carbon,l% Nafion, and 3% Pt) for the working electrode. Platinum is a catalyst for hydrogen gas oxidation, a redox reaction that occurs at open circuit potential (0 bias). Thus, the platinum in the wording electrode provides for improved sensitivity and selectivity of the electrochemical sensor with respect to hydrogen measurement. Nafion® is a type of conductive polymer (a sulfonated tetrafluoroethylene-based fluoropolymer-copolymer) that is added to carbon ink as an additive to improve conductivity and / or sensitivity. It should be understood, however, that the specific metals, polymers, coatings, inks, etc. that were used in the inventors’ experiments and prototype designs are not limiting. The inventors selected carbon ink, platinum, and Nafion for a variety of reasons (cost, complexity, and target application), but it is contemplated that other materials are also suitable for use, such as where other gas components are desired to be sensed (e g., hydrogen-based gases) and / or other cost or complexity constraints apply.

[0072] In one example embodiment, PTFE filter paper soaked in sulfuric acid (shown in Figure IB) was used to create an impregnated absorbent substrate by providing an electrolyte for making the gas measurement, though other hygroscopic substances / materials and other absorbentsubstrates (e.g., other papers, cellulosic materials, natural fibers, polymer-based and polymer fiber materials (woven or non-woven), ceramic-like materials, glass fibers, porous materials, spongelike materials, gels, membranes, etc.) are also contemplated to be utilized as an electrolyte. The paper was embedded in a solution of strong sulfuric acid and subsequently dried at room temperature. Strong sulfuric acid has a very low evaporation rate due to its low vapor pressure (0.001 mmHg at 25 °C) and thus after drying a non-visible layer of acid is still adsorbed to the filter pad. Moreover, sulfuric acid is hygroscopic, meaning it can absorb water from the environment keeping the pad highly conductive but without any visible liquid. As a result, a quasisolid electrolyte is formed that keeps a high conductivity rate at 0.056 (+ / - 0.013, n=6) S cm'1which can be comparable with the conductivity of Nafion 0.079 - 0.2 S cm'1. The pad’s high conductivity is maintained for at least a month even under continuous exposure to air.

[0073] Carbon as the material for an electrode offers certain advantages such as a stable potential with little changes as small as 1 mV per day, being even more stable than a standard silver chloride electrode (Ag / AgCl). While carbon electrodes may be sensitive to chloride concentrations which are ubiquitous in the environment, for the gas sensing application chloride is generally absent and thus carbon can be used as a quasi -reference electrode. Additionally, silver or silver chloride as a material for the reference electrode would not be ideal for use in a strong acidic environment as it will gradually dissolve in the acid. Although this process is slow it limits the sensor’s long-term performance.

[0074] In another embodiment, the electrochemical sensor is composed of a conductive carbon ink with platinum and Nafion as working electrode, and a bare conductive carbon ink as counter and reference electrodes respectively. In some configurations, the electrodes may be screen printed on the flat surface of a PCB. In other configurations, alternatives to a PCB may includeKapton or ceramic substrates. In yet further embodiments, a PCB is not used and the electrodes are self supporting or integrated into the housing of the device. In some configurations in which a PCB or other substrate is used, the PCB or substrate may form a water-tight seal within the housing of the device, and / or may be air permeable. A quasi-solid electrolyte is fabricated from a cellulose paper, previously moisturized with 6 pL of concentrated (4 mol L-l) sulfuric acid and dried at room temperature for at least 20 min. On top of the electrode system a commercial PTFE membrane is placed and pressure towards the electrodes is applied to maximize the conductivity (Figure 1). In some embodiments, the whole electrode system is protected by a safe 3D printed PETG case. The case is sealed by melting the 2 case components together with acetone.

[0075] During operation, the electrochemical sensor exhibits an electrical output via leads, which provide an analog signal to a processor for conversion into data indicative of hydrogen presence. The electrical output varies roughly linearly with increasing addition of H2 acting on the Ptcontaining working electrode, and thus the sensor can be thought of as acting like a potentiostat. In some embodiments, the particular materials chosen for the electrolyte and electrodes, and even the surface area and shape of the electrodes, may provide for different, even non-linear, changes in electrical output of the sensor. Thus, a memory associated with the processor monitoring the output of the sensor leads may include a lookup table or other conversion algorithm for interpreting H2 presence from output of the sensor.

[0076] Referring now to FIG. 2, a conceptual block diagram of a sensing device 200 is shown. The device 200 comprises an outer housing 202. The housing 202 defines a size and profile of the device 200, as the principal components thereof are stored within the house 202. As shown, the housing is shaped as roughly a flat disk or cylinder, though other shapes are contemplated, such as wider / thinner shapes, contoured shapes designed to follow the shape of the human body, etc.The housing is designed to be easily attachable to a user’s underwear, belt, harness, or skin-friendly adhesive, and thus may include contours, depressions, or other areas suitable for clips and / or attachments.

[0077] The housing 202 includes at least one opening 204. In some embodiments, other than the opening 204, the remainder of the housing may be sealed to prevent water or other unwanted intrusion. The opening 204 may directly interface with the electrochemical sensor 206, such as being sized and / or positioned to align with the electrodes of the sensor 206. In other embodiments, the opening 204 may include an air-permeable membrane, mesh, or other covering that still allows air infiltration. In some embodiments, more than one opening 204 may be utilized, such as on opposing or neighboring surfaces of the housing 202 to promote air flow over the sensor 206. Certain examples of sensor profiles, sizes, shapes, orientations, configurations and housing designs are shown in FIGs. 6A-C; however, none of these are intended to be limiting, and the housing 202 may define a variety of shapes and sizes, and influence configuration of the internal components according to form factors that are more optimal for positioning, subject comfort, and flatus gas detection. Similarly, the manner in which such sensors are affixed may change as well. For example, some sensors may be housed in a flexible / fabric body that can be adhesively worn by a user, some may attach to a belt loop, some may be worn by a pin to the user’s undergarments or clothing, or may be affixed on clothing through a magnetic or compression attachment (e.g., wherein one part of the attachment is on an opposite side of clothing from the sensor main housing), or may simply be part of a belt or other affixation means.

[0078] Sensor 206 may be a sensor as described above with respect to FIGs. 1A and IB. In some embodiments, the form factor of the sensor 206 may be such that no air, water, and / or foreign debris is able to pass the sensor 206 to other components of the device 200. For example, thecircumference of sensor 206 may interface with housing 202, and / or may interface with gaskets or other sealing means. The electrodes of sensor 206 (through an opening of sensor 206 and / or through the filter / electrolyte pad of the sensor) are disposed so as to receive air through opening 204. Sensor 206 may be connected to a power source 216 so as to generate an electrical potential or other characteristic across the electrodes. In other embodiments, the sensor 206 may not require a connection to a power source in order for measurements across the electrodes to be made.

[0079] The electrical output leads of sensor 206 may be connected to a processor 208. In some embodiments, the leads may first be connected to other electrical components such as analog-to- digital converters, filtering components, capacitive elements, or the like. Processor 208 may be a microcontroller or other similar computational resource that can read analog or digital data indicative of the output of the sensor 206. Processor 208 may run software that is stored on a memory, such as memory 210. For example, the software may cause the processor to detect and digitize the electrical output of the sensor 206 on a given sampling frequency. In some embodiments, the sampling frequency may be relatively low (e.g., once per every 5s, 10s, 30s, Im, 1.5m, 2m, 3m, 4m, or the like) while the readings from the sensor indicate a comparatively low level of hydrogen, and may increase in frequency whenever the readings from the sensor indicate a comparatively high level of hydrogen. The measurements may be stored in a memory such as memory 210, and supplemented with time stamp data based upon a time counter operated by the processor that is synchronized to a clock / time counter of an external device with which the device 200 may interface. The time stamp data may correlate to an actual time of day (e.g., based on a recognized clock), or may merely correlate to time elapsed since a given occurrence (e.g., the device was turned on).

[0080] As shown, the device 200 comprises one chemical sensor 206 for detection of gas characteristics. The single sensor 206 allows for measurement of hydrogen levels sufficient to make highly accurate and sensitive determinations concerning flatus emissions, without added complexity of multiple sensors. The capability to require only one sensor 206 for chemical measurements results in multiple advantages, such as reduced size of the device 200, reduced power consumption, and a higher degree of reliability.

[0081] In some embodiments, the software stored on memory 210 may also provide for further processing by the processor of the data generated by sensor 206 such that it is augmented and / or compiled and transformed into higher level assessments. For example, processor 208 may bin sensor data into chronological categories during a test, such as occurring pre-consumption of food of interest or after consumption. Similarly, processor 208 may convert the electrical signals of sensor 206 into imputed hydrogen levels at given times, using calibration data or known output curves. Processor 208 may also determine the occurrence of individual flatus emissions from sensor data. For example, as hydrogen levels increase to a certain elevated level (e.g., lOx of a given baseline level), a determination can be made that an emission was likely. Then, if the hydrogen level continues at an elevated amount for a given duration, then processor can categorize the detections made during that duration as being an individual emission. In other words, the processor would be programmed to identify spikes in hydrogen levels of a duration within a given window and designate those as indicating a flatus emission. In some embodiments, once an elevated level of hydrogen is initially detected, the processor may begin sampling the sensor output on a more frequent basis, such as multiple times per second, once per second, etc., until hydrogen levels return to a baseline level or near baseline level. Thus, if two emissions occur in succession, individual peaks in hydrogen levels can be categorized as separate occurrences. Based on theidentification of individual emissions, the processor can also determine duration of time to first emission, total number of flatus detected, frequency of emissions and time between given emissions, as well as individual and total volumes and intensities of emissions. As another example, data from sensor 206 may be used to monitor total volume of emissions, rate of emissions, and / or total number of emissions as indicators of gut-microbiome activity; thus, when the data demonstrates elevated gut-microbiome activity, various identifications can be made relevant to an individuals gut condition or sensitivities, as described herein.

[0082] In further embodiments, an optional additional sensor 214 may be included in the device 200, for purposes of assessing whether the patient was complying with the prescribed duration of wearing the sensing device 200. Thus, optional sensor 214 may include an accelerometer or other motion sensor, a temperature sensor, or a humidity sensor. Processor 208 may be programmed to supplement the gas sensor data 206 with an indication of whether the additional sensor output 214 suggested the user was wearing the sensor 200 when a given hydrogen reading took place. For example, if a temperature sensor 214 indicates the device 200 was likely touching or worn by a user (e.g., close to surface skin temperature) during a given timeframe (e.g., 2 or more consecutive temperature samples), then gas sensor data measured during that timeframe can be augmented with an indicator that the data is a valid measurement, and vice versa. In a similar way, if a movement or humidity sensor 214 provides measurements indicate the device 200 is being worn during a given timeframe, then gas data acquired during that time frame can be flagged as valid, and vice versa.

