System and method for obtaining desired oxygen dosing based on erythropoiesis model

Abstract

A computing system for determining a system training policy for a user is provided. The computing system includes a user device that uses one or more sensors to obtain an oxygen partial pressure (PO2) level of a user over a period of time. The user device provides previous PO2 levels to a personalized erythropoiesis model generation computing platform. The computing platform obtains individualized user data for the user, the individualized user data indicating one or more previous hematocrit and / or hemoglobin measurements for the user. The computing platform determines an individualized erythropoiesis model for the user based on the one or more previous hematocrit and / or hemoglobin measurements and the previous PO2 information, and employs the individualized erythropoiesis model to determine predicted hematocrit and / or hemoglobin measurements. The computing platform performs one or more actions based on the one or more predicted hematocrit and / or hemoglobin measurements.

Description

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This patent application claims priority to U.S. patent application No. 17 / 960,560, filed on October 5, 2022, the contents of which are incorporated herein by reference. Technical Field

[0003] (none) Background Art

[0004] Red blood cells (erythrocytes) are essential for the transport of oxygen in the body. Understanding the regulation of erythropoiesis (called erythropoiesis) is very important for treating patients and other users such as athletes in various clinical situations. For example, athletes may seek to reach a desired hemoglobin level and / or oxygen rationing level before a specific event such as a sports game. By reaching a desired hemoglobin level, athletes can exert their peak potential. In other examples, patients planning surgery will seek to reach a certain hemoglobin level before surgery to reduce the risk of surgery. In this way, users can take precautions and / or perform actions to reach the desired hemoglobin level, such as training at high altitudes, in a hypoxic room, and / or take other measures. However, it is difficult for users to accurately predict the hemoglobin level and / or oxygen rationing level they expect at a specific time in the future (for example, at a specific event such as a planned surgery or a sports event). Summary of the invention

[0005] This Summary is provided to introduce certain exemplary embodiments that are further described below. This Summary is not intended to identify key features or essential features of the present disclosure.

[0006] In one embodiment, a computing system is provided. The computing system includes a user device, comprising: one or more sensors configured to obtain oxygen partial pressure (PO2) levels; one or more first processors; and a first non-transitory computer-readable medium having first processor-executable instructions stored thereon, wherein the first processor-executable instructions, when executed by the one or more first processors, facilitate: obtaining the PO2 levels over a period of time using the one or more sensors; generating previous PO2 information for a user based on the PO2 levels obtained using the one or more sensors; and providing the previous PO2 information to a personalized erythropoiesis model generation computing platform. The computing system includes the personalized erythropoiesis model generation computing platform, which includes: one or more second processors; and a second non-transitory computer-readable medium having second processor-executable instructions stored thereon, wherein the second processor-executable instructions, when executed by the one or more second processors, facilitate: obtaining the previous PO2 information from the user device; obtaining personalized user data for the user, wherein the personalized user data indicates one or more previous hematocrit and / or hemoglobin measurements for the user; determining a personalized erythropoiesis model for the user based on the one or more previous hematocrit and / or hemoglobin measurements and the previous PO2 information, wherein the personalized erythropoiesis model indicates a personalized model parameter set for the user; using the personalized erythropoiesis model to determine one or more predicted hematocrit and / or hemoglobin measurements at one or more future times; and performing one or more actions based on the one or more predicted hematocrit and / or hemoglobin measurements.

[0007] In some cases, determining the personalized erythropoiesis model for the user includes inputting the previous PO2 information and the one or more previous hematocrit and / or hemoglobin measurements into a mathematical model for oxygen rationing to determine the personalized erythropoiesis model.

[0008] In some examples, the previous PO2 information indicates one or more hypoxic events experienced by the user during a previous time period, wherein each of the one or more hypoxic events indicates a PO2 level that is below a normoxic level.

[0009] In some embodiments, employing the personalized erythropoiesis model to determine the one or more predicted hematocrit and / or hemoglobin measurements includes: generating a plurality of future PO2 information, wherein each of the plurality of future PO2 information indicates one or more hypoxic events to be performed for the user in a future time period; inputting the plurality of future PO2 information into the personalized erythropoiesis model to determine a plurality of predicted hematocrit and / or hemoglobin measurements; and determining optimal future PO2 information from the plurality of future PO2 information based on comparing the plurality of predicted hematocrit and / or hemoglobin measurements with expected hematocrit and / or hemoglobin measurements associated with the user.