[0083] Device 200 may also include an onboard power source 216, as well as a communications module 212. Power source 216 may include lithium-ion or other rechargeable power supply or rechargeable battery type, low-cost coin cell batteries, super capacitors, and energy storagecapacitors (electrolytic dual layer capacitors (EDLC)). Communications module 212 may be wireless, such as a Bluetooth transceiver, local WLAN connection, customizable RF transceiver, or the like.

[0084] Referring now to FIG. 3, a system 300 is illustrated in a conceptual diagram. A sensor device 304 takes gas measurements for a given individual during a specific timeframe, such as a prescribed test, a given 24 hour period, waking hours only, sleeping hours only, or specific postmealtime periods. The device 304 may comprise a device such as device 200 of FIG. 2, or similar devices. The data 302 generated by the device 304 is indicative of gas characteristics of emitted flatus of the wearer. As described above, gas sensing data 302 may be augmented, supplemented, binned, or extracted into other forms of data. For example, in one embodiment gas sensing data may comprise a corresponding time stamp for each measurement. The gas sensing data 302 may be transmitted to a remove computational resource, such as a remote computer 310. Transmission of the gas sensing data 302 may occur via local Bluetooth connection, or via an Internet connection to a remote service, or other communications network 320.

[0085] In addition to gas sensing data 302, system 300 may optionally include the recording and transmission of additional user data 306. For example, additional user data 306 may be entered into a mobile device of the patient or into a patient’s EMR, or other user interface. The data 306 may include user information such as age, sex, prior medical history, or other medical record information. Data 306 may also include information entered by the wearer or a healthcare professional to indicate what foods / compounds were consumed and when. For example, data 306 may be extracted from a food journal kept by the wearer, or from healthcare records indicating when and what a patient was fed. In other embodiments, the additional user data 306 may include times at which fasting began, duration of fasting, times at which the user was in a restroom or ona toilet, and times at which various interventions were administered such as antibiotics, probiotics, elimination diets, etc.

[0086] Computing device 310 includes a processor 312, memory 314, communications system 316, user inputs 318, and a display 320. In some embodiments the additional user data 306 may be entered directly into computing device 310 by a user. Memory 314 may contain software that generates certain determinations based on the gas sensing data 302 and additional user data 304. For example, emittance information such as time, frequency, and intensity of flatus emissions can be determined from the gas sensing data 302. At a higher level, likelihoods of various gut disorders may also be determined from the sensing data 302, per the methods described below. The output of such processing may be displayed to a user via display 320 or a display screen, such as to a healthcare professional monitoring the wearer.

[0087] FIG. 4 is a flow diagram illustrating an example process 400 for determining a likelihood of a gut condition, in accordance with some aspects of the present disclosure. As described below, a particular implementation can omit some or all illustrated features / steps, may be implemented in some embodiments in a different order, and may not require some illustrated features to implement all embodiments. In some examples, an apparatus (e.g., device 200 / 304, processor 208 / 312 with memory 210 / 314, etc.) in connection with FIGS. 2 and 3 can be used to perform all or part of example process 400. However, it should be appreciated that other suitable processing hardware for carrying out the operations or features described below may perform process 400.

[0088] At step 402, the process 400 receives gas sensing data from a gas sensor. As described with respect to various hardware implementations described herein the gas sensor may be a wearable flatus sensor device, that has a sensor for detecting the presence and / or concentration ofgas components of flatus, such as via electrochemical sensing or metal oxide semiconductor-based sensing. For example, the gas sensor may generate data or effectuate a change in electrical characteristic that can be processed as gas sensing data that is indicative concentrations of gas components of flatus emissions, such as for example hydrogen (H2), carbon dioxide (CO2), nitric oxide (NO), hydrogen sulfide (H2S), as well as other gases. In some examples, the gas sensing data can include a time series of measurements that indicate a concentration of a gas in a subject’s flatus emissions during a given time period. Thus, the gas sensing data may be associated with a time stamp, counter increment, or other time data to form a dataset or stream of supplemented, timed gas sensing data.

[0089] The time series of measurements may be continuous, periodic, on-demand, or triggered by external criteria, and may have a constant or dynamic sampling rate. For example, if an amplitude or value of the gas sensing data remains relatively stable at a baseline value, the sampling rate may remain at a comparatively lower rate. But, if an elevated reading is obtained, the sampling rate may dynamically increase to generate more data for relevant flatus events and time periods, and vice versa. Likewise, the time period during with the measurements are acquired may vary according to the purpose for which process 400 is being used. For example, where process 400 is a challenge-type study in which a given food will be administered under controlled circumstances to measure its effect on gastrointestinal activity (as determined by various determinations described herein), the periodicity of the time series of measurements and duration of the measurement period may be adjusted to correspond to the duration and parameters of the study (e g., whether a baseline measurement will be obtained, or only post-consumption measurements, duration of the overall study, necessary time-fidelity of measurements of flatus emissions, etc.).

[0090] At step 404, the process 400 optionally receives adherence data indicating whether the subject was wearing the sensor. In some examples, the adherence data may indicate whether the sensor was turned on and operating correctly and / or if a subject was wearing the sensor while some or all of the measurements of the time series gas sensing data were acquired. For example, as described in more detail below, the adherence data may include signals corresponding to one or more accelerometers, temperature sensors, humidity sensors, external “touch” -based electrodes, or the like in the sensor or a garment in which the sensor is attached to. The signals may be compared to a set of threshold data, which may indicate if the sensor is appropriately worn by the subject. In some embodiments, the adherence data may be acquired continuously, only when gas sensing data is acquired, or at a different frequency than the gas sensing data (e.g., only once per 5 minutes, 15 minutes, 1 hour, 2 hours, etc. versus gas sensing data being every 0.25 seconds, 0.5 seconds, 1 second, 1.5 seconds, 2 seconds, 5 seconds, 10 seconds, 1 minute, etc.). In further embodiments, the user may be given the option to self-record adherence data, such as instances of going to the bathroom, showering, changing, etc. and may also have the option to self-report flatus information during the periods of time the sensor was not being worn in place.

[0091] At step 406, the process 400 optionally receives user data. In some examples, the user data may indicate a time (which can be a given time / date, or a signal that initiates or increments a counter relative to the time series gas sensing measurements) when an intervention of interest occurred; when a food of interest was consumed; images or data regarding regular food consumption; when a dietary regimen commenced; when a period of fasting commenced; etc. For example, in some examples, the user data may indicate a time at which the subject consumed a specific type of “challenge” food, such as a carbohydrate or amino acid. For example, a subject may input a time when they consumed food, as well as the type of food consumed. In someexamples, the time may be indicative of a period of time in which the subject was fasting and / or broke their fast. In further embodiments, the user data may include data entries such as confirmation of regimen following, confirmation of detected flatus emission (for validation purposes), etc.

[0092] At step 408, the process 400 optionally filters the gas sensing data to remove measurements that were acquired when the subject was not wearing the sensor, or other invalidating circumstances occurred (e.g., the user notes that they did not adhere to a fasting or dietary regimen). In some examples, the removal of this gas sensing data may ensure the determinations, as described below, correspond directly to relevant gas sensing data obtained by the subject during a desired timeframe.

[0093] At step 410, the process 400 identifies gas sensing data of the time series that are likely to correspond to occurrence of flatus emissions. For example, where gas sensing data indicates a comparatively lower value or baseline value, then increases beyond a threshold to a peak value for a period of time (such as shown in the graphs of FIG. 8), followed by dropping down to a lower / baseline value, process 400 can identify that peak as a confirmed flatus emission. In some examples, the process 400 may tag one or more flatus emission occurrences throughout the duration of the gas sensing data, or may cease upon detection of the first flatus data. For example, the process 400 may determine a total number of flatus occurrences during a specific time period, an intensity of the emission(s), as well as a volume associated with each emission. In some examples, a flatus emission may be detected by monitoring temperatures and intensities contained in the gas sensing data.

[0094] At step 412, the process 400 determines a time of emission corresponding to the occurrence of flatus. In some examples, the time may be used to determine a time elapsed between consumption of food (as received at step 406) and a flatus emission. Moreover, in some examples, the time may be used to determine a frequency of flatus emissions.

[0095] At step 414, the process 400 outputs an indicator of a likelihood the subject has a gut condition. For example, the process 400 may use the gas sensing data, alone or in combination with the user data received at step 406, to determine a likelihood of the subject have carbohydrate malabsorption, small intestine bacterial overgrowth (SIBO), inflammatory bowel syndrome diarrhea dominant (IBS-D), as well as other gastrointestinal conditions.

[0096] Methods of Use in Human Assessments

[0097] In some aspects, the present disclosure enables a variety of assessment and sensing methods that can provide data relevant for use by healthcare professionals in considering diagnoses of malabsorption, SIBO, or other gut disorders. For example, gut microbial gas production can be measured using the sensors, devices, and systems described above, and resulting data can be provided to healthcare professionals in a variety of manners.

[0098] Referring now to FIG. 5, an example protocol method 500 is illustrated in flow chart form. The process 500 illustrated in FIG. 5 could be utilized as a protocol framework for monitoring and diagnosing gut conditions relating to a given food or substance of interest, using a wearable flatus sensor. However, as further described below, adaptations of this protocol method 500 may be utilized for other gut condition monitoring and diagnosis, and / or for general tracking of gut response to a user’s ordinary diet over time.

[0099] At block 502, process 500 may include instructing a user to commence an initial dietary regime. In some embodiments, a software application running on a user device (such as, e.g., a mobile phone) may provide the user with software instructions on how to follow the initial dietary regime. This may entail instructing the user in regard to which types of foods to avoid, times of day to eat or not eat, overall food intake, certain food types that must or must not be eaten, as well as graphical depictions of example foods. The application may also instruct the user regarding the duration of the period in which the initial dietary regime should be followed, and when it will conclude. In some embodiments, the initial dietary regime may be tailored to the type of study to be conducted for the user, which may be defined by the user, selected by the user from among available pre-planned studies, or defined by a healthcare provider through an EMR integration or healthcare provider portal. For example, the user may be instructed to avoid fiber-rich foods and may be presented with examples of foods to avoid and / or pictures of such foods, based on the goal of the study. As another example, the user may be instructed to avoid gluten, sugar alcohols, or other dietarily sensitive ingredients. In some embodiments, the application may monitor the user’s compliance with the regime through self-reporting or by integrating with external dietary tracking tools. Furthermore, in alternative implementations of process 500, the initial dietary regime may be unnecessary, such as when a user confirms they have not eaten certain avoidance foods in the past 24 hours (or other time period) or when a general monitoring of regular diet is to be performed.

[0100] At block 504, the process may include instructing the user to begin a fasting period. This instruction may be delivered through the same user interface or application described in block 502. In some embodiments, the system may provide reminders and alerts as the fasting period approaches. For instance, a mobile application may generate notifications indicating when to stop eating or drinking, as well as warnings if the fasting period is not initiated on time. The durationof the fasting period may be preset by the system or customized based on the type of study being performed and / or the target carbohydrate or amino acid to be fasted. The fasting period may vary between different protocols, such as shorter periods for general gut health assessments or longer durations for studies requiring significant baseline reduction in gut activity, or may even be optional in some implementations.

[0101] At block 506, process 500 may include sending a notification or reminder to the user regarding the impending fasting period, and / or requesting information to confirm adherence to the initial dietary regime. In some embodiments, this may involve presenting the user with a checklist or questionnaire to verify compliance with dietary restrictions. For instance, the user may confirm whether they have avoided specific foods or followed required eating patterns. In some examples, process 500 may also offer dynamic feedback or suggestions if the user reports partial compliance, such as extending the preparation period or adjusting the fasting protocol. In some embodiments, the notifications, reminders, and / or requests for confirmation may be communicated to the user via graphical, textual, or audio prompts, such as via software running on or integrated with mobile or wearable devices of the user. In other embodiments, data concerning the user’s food consumption may be obtained from other sources, including: food entries in an in-patient individual’s medical chart; and / or environmental monitoring devices to independently verify dietary adherence, such as through blood glucose monitors or smart devices tracking food intake.

[0102] At block 508, process 500 may prompt the user to prepare or obtain a designated study food. For example, as the fasting period nears its conclusion, the user is prompted to prepare a designated study food. This step may involve detailed instructions provided through a user interface, including recipes, ingredient lists, and preparation techniques. In some embodiments, the study food may be pre-packaged and provided to the user, with preparation instructions tailoredto its use in the study. For example, the system may recommend a specific carbohydrate or protein source designed to stimulate gut activity for diagnostic purposes. The application may also provide timers or visual guides to assist the user in preparing the food correctly. Additionally, the system may confirm the user’s adherence to fasting and preparation instructions through questionnaires, user-uploaded photos, or other compliance verification tools.

[0103] At block 510, the user may be instructed to consume the prepared study food and confirm the proper placement of the wearable flatus sensor. The sensor may be positioned on or near the user’s undergarment, and the system may provide visual or video guides to ensure correct placement. In some embodiments, the wearable sensor may include a self-check feature, such as a notification confirming proper positioning based on data quality or environmental parameters. The system may also instruct the user on how to securely affix the sensor to prevent displacement during the study period. Notifications may alert the user if the sensor is removed or improperly aligned. In further embodiments, adherence sensing may include use of temperature sensor data, accelerometer / motion sensor signals, etc., as described herein.

[0104] At block 512, process 500 may acquire sensor measurement data. For example, the wearable sensor device (such as described above) measures flatus emissions during the study period. This may include continuous or interval-based measurements of electrical values of a sensor indicative of the presence of specific gas components, such as hydrogen-based compounds, nitrogen-based compounds, various sulfides, ammonia, methane, VOCs, or other diagnostic markers. For example, the sensor may include an electrochemical sensor as described above to detect total nitrogen content or total hydrogen content of the sensed gas. The system may also record the time and frequency of emissions to establish temporal relationships and / or patterns. In some embodiments, the sensor may incorporate additional environmental sensors, such asaccelerometers or temperature monitors, to validate that it is being worn correctly and operating within expected parameters. Data collected by the sensor may be stored locally or transmitted to a cloud-based system for real-time analysis and storage.

[0105] At block 514, process 500 may optionally detect indications of non-adherence or improper operation of the sensor. For example, if no data is received for an extended period or if recorded data deviates significantly from expected patterns (e.g., continuously high nitrogen or hydrogen readings that could not constitute a flatus emission due to duration, or atmospheric conditions due to high content of these gases), the system may alert the user to verify sensor placement, perform a diagnostic, seek technical support, or verify protocol compliance. Notifications may be delivered through the software application, email, or text messages, and may include troubleshooting steps or contact options for technical support. In some embodiments, the system may utilize a trained machine learning algorithms to identify anomalies and distinguish between user non-compliance and potential technical issues with the sensor.

[0106] At block 516, process 500 may cause the sensing device to output the data acquired during the study to a user interface and / or integrate it into an electronic medical record (EMR). The user interface may present the data in a visually intuitive format, such as graphs or charts showing flatus activity over time. In some embodiments, the data may include higher-order interpretations, such as likelihood scores for specific gut conditions. Integration with an EMR may allow healthcare providers to review the results and use them in conjunction with other diagnostic information. Data privacy and security measures, such as encryption and access controls, may be implemented to ensure compliance with healthcare regulations.

[0107] Thus, processes deriving from process 500 may provide a comprehensive and customizable method for collecting and analyzing gut gas emissions. By incorporating multiple layers of user guidance, compliance verification, and data analysis, the system ensures high-quality data acquisition for diagnosing conditions such as small intestinal bacterial overgrowth (SIBO), irritable bowel syndrome with diarrhea (IBS-D), and carbohydrate malabsorption. Furthermore, by integrating the user interface in a patient-accessible device (e.g., mobile device) and allowing for installation of the sensor within the patient’s clothing, studies and monitoring may be performed by the patients themselves, in their own homes / workplaces, during a normal routine, on their own schedules - benefits which have not been achievable using prior methods like breath testing.

[0108] With the overall highlights of process 500 in mind, various specific implementations and example studies will now be described.

[0109] For some methods, an initial data gathering study may be performed in order to train machine learning models, and / or to tune various thresholds and / or determination scales for specific dietary items, classes of individuals / patients, or even specific patients (e.g., in a precision or personalized treatment context). For example, several protocols for developing training data on which to tune various thresholds are further described below. In one such example, two groups of individuals may participate in the data gathering. One group is known to have conditions such as SIBO, IBS-D, or other causes of malabsorption for a given type of food item. The other group is known not to have such conditions. The groups are randomized and given either a placebo, or the food item known to cause gut distress for the condition group. Measurements can be made of time to first flatus emission, total number of emissions, volume / intensity of each emission, a microbiome activity index, and / or total volume / intensity of emissions during a given time period,such as 4 hours (more detail on the specific protocol is described below). Then, a regression, average, or machine learning algorithm can be applied to determine the likelihood that an individual exhibits symptoms of a condition of interest based upon their measured flatus activity. In another example, a more simplified protocol can be conducted to tune the threshold value(s). In such example, a FODMAP is chosen that is always malabsorbed by humans. A group of participants are given the FODMAP to consume, and the time to first flatus, total number of emissions, volume / intensity of each emission, and total volume / intensity of emissions during a given time period are measured for the group. Based on this, a threshold can be determined such that, if a given individual exhibits certain flatus activity when consuming another given FODMAP, they likely have a sensitivity to that FODMAP or likely have a condition such as SIBO, IBS-D, or other cause of malabsorption.

[0110] In one implementation, a method can be performed for measuring characteristics of a patient’s flatus that provide insight into the likelihood of malabsorption, such as malabsorption of a dietary carbohydrate. The characteristics that are determined through measurement of flatus can be interpreted to gain insight into gut microbial gas production. For example, in one method a patient may be instructed to avoid eating foods containing fermentable oligosaccharides, di saccharides, oligosaccharides, and polyols (FODMAPs) for at least 24 hours before the beginning of measurements. This could be extended to 72 hours for some patients dependent on background results from previous measurements. Patients may receive instructions on how to complete the test via instructions from a smartphone application or printed booklet. Alternatively, healthcare professionals may be given instructions via an Electronic Medical Record or similar notification means (e.g., in the circumstance of in-patient care) regarding what the patient can orcannot eat. In further examples, specific foods may be recommended or prescribed which are known to be unlikely to cause generation of excess gases of interest during the measurement.[0U1] Next, the patient can be instructed or caused to fast for 8-12 hours before beginning measurements. In certain cases, fasting could be extended to 24 hours dependent on background results from previous measurements and patient input. E.g., if the patient continues to report bloating sensations after 8 hours, the fasting period can be extended, or other interventions could be employed to reduce the presence of accumulated gas in the patient’s gut. At the end of the fasting period (or earlier, if desired), a wearable sensor such as described above can be attached to the patient’s underwear via any of the attachment options disclosed herein or other suitable means.

[0112] The wearable device may then begin acquiring sensor measurements. The sensor measurements may include (but are not limited to): a signal corresponding to the total concentration of hydrogen, volatile sulfur compounds, and volatile organic compounds present in flatus. Such compounds include but are not limited to hydrogen, ammonia, indole, skatole, methane, carbon dioxide hydrogen sulfide, methyl mercaptan (methanethiol), ethanethiol, dimethyl sulfide, trimethyl sulfide, dimethyl disulfide, dimethyl trisulfide. The device may also contain sensors to measure patient adherence to wearing. For example, temperature sensing, humidity sensing, and / or motion sensing may take place (e.g., via 6 axis accelerometer or 9 axis inertial measurement unit) to determine the times at which a patient is appropriately wearing the smart underwear device. For example, so long as temperature measurements remain within a range consistent with the sensor being attached to patient underwear (e.g., generally similar to surface skin temperatures for the relevant region of the patient’s body), and / or so long as humidity measurements remain within a given range, and / or so long as motion is detected by the motion sensor within a given periodicity, the sensor measurements of gas composition can be considered‘valid’ data for assessment of flatus. In contrast, if temperature suddenly drops (e g., to roughly ambient temperature), and / or motion ceases for a given period of time, data acquired by the gas sensor can be flagged as potentially not relevant or invalid. Similarly, a notification could be sent to the user to put the sensor back on, or a notification could be sent to a healthcare provider indicating that adherence to wearing the sensor is non-optimal.

[0113] At a certain point after the patient begins wearing the device, the patient can be instructed to or caused to consume a food or compound that is selected for the assessment. This time period can be predetermined (e.g., 1 hour), or can be dynamically assessed based on quality of initial measurements. The time period can also be selected at the time the test takes place, based on a healthcare provider’s confidence that the dietary restrictions had been adhered to by the patient.

[0114] In some embodiments, the food or compound may be a carbohydrate of interest, formulated as a beverage, capsule, or contained in a food. The carbohydrate comprises one or a combination of monosaccharides, disaccharides, oligosaccharides, polysaccharides, or polyols (sugar alcohols) including but not limited to: glucose, galactose, fructose, fucose, arabinose, xylose, maltose, sucrose, lactose, fructo-oligosaccharides (FOS), galacto-oligosaccharides (GOS), maltodextrin, inulin, chicory root extract, sorbitol, mannitol, erythritol, maltitol, lactitol, lactulose, and / or xylitol. In other methods, the food or compound of interest may consist essentially of (or comprise) an amino acid, such as a solution containing cysteine, methionine, tryptophan, or any other amino acid, or a mixture of amino acids.