[0010] In some cases, the second processor executable instructions, when executed by the one or more second processors, further facilitate: determining a system training strategy for the user based on the optimal future PO2 information, and wherein performing the one or more actions includes displaying the system training strategy on a display device associated with the personalized erythropoiesis model generation computing platform.

[0011] In some examples, the first processor executable instructions, when executed by the one or more first processors, further facilitate: generating additional PO2 information for a user based on the PO2 level obtained using the one or more sensors; and providing the additional PO2 information to the personalized erythropoiesis model generation computing platform.

[0012] In some variations, wherein the second processor executable instructions, when executed by the one or more second processors, further facilitate: determining whether to re-adopt the personalized erythropoiesis model after adopting the personalized erythropoiesis model; and re-adopting the personalized erythropoiesis model using the additional PO2 information from the user device.

[0013] In some cases, the first processor executable instructions, when executed by the one or more first processors, further facilitate: obtaining a mathematical model for oxygen rationing; obtaining group user data associated with a plurality of users; and generating a plurality of virtual user avatars based on the mathematical model and the group user data, wherein determining the individualized erythropoiesis model for the user is further based on the plurality of virtual user avatars.

[0014] In some examples, determining an individualized erythropoiesis model for a user is based on the following equation: in, Indicates the rate of change of hypoxia-inducible factor (HIF) signaling activity, β h(θ) represents the formation rate of HIF complex, indicates that HIF signaling is upregulated in response to hypoxia, which reduces blood hemoglobin concentration, k h represents the decay rate of the HIF signal, represents the blood hemoglobin concentration measured by the kidneys, represents the physiological set point of oxygenated hemoglobin concentration in the blood, and h represents the HIF signal.

[0015] In some variations, determining the personalized erythropoiesis model for a user is based on the following equation: in, represents the rate of change of EPO concentration, β e (θ) represents the basal erythropoietin (EPO) synthesis under steady-state conditions, represents additional EPO synthesis due to HIF signaling activation, and k e e represents the decay rate k e EPO decay.

[0016] In some cases, determining the personalized erythropoiesis model for a user is based on the following equation: in, Indicates the rate of change of the total amount of red blood cells. represents the differentiation flux from the precursor cell population to the erythroid population, and K ery represents the cell flux due to apoptosis.

[0017] In another embodiment, a method is provided. The method includes obtaining personalized user data for a user, wherein the personalized user data indicates one or more previous hematocrit and / or hemoglobin measurements for the user and previous oxygen partial pressure (PO2) information for the user; determining a personalized erythropoiesis model for the user based on the one or more previous hematocrit and / or hemoglobin measurements and the previous PO2 information, wherein the personalized erythropoiesis model indicates a personalized model parameter set for the user; using the personalized erythropoiesis model to determine one or more predicted hematocrit and / or hemoglobin measurements at one or more future times; and performing one or more actions based on the one or more predicted hematocrit and / or hemoglobin measurements.

[0018] In some cases, determining the personalized erythropoiesis model for the user includes inputting the previous PO2 information and the one or more previous hematocrit and / or hemoglobin measurements into a mathematical model for oxygen rationing to determine the personalized erythropoiesis model.

[0019] In some examples, the previous PO2 information indicates one or more hypoxic events experienced by the user during a previous time period, wherein each of the one or more hypoxic events indicates a PO2 level that is below a normoxic level.

[0020] In some embodiments, employing the personalized erythropoiesis model to determine the one or more predicted hematocrit and / or hemoglobin measurements includes: generating a plurality of future PO2 information, wherein each of the plurality of future PO2 information indicates one or more hypoxic events to be performed for the user in a future time period; inputting the plurality of future PO2 information into the personalized erythropoiesis model to determine a plurality of predicted hematocrit and / or hemoglobin measurements; and determining optimal future PO2 information from the plurality of future PO2 information based on comparing the plurality of predicted hematocrit and / or hemoglobin measurements with expected hematocrit and / or hemoglobin measurements associated with the user.

[0021] In some cases, the method further includes determining a system training strategy for the user based on the optimal future PO2 information, and wherein performing the one or more actions includes displaying the system training strategy on a display device associated with the personalized erythropoiesis model generation computing platform.