[0115] After the patient has consumed the food or compound of interest, the device can acquire continuous measurements for a measurement period. The measurement period may depend on the type of food / compound consumed, and / or the patient’s age, health and prior history ofmeasurements. In other embodiments, the measurement period may be dynamically determined. For example, a processor associated with the wearable device may determine whether a sufficient quantity of flatus has been measured and can send a notice to the patient or healthcare provider to signify when the test can be completed. In other examples, the measurement period may be dynamically determined by waiting until a flatus emission (or emissions) having certain characteristics has been detected and the continuing for a given period of time thereafter. In some embodiments, data may be acquired for 3 hours from the time the food / compound was consumed, for 3 hours from the time the first qualifying flatus emission is detected, or other durations such as 8, 9, 10, 11, 12, 14, 18, or 24 hours (whether from the time the device is turned on, from the time the food / compound was consumed, or from the time flatus emissions were detected).

[0116] During sensing, the wearable device may transmit sensor data to a separate processor, or may store and process data locally on the device itself or a device physically coupled to the wearable sensing device. Hardware and connection types for transmitting and / or storing the data may be as described above. All data may be acquired with time stamps synced with a clock / time used by the patient and / or healthcare provider so as to more accurately assess time differential from when the food / compound of interest was consumed and / or from emission to emission.

[0117] The acquired data can be processed (whether in real time during the assessment, or by subsequent analysis) to determine certain characteristics of the patient’s flatus emissions during the test. For example, temperature, humidity, and / or motion data may be used to bound the gas sensor data to determine which periods of time represent valid data, e.g., when the user was wearing the sensor. Similarly, other the sensor data can also be tagged or bounded to reflect data acquired before and after the food / compound of interest was consumed, time of day, etc. Thus, data relating to flatus emissions can include gas sensor data (e.g., electrochemical signals), butalso supplemental data and / or annotations such as time data, and various other flags corresponding to the sequence of events and or timing of the steps of the test itself.

[0118] The annotated / tagged gas sensor data can be analyzed to determine the number of emissions, total emission volume, intensity of emissions, gas concentrations of emissions, general / gross emission occurrences and emission characteristics during the measurement period, that will be relevant to determining a likelihood of gut disorders, such as malabsorption of the food / compound consumed. For example, a likelihood of gut malabsorption could be determined if at least two flatus missions are detected within 3 hours after consumption of the food / compound of interest. A flatus emission can be determined from the sensor data by comparison to baseline values: e.g., a 5x, 7x, 8x, lOx, 15x, or other multiple of baseline gas sensor detection values can be flagged as comprising an individual flatus emission. (A baseline value may be determined via a number of approaches, such as described herein). Furthermore, the degree or significance of the gut disorder (e.g., malabsorption) can be assessed by the combination of total number and volume of gut microbial gas production detected as flatus.

[0119] In some cases, if a flatus emission is detected in less than one hour (or other threshold) from consumption of the food / compound of interest, then the test can be flagged as faulty. If no flatus emissions, or a minimal amount of emissions, are reported within a given window (e.g., a 3 hour window after consumption of the food / compound of interest), a negative determination can be generated (e.g., the patient’s gut microbial gas does not reflect a likelihood of malabsorption of the food / compound). If an unusual or inconclusive number, volume, etc. of emissions is detected, the test can be flagged as invalid and / or a repeat test being necessary.

[0120] As another example, the systems and devices described herein can be utilized in a method for assessing the likelihood of SIBO in a patient. As described above in the method for assessing likelihood of malabsorption, a patient wearing the sensor device fasts, then consumes the food / compound of interest in a challenge study (e g., a carbohydrate challenge of consuming a carbohydrate that is a potential trigger food likely to cause malabsorption or other elevated gut microbiome activity to test whether the patient exhibits increased gastrointestinal / flatus emission activity as a result of consumption of the food of interest that is the subject of the challenge). Data measurements are taken in the same fashion as described in the previous method. However, the data is then processed using additional or alternative analyses. A likelihood of SIBO can be determined by time-weighting gas production data. The more flatus detected earlier in the measurement window (e.g., a 3 -hour measurement wind down), the more likely the patient has SIBO.

[0121] In yet further example methods, an initial baseline measurement may be acquired prior to fasting and consumption of the food / compound of interest. For example, in one method for determining a likelihood of IBS-D or SIBO, a patient is instructed to wear the sensing device for a longer time period during which a baseline is established of the patient’s frequency, volume, and timing of gut microbial gas production. In some embodiments, a mobile app, website, pamphlet, instruction booklet, or other means of delivering information to the patient, instructs the patient when to wear the device. The patient may also be instructed to record a food journal with precise times each food was eaten. The patient may even be given a specific dietary regimen and instructed regarding which foods to eat at which times or on which days. The initial baseline recording window may be one day, two days, or any other number of days such as up to seven days. The recordings may take place “continuously,” such as on a constant basis as frequently as possible(measuring at the fastest sample rate at all times), on a constant basis at a given periodicity or sample rate, at certain times of day on a continuous day to day basis, at a periodic rate only during certain hours of day (e.g., waking / sleeping hours), etc. From these measurements, a baseline flatus emission level may be determined by statistical feature extraction such as: the average number of emissions, the total volume or intensity of emissions, median or interquartile emissions, average or median microbiome activity index, etc. per day (or per hour, week, etc.); or more advanced statistical methods such as exponential smoothing trend decomposition, Bayesian estimations, Fourier transformations, etc. Data acquired during this baseline period may be utilized for purposes of comparison to future studies and / or communicated to the healthcare professional(s) monitoring the test, and / or summarization data may be transmitted that interprets the acquired data (e g., number, volume, frequency, composition of flatus emissions, etc.). Such a method can be utilized to diagnose SIBO, IBS-D, inflammation-induced malabsorption, and similar conditions, as well as to estimate gut microbial activity.

[0122] In additional example methods, a sensing device can be utilized in a modified version of the methods described above, for assessing efficacy of a treatment for IBS-D / SIBO via measurements of gut microbial gas production excreted in flatus. Such methods may include obtaining a baseline measurement of gut microbial gas production as set forth above (e.g., for 1-7 days). This may take place prior to or after the treatment regimen. In some cases, the baseline is acquired, then a patient is instructed to thereafter take a prescribed intervention, such as an antibiotic (e.g., rifaximin / xifaxan), or probiotic or prebiotic, or a dietary intervention (e.g., FODMAP avoidance, elimination diet, etc.). Shortly before or upon beginning the intervention, another measurement cycle is begun in which data is acquired for another period (e.g., another 1-7 days) and the patient is given suitable instructions concerning wearing of the device, recordingadherence to the intervention, and / or recording a food journal with times of consumption. In some methods, another baseline measurement cycle can be performed after completion of the intervention period, in order to reassess post-intervention symptoms.

[0123] In further examples, a method can be utilized to detect gut microbial utilization of dietary fiber or fiber supplements via measurements of gut microbial gas production excreted in flatus. Dietary fiber consumption has many benefits, and most people consume far less than the daily recommended amount of dietary fiber. One benefit of fiber is that gut microbes can ferment it into beneficial short-chain fatty acids (SCFAs). Using the systems and techniques described above, a method can be performed to determine whether an individual’s microbiome is capable (and the degree to which it is capable) of fermenting dietary fiber or fiber supplements via measurements of gut microbial gas production excreted in flatus. This determination can be based on the premise that the more gut microbial gas production that is detected via flatus, the better the dietary fiber or fiber supplement can be utilized by the gut microbiota. This method could be administered and monitored by a clinician or be performed directly by a consumer with instructions from a dietary / nutritional company such as via smartphone application.

[0124] In such a method, the participant may be instructed to consume dietary fiber or a dietary fiber supplement and record the time of consumption in the app, journal, website, EMR, etc. The participant should also avoid other fermentable foods (which will be listed in the patient instructions or app, or instructed directly by the clinician / dietitian) or to simply fast given period of time (the fasting or avoidance period may be as described in the methods above). The participant is then instructed to wear a sensing device (such as disclosed herein) for at least 3 hours, but ideally 6-12 hours after consumption of the dietary fiber. In some methods, the participant’s adherence to wearing the sensing device and the sensing device’s operability (e.g., turned on andsending valid data) can be monitored. During the measurement period, the extent of gut microbial gas production excreted in flatus is determined via output of the sensing device. The degree of flatus detected can be used to interpret utilization of the fiber: the more flatus detected, the more the fiber must have been utilized. The degree of flatus detected can be measured through one or more of several indicators: number of individual flatus emissions, duration of detected emissions, magnitude of emissions, and total volume of each / all emissions. These measurements can be derived from the output of the sensing device, and reported to the user or clinician (the derived measurements may be determined by the processor of the sensing device itself or from a separate computing device receiving raw data from the sensing device). The method may then provide an estimated utilization or range of utilization, based on accumulated data from study participants. In further examples, if necessary, the procedure can be repeated with alternative fiber compositions.

[0125] In another example, a method may be performed by which personalized nutrition recommendations can be provided to a user, as a way to reduce IBS-D / SIBO symptoms caused by excessive gut microbial gas production. One way that patients suffering from IB S-D / SIBO attempt to reduce their symptoms is through dietary interventions. However, their approaches are often anecdotal and depend upon the individual’s ability to accurately recall / record how they felt after a period of time during which they were on an interventional diet. The methods and systems provided herein, however, can objectively measure gut microbial gas production and relate it to diet to make evidence-based and personalized recommendations to avoid foods that cause excessive symptoms while maintaining maximum dietary variety.

[0126] A smartphone app or other user interface may provide a way for a user to input the types and quantities of food consumed, and the time at which the food was consumed. In this manner, a timed food journal is kept. In time synchrony, data is collected from a sensing device asdescribed herein, connected to the smartphone. As the smartphone app records data and determines that an unusual, rapid, or excessive amount of flatulence was detected, the app can correlate the episode with the foods that were eaten. Using a database of ingredients for the foods eaten (e.g., FODMAP content of food), the app can begin to correlate diet with gut microbial gas production. From this, the app can generate personalized suggestions of “trigger” foods that cause excessive gut microbial gas production based on data from the smart underwear device. In addition, the app can allow a user to query whether a given food item would be likely, or unlikely, to cause gut distress based on similarity to other foods that were known by the app to cause an elevated gut-microbiome activity level, or onset of other gastrointestinal distress. Similarly, the app can generate personalized suggestions of foods that will not cause symptoms based on data from the smart underwear device.

[0127] In some examples, a user may report their stress level into a software application, to be correlated with gas sensing data, for uses in determining correlation between mental state and gutmicrobiome activity or other gut conditions and disorders. (As used herein, “stress” may generally include mental stress, emotional stress, states of anxiety, etc.). In some embodiments, the user may respond to standardized anxiety or stress questionnaires, such as: a state-trait anxiety inventory (STAI); a visual analog scale for anxiety (VAS-A); a positive and negative affect schedule (PANAS); or a more customized set of questions asking the user to simply rate their stress level on a numeric or word-based scale. From these questions, the software application can generate reported stress level indicators, such as no / low / medium / high / very high stress states. Alternatively, the user could simply input a narrative of how they are feeling, and a software application such as a language model can generate a categorization of the reported stress level(e.g., no, low, medium, high, very high) or directly input their stress level.