[0022] In some examples, determining the personalized erythropoiesis model for a user is based on the following equation: in, Indicates the rate of change of hypoxia-inducible factor (HIF) signaling activity, β h (θ) represents the formation rate of HIF complex, indicates that HIF signaling is upregulated in response to hypoxia, which reduces blood hemoglobin concentration, k h represents the decay rate of the HIF signal, represents the blood hemoglobin concentration measured by the kidneys, represents the physiological set point of oxygenated hemoglobin concentration in the blood, and h represents the HIF signal.

[0023] In some variations, determining the personalized erythropoiesis model for a user is based on the following equation: in, represents the rate of change of EPO concentration, β e (θ) represents the basal erythropoietin (EPO) synthesis under steady-state conditions, represents additional EPO synthesis due to HIF signaling activation, and k e e represents the decay rate k e EPO decay.

[0024] In some cases, determining the personalized erythropoiesis model for a user is based on the following equation: in, Indicates the rate of change of the total amount of red blood cells. represents the differentiation flux from the precursor cell population to the erythroid population, and K ery represents the cell flux due to apoptosis.

[0025] In yet another embodiment, a personalized erythropoiesis model generation computing platform is provided. The computing platform includes: one or more processors; and a non-transitory computer-readable medium having processor-executable instructions stored thereon, wherein the processor-executable instructions, when executed by the one or more processors, facilitate: obtaining personalized user data for a user, wherein the personalized user data indicates one or more previous hematocrit and / or hemoglobin measurements for the user and previous oxygen partial pressure (PO2) information for the user; determining a personalized erythropoiesis model for the user based on the one or more previous hematocrit and / or hemoglobin measurements and the previous PO2 information, wherein the personalized erythropoiesis model indicates a personalized model parameter set for the user; using the personalized erythropoiesis model to determine one or more predicted hematocrit and / or hemoglobin measurements at one or more future times; and performing one or more actions based on the one or more predicted hematocrit and / or hemoglobin measurements.

[0026] Further features and aspects are described in more detail below with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1A is a simplified block diagram depicting an exemplary computing environment according to one or more examples of the present application.

[0028] Figure 1B is a graphical representation of partial pressure of oxygen (PO2) information according to one or more examples of the present application.

[0029] Figure 2 According to one or more examples of this application Figure 1A A simplified block diagram of one or more devices or systems within an exemplary environment.

[0030] Figure 3 is a flow chart of an exemplary process for monitoring a user's hematocrit and / or hemoglobin concentration using a personalized erythropoiesis model according to one or more examples of the present application.

[0031] Figure 4 is a flow chart of an exemplary process for generating an erythropoiesis model according to one or more examples of the present application.

[0032] Figure 5A is a block diagram of an exemplary mathematical erythropoiesis model according to one or more examples of the present application.

[0033] Figure 5B One or more examples of the present application are described. Figure 5A A block diagram of the blocks within an exemplary mathematical model.

[0034] Fig. 6A Figure 2 from a previous study of Chilean soldiers shows a schedule of intermittent hypoxic exposure.

[0035] Figure 6B-6K A graphical representation of an erythropoiesis model generated for use according to one or more examples of the present application is shown. DETAILED DESCRIPTION

[0036] Exemplary embodiments of the present application provide generating an erythropoiesis model for one or more users and using the erythropoiesis model to adjust the hematocrit and / or hemoglobin concentration of the user. Modeling erythropoiesis using a selected erythropoiesis stimulating agent (ESA) administration regimen (including using iron homeostasis), using the initially selected ESA administration regimen to predict the hematocrit and / or hemoglobin (HGB) concentration of the patient, using the model to determine one or more different ESA administration regimens so that the model predicts the hematocrit and / or hemoglobin concentration of the patient within a desired range, and using the determined ESA administration regimen to administer ESA to the patient. Examples are described in U.S. Patent Nos. 10,319,478, entitled "SYSTEM AND METHOD OF MODELING ERYTHROPOIESIS AND ITS MANAGEMENT", and U.S. Patent No. 9,679,111, entitled "SYSTEM AND METHOD OF MODELING ERYTHROPOIESIS INCLUDING IRON HOMEOSTATSIS", the contents of which are incorporated herein by reference in their entirety. U.S. patent application No. 17 / 960,305, filed on October 5, 2022, entitled “SYSTEM AND METHOD FOR ADJUSTING HYPOXIA-INDUCIBLE FACTOR STABILIZERTREATMENT BASED ON ANEMIA MODELING,” is also incorporated herein by reference in its entirety.