[0128] Furthermore, systems and methods contemplated herein may be configured to determine a temporal relationship between elevated stress levels (e.g., high or very high stress indicators) and changes in gut-microbiome activity. For example, in a method for evaluating stress-induced gut disorders (like stress-induced IBS (including IBS-D and IBS-M), stress-induced SIBO, functional dyspepsia, stress-related enteropathy, stress-induced imbalances in gut microbiota, etc.), a user may wear a flatus sensor, such as described herein for a given time period (e.g., several hours, a few days, a week, several weeks, etc.). During that time period, a time series of flatus data is recorded as described above, and the patient is periodically prompted (via a user interface, such as a software application on a mobile device) to report or otherwise give indications of their perceived stress level. In some examples, the prompts may be given on a periodic basis throughout the study period, or may dynamically increase / decrease in frequency corresponding to changes in flatus emission activity. A baseline flatus level or gut-microbiome activity index may be determined, as described above, during periods of time in which the user reports a state of low or no stress. Then, for periods of time in which the user reports elevated stress, a system or method can correlate the gas sensing data shortly before, during, and after those periods of time. The flatus data correlating to these periods of elevated stress can be compared against the baseline, to determine impact of stress on gut activity. This may include quantifying a magnitude of deviation from baseline during periods of elevated reported stress levels, as well as various temporal relationships between onset of elevated stress level and changes in gut-microbiome activity, such as a temporal delay between onset of elevated stress and the subsequent gut-microbiome at and after the onset as well as throughout and after the period of elevated stress persists; as well as a duration of elevated gutmicrobiome activity following stress onset. From these determinations, a gut-brain reactivity profile can be made for use in diagnosing, evaluating, and / or treating gut-brain disorders.Example Outputs, User Interfaces, and Diagnoses

[0129] Various types of outputs and information can be provided to users as a result of use of the foregoing systems and methods. In some embodiments, a user (such as a clinician, dietician, etc.) is provided data concerning flatus activity (e.g., time to first emission, total number of emissions, time of day of each emission, volume / intensity of each emission, and total volume / intensity of emissions during a given time period). The user may also be provided overall statistics or graphs, showing averages and trends of the wearer’s flatus activity, such as the types of ventograms set forth in FIG. 8. (As can be seen in FIG. 8, graphs are depicted showing normalized, observed flatulence frequency and intensity as a function of time, over the course of a week for both a healthy person (left) and a person suspected of having SIBO (right)). In other embodiments, higher order determinations may be provided to a user, such as a likelihood of malabsorption (e.g., expressed a percentage or Boolean with confidence), or a specific diagnosis of a given condition. In such case, a rationale for the given likelihood or diagnosis may also be provided, such as an output of an explainable Al method or an indication that the wearer’s flatus activity exceeded certain thresholds, resulting in a diagnosis or the given likelihood. For example, if a user’s first flatus occurred within the expected timeframe for malabsorption after a “challenge” study consumption of a food of interest, the systems and methods herein can provide a diagnosis to the user, such as “the wearer has malabsorbed “X” grams of the given carbohydrate consumed in the test.” The number and frequency of flatus will also be used to confirm the result. However, the in many cases the inventors have determined that timeframe is more crucial to a malabsorption diagnosis.

[0130] In other methods, a person wears the device continuously and tracks their consumption of all foods the person has eaten in an app. In some embodiments, the person may take pictures of the foods they eat, and a software application will leverage an Al food-recognition model to determine what the food is and its ingredient profile. In other embodiments, the person may answer a questionnaire or input information concerning the amount, type, and / or nutritional information of all foods they eat during a study period. Given that continuous tracking may occur over a long period of time, a person that inputs a substantial majority (e.g., 65%, 70%, 75%, 80%, 85%, 90%, 95%, 100%, etc.) can be considered to have provided information regarding “all” foods eaten. In this approach, the systems and methods described herein will extract information from flatus data using Al to identify which foods are likely causing malabsorption. In this case, the user interface can provide a score that reflects the likelihood of which type of food could be malabsorbed based on the food diary, in addition to the ventogram collected over a long period of time. In further embodiments, these scores may be updated on a continuous and automatic basis, such as every time a 4-hour window has passed after a food item was consumed. As new food items are determined to be malabsorbed, or potentially malabsorbed, an alert can be sent to the user, along with personalized recommendations.

[0131] In yet further embodiments, an app developer may anonymize (pursuant to proper patient consent) data acquired from a use of the sensing device and app and train a deep learning algorithm to predict which types of food items are likely to be malabsorbed, which food items should be avoided by similar individuals with SIBO, IBS-D, or similar diagnoses, and which food items are generally tolerated by similar individuals.Experimental Studies and Prototypes

[0132] To address the need for practical non-invasive tools to measure gut microbial gas production, the inventors developed a “Smart Underwear” device: a wearable sensor system for continuous, autonomous, longitudinal monitoring of flatus composition, frequency, and intensity. The Smart Underwear prototype device comprised a small sensor module that snaps to the exterior of underwear types adjacent to the perineum. It incorporates electrochemical gas sensors to measure gases produced by gut microbes excreted in flatus, along with a temperature sensor and accelerometer to track if the device is being worn. The device allows for novel, non-invasive, realtime measurements of flatus over extended periods during daily activities, providing unprecedented capabilities for monitoring gut microbial gas production.

[0133] Referring now to FIGS. 6A-C, a discussion of various prototype designs will be given. An initial design of the Smart Underwear featured an open-source microcontroller (Ambiq Apollo 3 Artemis) attached to the individual's waistband for a lithium-ion battery. The sensors were positioned in the perineal region, connected to the backpack via a wire threaded through the underwear, as shown in FIG. 6C. While this prototype was functional and had the advantage of a high-capacity rechargeable battery, its large size and the use of a wire were uncomfortable and unreliable. The sensor was affixed to the underwear using double-sided tape, which was not always reliable and required frequent replacement during wash cycles. Subsequent iterations solved all these problems by reducing the device’s size and using custom-made 3D-printed peg for attachment.

[0134] Another design concept featured a round shape, in which a coin cell battery was incorporated into the sensor itself, with the entire device affixed near the perineum, using a peg clasp through the fabric of the underwear, as shown in FIG. 6B. The peg may be a standard size,or may be custom sized / molded to be comfortable to a user (e.g., a mold process, 3D printing, etc.).

[0135] And, yet another prototype utilized the same ‘single device’ design, with peg clasp attachment, and which features a square shape slightly larger than a quarter dollar (e.g., approximately 26 x 29 x 9mm) and is powered by two silver oxide coin cell batteries with enough capacity to run the device for 12 hours (see FIG. 7 exploded views of coin cell battery arrangement designs). Power consumption was measured, showing an average of 173 pA in sleep mode and 918 pA in powered-on mode, respectively (see graph in FIG. 7).

[0136] These prototypes may have one or two main sensing components for gas sensing, along with optional temperature, humidity, and accelerometer sensors for tracking when the device is being worn. In addition, the device connects via Bluetooth to a cellphone for data transmission. Thus, devices contemplated herein may have at least one gas sensor, at least one temperature sensor, at least one humidity sensor, at least one accelerometer, at least one wireless communication module, a processor, and onboard memory.

[0137] In terms of sensing capability, the inventors considered utilizing commercial off-the-shelf electrochemical sensors. However, one drawback of these sensors is their size compared to other types of sensors. After some experimentation, it was determined that this limitation hindered further miniaturization of the device, yet the size would pose negative comfortability and user acceptance. So, the inventors proceeded to development of custom sensors that can provide suitable accuracy while remaining comfortably small.

[0138] In particular, the inventors found that electrochemical sensors were particularly useful, as they can operate in low-power applications while providing high sensitivity and accuracy across awide range of gas concentrations at a relatively low cost. Among various electrochemical techniques, amperometric methods were used, as they presented advantages for use in miniaturized devices because of their simplicity and high sensitivity. In this approach, an open circuit potential (0 bias) is maintained while the current is continuously recorded over time. If hydrogen gas is present, its oxidation at the working electrode generates an electric current, causing a positive peak. The sensor's sensitivity is determined by the capacity of H2 molecules that can be oxidized at the working electrode, catalyzed by the presence of platinum particles in the carbon ink. The electrochemical reaction mechanism can be summarized as follows:O2+ 4e" + 4H+2H2O (1)2H24H++ 4e~ (2)2H2+ O2-> 2H2O (3)

[0139] The four-electron transfer produced a current that was detected and amplified by the potentiostat.

[0140] To form a miniaturized electrochemical sensor, the inventors utilized in some designs a disposable vinyl stencil from which the electrode pattern can be produced using an automatic cutting machine assembled by PCBway®, featuring an immersion gold layer (ENIG) and applied to a prefabricated printed circuit board (PCB), which can serve as both the support structure and the connection interface to the primary device. Bare conductive carbon ink may be employed to create the counter and reference electrodes. The conductive ink (MG Chemicals®) may be screen printed onto the surface of the PCB through the stencil and subsequently dried for approximately 1 minute using a heat gun at 150°C.

[0141] In some examples, over the working electrode area, 2 pL of conductive Vulcan carbon ink(DURA-ink® Pt40), containing 3% platinum and 1% Nafion® w / w, can be deposited and driedfor approximately 1 minute at 150°C. Once fully dried, the heat gun temperature may be increased to 300°C for a few seconds, causing the vinyl stencil to melt and detach from the PCB, leaving behind the desired electrode pattern. Any residual vinyl accumulated at the PCB edges may be manually removed.

[0142] To aid in the electrochemical sensing, in some examples, a solid electrolyte can then be fabricated using a polytetrafluoroethylene membrane, typically used for biological culture sealing, which had been pre-moistened with 6 pL of concentrated (5 mol L1) sulfuric acid (Sigma-Aldrich, 95-98%) and allowed to dry at room temperature for at least 1 hour. The PTFE membrane may be visually inspected to ensure it is completely dry before use. The conductivity of the membrane can also be checked before use, falling within the range of 4-100 kQ cm-1.

[0143] The PTFE membrane (Sigma-Aldrich®) can then be placed over the electrode system, followed by a layer of Teflon tape (PTFE, EverFlow®), which can be wrapped around the PCB with slight pressure to ensure close contact between the membrane and the electrodes. The entire electrode assembly may be encased in a protective 3D-printed PETG (Overture®) housing, ensuring that the case does not come into direct contact with the sensor pad to prevent any interference. The case may be sealed by bonding the top and bottom parts by wetting them with acetone.

[0144] In alternative embodiments, the inventors contemplated that a metal oxide-based sensor may be utilized in place of or in addition to the electrochemical sensor. In such designs, rather than a platinum / Nafion compound deposited to form a working electrode (and bare carbon ink to form the other electrodes), a metal oxide sensing layer may serve as the working gas sensor. For example, a metal oxide sensor may be electrically connected to a receive an applied referenceelectrical signal, with a portion of the sensor (e.g., a semiconductor sensing layer serving as the ‘sensing element’ that is exposed to contact with flatus gas, such as being exposed to hydrogenbased gases of interest). In some examples, the gas sensor may be a metal oxide semiconductor sensor. The composition of the metal oxide sensor may be configured such that it would exhibit a change in electrical resistance when target gas molecules interact with the sensor. In other words, instead of exhibiting an altered electrical characteristic (e.g., current or voltage) as a result of redox reactions at the working electrode as would be the case in an electrochemical sensor, a metal oxide sensor would comprise a metal oxide sensing layer that adsorbs gas molecules which alters the material in a way that affects its resistance.

[0145] Thus, implementing a detection scheme utilizing a metal oxide sensor would entail a modification of how the processor of a sensing device interacts with the gas sensor / electrode. For instance, the processor would apply a current / voltage at a reference level across the metal oxide sensor via a first lead and then detect change in resistance or conductance caused by changes in the sensor. This could be done by, for example, sensing the electrical characteristics of the signal after it has been applied over the metal oxide sensor to derive resistance or conductance, and output a signal indicative of the presence of target gas molecules as a function of resistance or conductance change. Calibration data could be utilized to correlate resistance / conductance change values to concentrations of the gas of interest. By focusing on change values, rather than absolute values, of resistance / conductance, any saturation or long term changes in the metal oxide sensor due to accumulation of gas adsorption can be overcome.

[0146] In prototype devices, the inventors utilized an analog front-end (ATE) integrated circuit (IC) that is compatible applications involving chemical and gas sensing known as the LMP91000 (Texas Instruments). Regardless of sensor types contemplated herein, the IC can be configured todetect, filter, and process gas sensing signal values to assess the content, confirmed occurrences, frequency, and / or intensity of flatus emissions. In the prototypes, the LMP9100 can function as a miniaturized version of a potentiostat by converting and amplifying the small current generated from the chemical reaction into a measurable voltage, which was then translated into a digital signal by the microcontroller's ADC. Consequently, the gain of the trans-impedance amplifier within the LMP91000 directly influenced the sensitivity detected by the microcontroller. The inventors studied the sensor performance at different gain levels, showing that higher amplification resulted in a stronger signal output without a significant positive impact on noise (see FIG. 9, comparison of gain levels). This allows for the detection of hydrogen concentrations in flatus over a wide range, although excessively high concentrations can saturate the signal. The inventors determined that a suitable trade-off design utilized gains of 2 and 6 (corresponding to 3.5 kQ and 35 kQ trans-impedance amplifying resistances) for detecting low and high hydrogen concentrations, respectively.

[0147] In experimental tests, sensor sensitivity depended on several factors, including circuit design, amplification, noise reduction techniques, front-end capability, ADC resolution, temperature, humidity, catalyst concentration, and the surface area of the working electrode. In some experiments utilizing electrolytic sensors, electrolyte volume also played a significant role in gas sensor sensitivity since the gas had to dissolve in the electrolyte before reaching the surface of the working electrode. From this, the inventors determined that electrochemical gas sensors operate better in thin-layer diffusion regimes, as a larger electrolyte volume would slow the response time, increase the ohmic drop, and dilute the analyte. Paradoxically, traditional waterbased electrochemical gas sensors require large reservoirs to prevent evaporation. This illuminates a disadvantage of these types of sensors: their relatively large size needed to accommodate thereservoir compared to other sensor types. This limitation can be partially addressed by using conductive polymers that act as solid electrolytes, that do not evaporate, although these typically still need to be in a wet environment to maintain desirable conductivity. The most common example of such electrolytes is the use of conductive polymers like Nafion, which perform well in high-humidity environments but lose conductivity when completely dried. These polymers are relatively expensive, and the cost is reflected in the price of the sensor.

[0148] Thus, the inventors developed a suitable alterative design, using a concentrated sulfuric acid dispersed in a porous PTFE membrane to act as a pseudo solid electrolyte with a good and minimal evaporation, enabling the fabrication of custom sensors using inexpensive materials and 3D printing technology. This approach allows reduction in the size and cost of devices contemplated herein.

[0149] Electrochemical gas sensors were fabricated using carbon conductive ink for the counter and reference electrodes, and a platinum-Nafion-carbon ink (1% Nafion and 3% Pt) for the working electrode. Platinum (Pt) is a catalyst for hydrogen gas oxidation, a redox reaction that occurs at open circuit potential (0 bias) and is primarily responsible for the sensitivity and selectivity of the electrochemical sensor. Nafion is a polyfluorinated conductive polymer that can function as a solid electrolyte when it has a sufficiently large thickness, as well as a binder to improve the mechanical stability of the carbon ink.43 46. As mentioned before, the inventors employed PTFE filter paper embedded in a strong sulfuric acid solution. The 5 mol L1concentration guarantees a very low evaporation rate due to its low vapor pressure (0.001 mmHg at 25 °C), so even after drying, an invisible layer of acid remained adsorbed to the filter pad. Moreover, sulfuric acid is hygroscopic, which keeps the pad highly conductive. As a result, a quasi-solid electrolyte was formed, maintaining a high conductivity rate of 0.056 ± 0.013 S cm1(n=6) (FIG. xxxx), comparable to the conductivity of Nafion (0.079 - 0.2 S cm ’). The pad maintained high conductivity for at least a month, even when continuously exposed to air.

[0150] The non-conventional use of carbon as a quasi-reference electrode provides for a stable potential, with changes less than 1 mV per day, making them even more stable than standard silver / silver chloride (Ag / AgCl) electrodes. The main reason this electrode is not commonly used is its sensitivity to chloride concentrations, which are ubiquitous in the environment. However, for gas sensing applications where chloride is absent, carbon can be used as a quasi-reference electrode. This decision was also motivated by the fact that silver or silver chloride pseudoreferences cannot be used in a strong acidic environment, as they gradually dissolve in acid. Although this process is slow, it limits the sensor's long-term performance. The use of bare carbon as a reference electrode has limitations. The potential difference between the working electrode, containing Pt particles, and the reference electrode may lead to high transient currents at the beginning of the amperometric measurement, increasing the waiting time for baseline stabilization. To minimize this effect, the amount of Pt deposited on the electrode should be balanced between desired sensitivity and electrode stability. Moreover, if the transient current is relatively large and the amplifier gain in the analog front-end is set too high, it can lead to signal saturation. This issue was addressed by allowing an initial stabilization time of 10 to 20 minutes. A layer of PTFE tape was applied over the electrochemical electrodes and electrolyte pad with slight pressure to ensure full contact over the entire electrode area. This layer provided additional protection, reduced water evaporation, and strengthened the connection between the electrode system and the electrolyte. Finally, sensor connections were completed with a 3D-printed PETG case, sealed with acetone, to protect the sensor (Fig. 10 A-D). Figure 10 E and H show the voltamogram of a custom carbon electrode with ferrocyanide, using a standard Ag / AgCl electrode as a reference, demonstrating thatthe conductivity of the carbon ink is sufficient to perform electrochemical experiments for several reduction and oxidation cycles. A comparison between the Ag / AgCl reference electrode and the carbon quasi-reference electrode showed that when using the carbon quasi -reference electrode, the potential was nearly identical to that observed with the standard Ag / AgCl electrode. This suggests that under these conditions, the carbon reference electrode behaves similarly to a silver chloride electrode. The sensor was evaluated with a benchtop potentiostat, showing a linear range between 0-2000 ppm of H2 (Figs. 10F and 101) and was able to detect real flatus, as shown by the overlapping signals with a device containing both custom and commercial sensors (Fig. 10G). The effect of temperature and humidity was studied, showing a significant baseline deviation above 30°C (Fig. 10 J). This will not pose any problem for the device, as the expected wearing temperature should not exceed body temperature (36°C), unless the device is used in extreme ambient conditions. On the other hand, humidity showed a larger contribution to baseline drift (Fig. 10M). Despite these variations, the sensor still worked consistently across a wide range of humidity, with sensitivity maximized at intermediate values (Fig. 10K). This variation can be explained by the effect on the electrolyte: at lower humidity, water adsorption decreases in the strong acidic filter pad, affecting the electrochemical conductivity of the cell. (Fig. 10L). To address the issues of adjusting the baseline over long periods and the initial stabilization time, the device software was programmed to recalibrate the baseline threshold every 5 minutes.

[0151] Fig. 10A-C show a schematic illustration and image of the custom sensor featuring a pseudo-solid electrolyte fabricated with strong sulfuric acid adsorbed on a PTFE filter pad. The electrolyte is deposited onto a three-electrode system that is screen-printed and attached to a custom PCB for electrical connections. An additional PTFE membrane is applied to minimize water evaporation, and a 3D-printed PETG box is sealed around the electrodes for strongmechanical protection, leaving only a small hole for gas diffusion. Fig. 1 OD is a photo of the actual electrode and a complete sensor image matching the design. FIG. 10E illustrates a cyclic voltammetry of the fabricated electrode with the ferro / ferri cyanide couple compared to a standard Ag / AgCl electrode. Fig. 10H illustrates a signal comparison of the ferro / ferri cyanide couple on the custom carbon electrode versus the standard Ag / AgCl reference electrode and the carbon pseudo-reference electrode, demonstrating signal overlap. Figs. 10F-I illustrates a calibration curve of the custom sensor characterized on a benchtop potentiostat, showing a linear range with increasing additions of EE. Fig. 10G illustrates a comparison between commercial and custom sensors, both attached simultaneously to the same Smart Underwear and tested on a human subject. Fig. 10J illustrates the effects of temperature and (M) relative humidity on custom sensors, illustrating sensor variation with relative humidity and minimal variation with temperature. Fig. 10K illustrates the impact of relative humidity variation on sensor baseline, showing performance with equal concentrations of FE at 1660 ppm at different relative humidity levels measured with a benchtop potentiostat. Fig. 10L illustrates a measurement of FE gas concentration (1660 ppm) with the Smart Underwear over 12 different days, showing small baseline variations attributed to changes in day-to-day relative humidity and temperature.

[0152] In some examples, custom sensors can be characterized using a PSTAT 910 potentiostat (Metrohm®) with commercial NIST-certified gas tanks (Gasco®), which includes 3% hydrogen (EE) in nitrogen (N2), 50 ppm of hydrogen sulfide (EES) in pure (99.99%) nitrogen (N2), and carbon dioxide (CO2), 2.5% methane (CEE) in 20.9% O2, and N2 in air, 50 ppm nitric oxide (NO) in N2, and 25 ppm nitrogen dioxide (NO2) in N2. Sensor characterization and testing may be conducted using the "flatus simulator." Briefly, a mass flow regulator and a microcontroller can be used to release a desired concentration of gases into the measurement chamber where the sensors aretested. This device can be employed for sensor calibration and interference studies. For example, in some examples, the linear range was evaluated between 332-1992 ppm for hydrogen detection, and selectivity was tested against common gases found in flatus, such CH4, CO2, and less concentrated gases such as H2S, NO, and NO2.