[0037] As will be explained in further detail below, the present disclosure describes systems, devices, and methods for generating and using erythropoietic models. For example, the new erythropoietic model is capable of predicting an individual's HGB dynamics as a function of EPO levels and other user-specific characteristics. These user-specific characteristics can be identified after a system perturbation (e.g., a hypoxic period) is performed on the user. User exposure to lower air pressures (e.g., at high altitudes) results in lower inspiratory O2 partial pressure, PO2. Similarly, user exposure to an atmosphere with reduced ambient fractional oxygen (O2) concentrations also results in lower inspiratory PO2. This lower inspiratory PO2 results in hypoxemia, which is a powerful physiological stimulus for EPO synthesis. Measuring HGB and EPO levels before, during, and after PO2 perturbations allows identification of key biological parameters of the new erythropoietic model, thereby generating a user-specific avatar. The avatar can be used to predict the user's temporal and quantitative HGB response to intermittent PO2 reductions.

[0038] Among other advantages, the present disclosure uses an erythropoiesis model to accurately predict the expected hemoglobin and / or oxygen rationing levels of future times. For example, because each user is different, the erythropoiesis model is used to better predict the response of a specific individual to certain events and how those events will affect the individual's hemoglobin and / or oxygen rationing levels, which in turn can more accurately predict or determine the hemoglobin and / or oxygen rationing levels of future moments. For example, hemoglobin is the primary means of transporting oxygen from the lungs to peripheral organs and tissues. Although low HGB levels can cause fatigue and physical weakness, higher than normal HGB levels (to a certain extent) can improve physical performance and endurance. These features are desirable in some users, particularly those users who are expected to benefit from a higher hematocrit at a given point in time (e.g., athletes before sporting events, special forces, and / or high endurance sports). Additionally, these features are desirable in situations where significant blood loss is anticipated (e.g., future major surgery), in situations where blood donation is preferred (e.g., in preparation for surgery where high blood loss is anticipated), in users who cannot be administered ESAs due to pre-existing conditions (e.g., vascular access obstruction, history of seizures, allergic reaction to ESAs, hypertension and / or hypercoagulopathy), and / or in users who are unwilling to receive such medications and instead prefer to increase red blood cell mass by physiological rather than pharmacological means.

[0039] Figure 1A1 is a simplified block diagram depicting an exemplary computing environment (e.g., a computing system) according to one or more examples of the present application. Environment 100 includes an external measurement system 104, a network 106, a personalized erythropoiesis model generation computing system 108 (e.g., a personalized erythropoiesis model generation computing platform), and a user device 110 (e.g., a mobile device and / or a wearable device). Although the entities within environment 100 may be described below and / or depicted in the figures as a single entity, it should be understood that the entities and functions discussed herein may be implemented by and / or include one or more entities.

[0040] Entities within the environment 100, such as the external measurement system 104, the personalized erythropoiesis model generation computing system 108, and the user device 110, can communicate with other systems within the environment 100 via a network 106. The network 106 can be a global area network (GAN), a wide area network (WAN), a local area network (LAN), or any other type of network or network combination such as the Internet. The network 106 can provide a combination of wired, wireless, or wired and wireless communications between entities within the environment 100. Additionally and / or alternatively, one or more entities within the environment 100 can communicate with each other without using the network 106. For example, the external measurement computing device 104 and the user device 110 can communicate with each other via one or more wireless protocols (e.g., WI-FI) and / or wired connections.

[0041] The user device 110 may be, for example, a smart phone and / or a wearable device owned and / or operated by a user. The user may use the user device 110 to provide information to other entities of the environment 100, such as the personalized erythropoiesis model generation computing system 108, and the user device 110 may receive information from other entities. For example, the user device 110 may receive user input from the user, such as the user's hemoglobin or hematocrit concentration; event information, such as hypoxic events (e.g., high-intensity training events, high-altitude events, hypoxic chamber events, high-endurance events, and / or other events in preparation for performance events) and performance events (e.g., future surgery, athletic competitions, blood donations, other medical events, and / or other types of performance events); and / or future times for event information (e.g., the date on which the performance event is scheduled to occur). The hypoxic event may include and / or indicate PO2 information of the user. Depending on the circumstances, the PO2 information may refer to the environment and / or inspiration. For example, for high altitude training, ambient PO2 may be appropriate, whereas inspired PO2 may be more appropriate when ambient PO2 is normal and the low PO2 event is the result of using a high altitude training mask. Figure 1B is a graphical representation of reduced oxygen partial pressure (PO2) information according to one or more examples of the present application. In particular, Figure 1BA graphical representation 150 is shown of a time series of a user's PO2 levels (e.g., the user's PO2 levels over a set time period). For example, the graphical representation 150 includes a normoxic level 152 (e.g., the user's normal oxygen level) and an observed PO2 level 154 for the user. In two different instances, the user's observed PO2 level 154 drops below the normoxic level 152. Each of these two instances may indicate a hypoxic event (e.g., "Hypoxic Event #1" and "Hypoxic Event #2"). As such, the PO2 information may indicate one or more hypoxic events over a period of time, such as the example shown in the graphical representation 150. In some variations, the user may provide user input indicating a hypoxic event (e.g., PO2 levels over a period of time), and the user device 110 may generate the PO2 information based on the user input.