[0153] Figs. 1 1A-B illustrate a calibration curve of the inventors’ sensor design against a commercial sensor. Data was collected using a flatus simulator. The graphs showed a linear range with increasing concentrations of hydrogen gas. Fig. 11C illustrates the sensor response time, showing that 90% of the overall signal is reached in ca. 18s while the maximum is reached after 42s. The relaxation time was in the same range and on average lasted ca. 23 s to return to the previous baseline at 22°C and 52% of relative humidity. Fig. 11D illustrates an interference of other gases commonly founded in the flatus. Signal is normalized with respect to hydrogen. The sensor showed a good selectivity towards H2 and only a significant interference was observed with hydrogen sulfide (H2S). However, the concentration of this gas is commonly absent or very low on healthy people and only significant concentrations has been reported for patients that suffer an inflammation in the gut. Figs. 11E-H illustrate the same experiments with the custom sensor for comparison, only a significant interference from Nitric Oxide (NO) was detected, however like H2S this gas was reported in very small concentrations only on patients with severe inflammation.

[0154] Hydrogen is one of the most abundant gases in flatus and thus interference from other gases is expected to be negligible. Nevertheless, selectivity was evaluated, and data was plotted relative to the hydrogen signal. The results indicated that hydrogen sulfide was less sensitive in the custom sensors than in the commercial sensor. No significant levels of carbon dioxide, methane, or nitrogen dioxide were measured in the custom sensors; however, a significant interference from nitric oxide was detected. This gas is typically absent in healthy individuals butcan occur as a consequence of gut inflammation, such as in patients with inflammatory bowel disease (IBD), celiac disease, or have recently undergone chemotherapy. This means that in such patients, the signal from hydrogen could be overestimated, but flatus detection will still be possible.

[0155] The limit of detection (LOD) and limit of quantification (LOQ) of the custom sensors were 58 and 254 ppm, respectively, with a sensitivity of 32 nA ppm '. These values are sufficiently suitable for the desired Smart Underwear application.

[0156] Wearing detection algorithm: Based upon survey results from the inventors’ experiments, it was determined that temperature was the most useful indicator of device adherence / usage; however, the inventors posited that it should not be fully relied upon as the sole indicator of adherence. This can be seen in the arrow in Fig. 12, where the temperature did not increase, but the accelerometer indicated that the person was wearing the device. The accelerometer was configured in low-power mode with a sniff mode feature, allowing it to operate independently from the main microcontroller. If movement is detected, it triggers an interruption, and the state is saved in the non-volatile (NOR Flash) memory within a 5-minute interval. This method was designed to conserve power and optimize battery life.

[0157] Fig. 12A illustrates the temperature and accelerometer outputs. The arrows indicate the gaps in the accelerometer readings that correspond to moments when temperature readings fell below the baseline threshold, arbitrarily selected as room temperature. Fig. 12B illustrates the results from the corresponding wearing algorithm calculation using the combined readings of the accelerometer and the temperature sensor. Fig. 12C illustrates the hand recorder wearing time,illustrating the necessity of the accelerometer (black arrow) for accurately assessing whether the device is being worn.

[0158] The algorithm used in the inventors’ experiments detects whether the device is being worn by combining two sensor inputs: temperature and accelerometer data. The approach assigns a probability score based on the temperature deviation from a baseline and augments it with an accelerometer score, provided that movement is detected over a specific period. The baseline temperature is set at an ambient level and the probability of wearing the device is calculated based on how much the recorded temperature exceeds this baseline. A higher temperature suggests closer contact with the body, indicating that the device is being worn. The probability score ranges from 0 to 90, depending on the temperature.

[0159] To enhance detection accuracy, accelerometer data is incorporated. The accelerometer score is only added if movement is detected within a particular time window: it requires movement to be detected before, during, and after the current time step. If this condition is met, an additional score is added, reflecting the assumption that movement implies the device is in use. The accelerometer contributes 10% to the overall wearing probability. This ensures that temperature remains the primary factor, while movement acts as a secondary indicator.

[0160] The final probability of wearing the device is the sum of the temperature-based probability and the accelerometer score, with the accelerometer score only considered when the movement condition is met. This logic is applied to each data point in the input file and the results are saved in a new output file.

[0161] The Microbiome Activity Index: A high gut microbiome activity is closely correlated with hydrogen gas production, reflected in the increase of both flatus frequency and hydrogenconcentration. It is essential to consider these two variables together. The accumulated gas can be expelled in a few high-intensity events or through a series of smaller flatus. Consequently, relying solely on counting flatus or measuring only the sensor output does not provide a complete picture of gut microbiome activity. Additionally, it was found that high concentrations of hydrogen in the gut can sometimes saturate the sensor. To address this issue, the absolute value of the sensor signal's first derivative was used to offer a more accurate assessment of flatus intensity. By measuring the rate of change rather than the signal output, the baseline contribution was reduced, and flatus was identified more accurately. While this method provides a clearer graphical representation, it may overlook cases where low but prolonged flatus also indicates high microbiome activity. To capture this aspect, a new metric is introduced, defined as the Microbiome Activity Index. Mathematically flatus time and counter can be defined as: Microbiome Activity Index = J O w( |S(t)|), where | S(t) | is the absolute value of the electrochemical sensor first derivative signal, W(x) is a binary function that takes the values of 1 if x > B where B is the dynamic baseline threshold below which flatus are not considered, At; are the discrete time intervals between consecutive data points, and the Microbiome Activity Index.

[0162] The physical meaning of the Microbiome Activity Index is the number of events where the signal exceeded the baseline threshold. By focusing on the number of data points collected rather than the intensity value, a better representation of the overall microbiome activity is provided. This empirical metric allows balance between moments of minor gas release and substantial expulsion events that might otherwise go unnoticed. This approach was tested on 38 participants in the GUMDROP study described below.

[0163] GUMDROP Study: A single-blinded crossover study comparing gumdrop candies containing inulin with sham gumdrops revealed a significant difference in outcomes both withinindividual participants and across the aggregated data from the 38 participants. Inulin was chosen because it is a universally non-absorbed FODMAP, ensuring that everyone should malabsorb it.

[0164] Statistical analysis was conducted using a one-sided Wilcoxon signed-rank test, with the null hypothesis (Ho) stating that the Microbiome Activity Index from the sham gumdrop arm is smaller than that from the inulin gumdrop arm. The results supported the rejection of Ho, with a p- value of 3.6 x 1011and a statistical test value of 5.0, indicating a notable effect of inulin on the Microbiome Activity Index (Fig. 13). Overall, 94.7% of the participants showed higher microbiome activity when consuming the inulin gumdrop, strongly suggesting that the device can be used for diagnosing carbohydrate malabsorption. The remaining 5.3% who did not follow the expected trend are attributed to either poor compliance with a low FODMAP diet or very long transit times that could delay the appearance of flatus within the expected 8 hours of wearing the device. One participant reported mild diarrhea after eating the sham gumdrop, while five participants reported the same symptom after consuming the inulin gumdrops, suggesting that all the flatus events may not have been captured. Surprisingly, 31.6% of participants also reported experiencing bloating, diarrhea, or constipation after consuming the sham gum drops, while 65.8% reported similar feelings after consuming the bag with inulin gummies. These results suggest that inulin gummies clearly change gut microbiome activity and, consequently, the gut perception of symptoms associated with bloating, diarrhea, or constipation. This survey result also highlights the potential unreliability of surveys, as it is highly unlikely that 31.6% of participants truly experienced symptoms after consuming only six regular gumdrops. This value may be attributed to fear and expectation of gastrointestinal discomfort after consuming the gumdrops, which could have led to an overestimation of perceived symptoms following the first bag of sham gumdrops.

[0165] It is worth noting that even with a low FODMAP diet restriction, the device still detects a low but quantifiable Microbiome Activity Index value after consuming the sham gumdrops (represented by the blue bars in Fig. 13 D). This outcome suggests that either the sugar content in the gumdrops exceeds the gut's absorption capacity, or a more restrictive dietary regimen would have been necessary to completely eliminate the availability of fermentable foods that cause flatus. This may prevent false positives, particularly when testing a single carbohydrate without a baseline comparison using the sham gumdrop.

[0166] By setting a Microbiome Activity Index threshold at 256, only 7 out of 38 participants had a Microbiome Activity Index above this threshold after consuming the sham gum drop, whereas 35 out of 38 exceeded the threshold after consuming the inulin gumdrop. This corresponds to an expected specificity of 81.6% and a sensitivity of 91.1%. These results highlight the effectiveness of the threshold, which will be refined and validated in future studies to further enhance diagnostic accuracy.

[0167] Fig. 13 A illustrates an example from a participant in the GUMDROP study showing flatus intensity across the two arms: first with the sham gumdrops (blue line) and then with the inulin gumdrops (orange line). Fig. 13B illustrating the wearing tracker, indicating that the device was effectively worn during both arms (threshold probability > 51%). Fig. 13C illustrates the Microbiome Activity Index calculated for each participant, demonstrating high microbiome activity during the inulin arm in 94.7% of cases (n=38). Fig. 13D is an example schematic representation of the GUMDROP study. Participants avoided high-fiber foods starting two days before and throughout the entire study. Fig. 13E illustrates a Wilcoxon one-tailed signed-rank test showing a p-value of 3.6 x 10 ' 1 and a statistic value of 5.0.

[0168] This work introduces a novel device and method for real-time measurement of gut microbiome activity by detecting hydrogen produced during bacterial fermentation and expelled as flatus. This device has broad potential applications, including the clinical diagnosis of food malabsorption, early detection of gastrointestinal diseases, and monitoring the efficacy of treatments for bloating and excessive flatulence. Notably, it is poised to revolutionize drug discovery in the field of gas and bloating relief by providing a quantitative measure of drug and supplement efficacy, replacing the reliance on subjective surveys that currently dominate research. Additionally, this device offers a compelling alternative to traditional breath tests, which are infrequently performed due to their lack of sensitivity and the inconvenience they pose to patients. In contrast, the device described herein can be comfortably used at home, featuring built-in capabilities for tracking user compliance with necessary dietary preparations. Significant advancements were also made in the fabrication of custom electrochemical gas sensors, resulting in reduced costs and sizes compared to commercial counterparts. In some examples, the malabsorption validation can be expanded to include more types of carbohydrates, such as lactose and fructose.

[0169] Several device prototypes were developed to reduce size, and improve comfort and reliability. In some examples, the device can comfortably snap to the outside of a participant's underwear, adjacent to the rectum in the perineal zone. This placement allows the device to be directly in the stream of flatus, enabling it to passively measure flatus frequency, volume, and composition. The Smart Underwear may be compatible with nearly all types of underwear, by allowing users to choose between four peg sizes adjusted for different underwear thicknesses.