[0042] Additionally and / or alternatively, in some examples, the user device 110 may include one or more sensors and / or other devices configured to obtain hypoxic events and / or generate PO2 information. For example, the user device 110 may be a wearable device including one or more sensors configured to obtain oxygen saturation and / or partial pressure of the user's surrounding environment. Based on the obtained sensor data, the user device 110 may generate PO2 information. For example, the user device 110 may be a wearable device and / or a mobile device worn by the user. The user device 110 may continuously measure the PO2 level of the environment around the user and generate PO2 information based on the continuous measurement values.

[0043] Additionally and / or alternatively, the user device 110 may receive information from other entities of the environment 100, such as from the personalized erythropoiesis model generation computing system 108. For example, the personalized erythropoiesis model generation computing system 108 may generate oxygen rationing information and provide it to the user device 110. In some cases, the oxygen rationing information may indicate a measured hemoglobin or hematocrit concentration and / or a predicted hemoglobin or hematocrit concentration of the user at a future time (e.g., when the performance event is scheduled to occur). Additionally and / or alternatively, the oxygen rationing information may indicate one or more hypoxic events to be performed by the user in order to obtain a desired hemoglobin or hematocrit level at the future time when the performance event is scheduled to occur.

[0044] In some variations, the oxygen rationing information may include an erythropoiesis model for the user. For example, the oxygen rationing information may include a generic erythropoiesis model (e.g., an erythropoiesis model that is not specific to an individual). The user device 110 may be configured to individualize the erythropoiesis model for the user using individualized model parameters specific to the user before using the individualized erythropoiesis model to predict a hemoglobin or hematocrit level for the user. Additionally and / or alternatively, the oxygen rationing information may include an individualized erythropoiesis model having individualized model parameters specific to the user. Generating a generic or individualized erythropoiesis model for a user and using the erythropoiesis model to predict a hematocrit and / or hemoglobin level for the user will be described in further detail below.

[0045] The user device 110 may be and / or include, but is not limited to, a desktop, a laptop, a tablet, a mobile device (e.g., a smartphone device, or other mobile device), a smart watch, an Internet of Things (IoT) device, a wearable device, or any other type of computing device, which typically includes one or more communication components, one or more processing components, and one or more memory components. The user device 110 may be capable of executing software applications managed by, communicating with, and / or otherwise associated with an enterprise organization.

[0046] The personalized erythropoiesis model generation computing system 108 is a computing system that generates one or more personalized erythropoiesis models. For example, the personalized erythropoiesis model generation computing system 108 includes one or more computing devices, computing platforms, systems, servers, and / or other devices capable of performing functions and / or actions such as generating one or more personalized erythropoiesis models for a user.

[0047] In some cases, the personalized erythropoiesis model generation computing system 108 may communicate, for example, with the external measurement system 104 and / or the user device 110. For example, the personalized erythropoiesis model generation computing system 108 may provide one or more generated personalized erythropoiesis models to the user device 110. In some examples, the personalized erythropoiesis model generation computing system 108 may include a display device configured to display information associated with the erythropoiesis model.

[0048] The personalized erythropoiesis model generation computing system 108 can be implemented using one or more computing platforms, devices, servers, and / or equipment. In some examples, the computing system 108 may include and / or be connected to a display device configured to display information. In some variations, the personalized erythropoiesis model generation computing system 108 may be implemented as an engine, a software function, and / or an application. In other words, the functionality of the personalized erythropoiesis model generation computing system 108 may be implemented as software instructions stored in a storage device (e.g., a memory) and executed by one or more processors.