[0170] The device estimates the probability that it is worn and whether it is being worn by combining readings from an accelerometer and a temperature sensor. In some examples, all of theelectronic components can be housed in a custom four-layer PCB (assembled by PCBway). The device may be powered by two silver oxide batteries, commonly used in earbuds, avoiding the use of lithium-ion batteries due to their known hazards 54. The custom battery retainers can be made of brass and cut from a single 0.02-inch thick foil using a metal binder jetting system, with both batteries separated by a custom-made 3D-printed PETG separator.

[0171] One Smart Underwear prototype device utilized a microcontroller (Ambiq Apollo 3 Artemis, Sparkfun®) with low-energy Bluetooth (BLE) capability, which facilitates the connection between the device and the phone application used to retrieve the data. The system also includes a low-power serial flash memory (IS25LP064D, IS SI®), a ferroelectric random-access memory (MB85RC64TA, FRAM, Fujitsu®), and two electrochemical sensors (EC-Sense®) connected to two configurable analog front ends (LMP91000, Texas Instruments®), functioning as small potentiostats. These sensors send analog signals converted by an embedded analog-to- digital converter (ADC) in the Artemis microcontroller. Additionally, a temperature sensor and an accelerometer (MC3630, Memsic®) are used for wearing detection.

[0172] The microcontroller remains in low-power sleep mode most of the time, waking up approximately every five minutes to save the baseline signal or whenever a flatus event causes an interruption to the ADC system via its built-in direct memory access (DMA) feature. The accelerometer can also trigger an interruption every two sleep and power-up cycles to assist in wear detection. The device is powered by two silver oxide batteries (357 / 303 Energizer®). Power consumption was measured using a Power Profiler Kit II (Nordic Semiconductor®).

[0173] Based upon the inventors’ work, including various experiments and designs for SmartUnderwear devices, it can be seen that the systems and methods described herein presentsignificant advantages for clinical applications, including the diagnosis of food malabsorption, early detection of gastrointestinal diseases, and monitoring the efficacy of treatments related to bloating and excessive flatulence. Additionally, such systems and methods are contemplated for use in certain drug discovery processes, such as for treatments targeting gas and bloating by providing objective, quantitative data (a marked improvement over the subjective surveys currently used in research). In some examples, these systems and methods could also replace or supplement traditional breath tests, offering a more comfortable and accurate alternative for patients, enabling long-term home-based monitoring with improved user compliance and comfort. And, the systems and methods provided herein may be utilized to unlock individualized, selfcontrolled food studies which individuals can undertake themselves to ascertain their sensitivities to various foods, ingredients, and components thereof.

Claims

Claims1. A device comprising: a housing having a profile allowing it to be affixed to a user’s underwear in proximity to a perineal region of the user while the underwear is worn; a processor positioned within the housing; a gas sensor positioned within the housing and comprising at least one sensing element, wherein the at least one sensing element has a composition that interacts with at least one gas component of flatus emissions from the user to generate an electrical signal indicative of a presence or a concentration of the gas component; a communication module positioned within the housing and connected to the processor to communicate data from the processor to a remote computing device; and a memory positioned within the housing and having stored thereon a set of instructions which, when executed by the processor, cause the processor to: read the electrical signal generated by the gas sensor; record flatus gas data based upon the electrical signal; associate the flatus gas data with time data corresponding to a time the electrical signal was generated, to generate timed flatus gas data; and output the timed flatus gas data via the communication module.

2. The device of claim 1, further comprising a power source positioned within the housing and connected to provide power to the processor, communication module, and memory throughout a study implemented via the device having a duration of at least twelve hours.

3. The device of claim 2, in which the power source comprises an electrolytic dual layer capacitor (EDLC) connected to provide rechargeable power to the device.

4. The device of claim 1, wherein the gas sensor is positioned in in alignment with an opening in the housing near the user’s perineal region, the opening configured to allow gas penetration to the at least one sensing element.

5. The device of claim 1 , comprising a temperature sensor and a motion sensor, both disposed within the housing and connected to provide signals to the processor; and wherein the set of instructions further cause the processor to: generate an indication of whether the device is being worn by the user based upon the temperature sensor signals and the motion sensor signals.

6. The device of claim 1, wherein the gas sensor is an electrochemical sensor comprising a reference electrode and a counter electrode, and wherein the at least one sensing element is a working electrode comprising a material that exhibits a change in electrical current due to redox reactions when exposed to hydrogen-based gases.

7. The device of claim 6, wherein the working electrode is a platinum-based electrode, and the reference electrode and counter electrode comprise conductive carbon.

8. The device of claim 1 wherein the gas sensor is a metal oxide semiconductor sensor and the at least one sensing element comprises a metal oxide sensing layer of the metal oxide semiconductor sensor, wherein the metal oxide semiconductor sensor exhibits a change in electrical resistance in response to adsorption of hydrogen-based gases at the metal oxide sensing layer.

9. The device of claim 6 further comprising a quasi-solid electrolyte disposed in contact with the working electrode, reference electrode, and counter electrode.

10. The device of claim 9 wherein the quasi-solid electrolyte comprises an absorbent substrate impregnated with a hygroscopic substance.

11. A system for determining a likelihood of a gut condition, the system comprising: a processor; a communication module; and a memory in communication with the processor and storing software instructions that cause the processor to:receive gas sensing data from a gas sensor, the gas sensing data comprising a time series of measurements indicative of concentration of a gas in a subject’s flatus emissions during a given time period; receive adherence data indicative of whether the subject was wearing the gas sensor when measurements of the time series were acquired; receive additional user data, indicative of a time when consumption of a food of interest occurred relative to the time series of measurements; determine at least one of: a timing of a first occurrence of a flatus emission after the food consumption of interest occurred, based on the gas sensing data; a microbiome activity index; or a total number of flatus emissions during the given time period after the food consumption of interest occurred; and output to a user an indicator of an effect of the food consumption on the subject.

12. The system of claim 11, wherein the software instructions further causes the processor to, in association with receiving the additional user data, receive a time at which the subject consumed a food comprising a target carbohydrate or amino acid after having fasted.

13. The system of claim 11, wherein the software instructions further causes the processor to output to a user an indicator of a likelihood the subject has a gut condition comprising carbohydrate malabsorption.

14. The system of claim 11 wherein the given time period is at least three hours.

15. The system of claim 11 wherein: the given time period is at least one week; the time series of measurements provide flatus monitoring throughout the user’s daily activities; andthe microbiome activity index is calculated over the given time period relative to all foods eaten by the user.

16. The system of claim 15 wherein the additional user data is indicative of all foods eaten by the user during the given time period, and wherein the software instructions further cause the processor to identify a trigger food that, when eaten by the user, results in an increase in the microbiome activity index over a baseline microbiome activity index of the user by a threshold amount.

17. The system of claim 16 wherein the additional user data indicative of all foods eaten by the user is determined via a user interface allowing the user to upload photos of the foods eaten for processing by a food-recognition Al model, or enter nutritional information for the foods eaten; and wherein the software instructions further cause the processor to determine ingredients of the foods eaten that may cause increased gastrointestinal activity.

18. The system of claim 11 wherein the adherence data is derived from at least one of an output of a temperature sensor disposed within a housing with the gas sensor or an output of a motion sensor disposed within the housing with the gas sensor.

19. The system of claim 11 further comprising a display screen of a mobile device, and wherein the software instructions further cause the processor to: prompt the user via the display screen to ensure the gas sensor is being worn at a location proximate to a perineal region; prompt the user to commence a dietary regimen; and prompt the user to consume the food of interest at a given time after commencement of the dietary regimen.

20. The system of claim 11 wherein the software instructions further cause the processor to monitor for peaks in a signal value of the gas sensing data indicative of a flatus emission, and to determine a confirmed flatus emission occurred when the signal value increases from a baselinelevel to a peak value exceeding a threshold value for nor more than a given period of time, then drops back to the baseline level.

21. The system of claim 11 wherein the microbiome activity index is indicative of a number of events in which the gas sensing data exceeded a threshold, and is calculated using an absolute value of a first derivative of the time series of measurements.

22. A method for diagnosing a gut disorder comprising: causing a subject to fast for an initial fasting period; providing a wearable flatus sensor to be worn by the subject; administering a dietary intervention to be consumed by the subject, and recording a time the dietary intervention was consumed; continuously recording flatus data from the wearable flatus sensor throughout a measurement period, the measurement period comprising at least one hour; determining flatus activity based on the flatus data, the flatus activity including at least one of: time elapsed from consumption of the dietary intervention to first flatus emission, total number of flatus emissions, frequency of flatus emissions, intensity of flatus emissions, time distribution of flatus emissions, flatus emission volume, and total volume of all flatus emissions; and based on the flatus activity and time of consumption of the dietary intervention, providing a diagnosis of the gut disorder.

23. The method of claim 22, wherein the dietary intervention is a carbohydrate, and the gut disorder is malabsorption of the carbohydrate.

24. The method of claim 22, wherein the gut disorder is at least one of small intestine bacterial overgrowth (SIBO) or irritable bowel syndrome diarrhea dominant (IBS-D), and the diagnosis is made by providing the flatus activity and information concerning the dietary intervention to a trained machine learning algorithm configured to determine a likelihood of SIBO or IBS-D based on training data of past patients having SIBO or IBS-D and past patients that did not have SIBO or IBS-D.

25. The method of claim 22 wherein the dietary intervention comprises a carbohydrate challenge utilizing at least one of glucose or fructose, and the gut disorder is secondary malabsorption in individuals with at least one of: recent chemotherapy, Celiac disease, or inflammatory bowel disease.

26. The method of claim 22 wherein the wearable flatus sensor comprises the device of any of claims 1-10.

27. The system of claim 11 wherein the gas sensor is the device of any of claims 1-10.

28. The system of claim 19 wherein the processor, communication module, and memory are part of the mobile device, the gas sensor is the device of any of claims 1-10, and the system allows a subject to assess food impact outside of, and without involvement of, a healthcare clinic.

29. A method for evaluating a gut disorder, comprising: providing a wearable flatus sensor to be worn by a subject during a study period; recording flatus data from the wearable flatus sensor during the study period; receiving an indication of a reported stress level of the subject during the study period; establishing a baseline gut-microbiome activity index by analyzing flatus data during periods of low or no reported stress level; measuring stress-induced changes in gut-microbiome activity by comparing flatus data during periods of elevated reported stress level against the baseline gut-microbiome activity index; quantifying a magnitude of deviation from baseline during periods of elevated reported stress level; determining temporal relationships between onset of elevated stress level and changes in gut-microbiome activity, including a temporal delay between stress onset and gut-microbiome response and a duration of elevated gut-microbiome activity following stress onset; andgenerating a gut-brain reactivity profile based on: the magnitude of gutmicrobiome activity changes during periods of elevated reported stress level; the temporal delay between stress onset and gut-microbiome response and the duration of elevated gut-microbiome activity following stress events.

30. The method of claim 29 wherein the wearable flatus sensor comprises the device of any of claims 1-10.

Images