[0049] The external measurement system 104 can be and / or include, but is not limited to, one or more desktops, laptops, tablets, mobile devices (e.g., smartphone devices, or other mobile devices), processors, controllers, or any other type of computing device, which typically includes one or more communication components, one or more processing components, and one or more memory components.

[0050] The external measurement system 104 can be configured to measure and / or obtain measurements associated with the user. For example, the external measurement system 104 can be a laboratory system configured to measure the user's hemoglobin or hematocrit level and / or erythropoietin level based on a blood sample of the user. After obtaining the measured hemoglobin or hematocrit level and / or erythropoietin level, the external measurement system 104 can be configured to provide the measured hemoglobin or hematocrit level to other entities within the environment 100.

[0051] In some cases, a single entity may perform the functions of the user device 110 and the personalized erythropoiesis model generating computing system 108. For example, the personalized erythropoiesis computing system 108 may generate a personalized erythropoiesis model and use the personalized erythropoiesis model to predict the user's hematocrit and / or hemoglobin level at a future time.

[0052] It should be understood that Figure 1A The exemplary environment depicted in is merely an example, and the principles discussed herein may also be applicable to other situations—including, for example, other types of institutions, organizations, devices, systems, and network configurations.

[0053] Figure 2 According to one or more examples of this application Figure 1A A simplified block diagram of one or more devices or systems within an exemplary environment. For example, device / system 200 may be Figure 1AThe personalized erythropoiesis computing system 108 and / or user device 110 of the present invention. The device / system 200 includes a processor 204, such as a central processing unit (CPU), a controller and / or logic, which executes computer executable instructions for performing the functions, processes and / or methods described herein. In some examples, the computer executable instructions are stored locally and accessed from a non-transitory computer readable medium such as a storage device 210, which can be a hard drive or a flash drive. The read-only memory (ROM: Read Only Memory) 206 includes computer executable instructions for initializing the processor 204, and the random access memory (RAM: Random-Access Memory) 208 is the main memory for loading and processing instructions executed by the processor 204. The network interface 212 can be connected to a wired network or a cellular network and a local area network or a wide area network, such as the network 106. The device / system 200 may also include a bus 202 connecting the processor 204, the ROM 206, the RAM 208, the storage device 210 and / or the network interface 212. Components within the device / system 200 may communicate with each other using the bus 202. The components within the device / system 200 are merely exemplary and may not include every component, server, device, computing platform, and / or computing device within the device / system 200. Additionally and / or alternatively, the device / system 200 may also include components that may not be included within every entity of the environment 100.

[0054] Figure 3 is a flow chart of an exemplary process 300 for monitoring a user's hematocrit and / or hemoglobin concentration using an erythropoiesis model according to one or more examples of the present application. The process may be performed by, for example, Figure 1A The process 300 may be performed by a user device 110 and / or a computing device of the personalized erythropoiesis model generation computing system 108 depicted in FIG. However, it will be appreciated that any of the following blocks may be performed in any suitable order and that the process 300 may be performed in any suitable environment and by any suitable device or system. Figure 3 The descriptions, illustrations, and processes of are merely exemplary, and process 300 may use other descriptions, illustrations, and processes to monitor a user's hematocrit and / or hemoglobin concentration.

[0055] For example, in some variations, a computing device (e.g., computing system 108 and / or user device 110) may obtain individualized user data for a user (e.g., a particular user's previous hypoxic events, previous hematocrit / hemoglobin measurements, the user's previous erythropoietin level measurements, gender, weight, the altitude at which the user typically resides, and / or other characteristics of the user). The computing device may determine a user-specific erythropoiesis model (e.g., an individualized erythropoiesis model) for the user based on the individualized user data (e.g., the user's individual data).

[0056] In some examples, the computing device may initially use a generic erythropoiesis model based on user characteristics (e.g., the user's gender or weight). For example, the computing device or a separate computing entity may generate a generic erythropoiesis model based on population data associated with multiple users. The computing device may then obtain the generic erythropoiesis model and modify / adjust the generic erythropoiesis model to be more appropriate for the user (e.g., generating an individualized erythropoiesis model). This will be in Figure 4 Detailed description is given in further detail in .

[0057] At block 302, a computing device (e.g., computing system 108 and / or user device 110) obtains individualized user data for a user. The individualized user data indicates one or more previous hematocrit and / or hemoglobin measurements for the user and / or previous PO2 information for the user. For example, in some examples, user device 110 may generate previous PO2 information for the user and provide the generated previous PO2 information for the user to computing system 108. For example, referring to Figure 1B , the user device 110 can use one or more sensors to measure the PO2 level over a period of time. Using these measurements, the user device 110 can generate previous PO2 information, which can include one or more hypoxic events over a period of time. In addition, as shown, hypoxic events can be at different PO2 levels. Examples of hypoxic events include high altitude events, hypoxic chambers that simulate high altitude conditions, altitude training masks, and the like. In some cases, the computing system 108 and / or the user device 110 can receive hemoglobin and / or hematocrit measurement levels from the external measurement system 104. For example, the external measurement system 104 can be associated with a blood laboratory that is configured to measure the hemoglobin and / or hematocrit measurement levels of the user's blood and provide the measured values ​​to the computing system 108 and / or the user device 110.

[0058] Additionally and / or alternatively, individualized user data may include, but is not limited to, the user's hemoglobin, hematocrit, mean corpuscular hemoglobin content, erythropoietin level, gender, height, weight, residential altitude, smoking status, clinical parameters such as oxygen saturation, treatment time, bleeding events, transfusions, hospitalizations, illnesses, and / or lifestyle choices of the user that alter the combination of oxygen and HGB, and the like.

[0059] In some examples, previous hematocrit and / or hemoglobin measurements for a user may indicate multiple different hematocrit and / or hemoglobin measurements. For example, a user may seek to achieve a specific elevated hemoglobin level (e.g., an HGB target) at a predefined time point and may experience repeated hypoxic events. The user device 110 and / or external measurement system 104 may obtain and / or determine hemoglobin levels before, during, and after a hypoxic event in order to measure the user's response to hemoglobin and / or EPO in response to these perturbations.

[0060] At box 304, the computing device determines an individualized erythropoietic model for the user based on one or more previous hematocrit and / or hemoglobin measurements of the user, the user's previous PO2 information, and a mathematical model for oxygen rationing (e.g., the mathematical model described below in FIG. 5). The individualized erythropoietic model indicates an individualized set of model parameters for the user (e.g., an individualized virtual avatar). For example, the individualized set of model parameters may be a value, function, or constant associated with a variable of the mathematical model. For example, the mathematical model described in further detail in FIG. 5 includes a plurality of mathematical equations and user parameters (e.g., variables). The computing device may use individualized user data to determine the user parameters (e.g., variables) for the mathematical model. In other words, each user and / or user group (e.g., a group of users with similar user characteristics (such as weight, height, gender, etc.)) may react slightly differently to a hypoxic event. Therefore, a first user and / or user group performing a hypoxic event may have a different effect (e.g., a different hematocrit / hemoglobin concentration level) than another user and / or user group. The mathematical model may be associated with a first set of variables that are non-user / group specific and a second set of variables that are user / group specific. The second set of variables may be a set of individualized model parameters indicated by the virtual user avatar. Thus, the computing device generates an individualized virtual avatar by inputting individualized user data (e.g., previous hemoglobin / hematocrit measurements and / or previous PO2 information) into the mathematical model to determine the set of individualized model parameters.

[0061] The personalized parameter set for the user may include, but is not limited to, production of the hormone EPO, RBC lifespan, parameters indicating the user's specific range of hemoglobin and EPO response to hypoxic events, parameters indicating the user's basal hemoglobin level in the absence of anemia medication, and / or iron-related parameters.

[0062] Partial Pressure It can be described in the following equation (1.7). For example, the partial pressure can be associated with PO2 information (e.g., the previous PO2 information obtained at block 302). For example, the partial pressure can be a time series function of the PO2 information (e.g., the time series function of the graphical representation 150).

[0063] RBC lifespan It can be described in the following equation (1.17). RBC lifespan can indicate the lifespan of the red blood cells for the user. For example, the computing device can determine the RBC lifespan based on the apoptosis rate for the cell population, the EPO for downregulating the apoptosis of the progenitor cells, the basal EPO production, and / or the new cell lysis rate regulated by the hormone EPO. RBC lifespan can indicate the average lifespan of the red blood cells.

[0064] Set point for oxygenated hemoglobin concentration in blood It can be described in the following equations (1.6) and (1.7). For example, the computing device can determine the set point based on the blood hemoglobin concentration as measured by the kidney, the number of red blood cells, activation and inhibition functions, the upregulation of the HIF signal in response to hypoxia due to reduced oxygenated hemoglobin concentration in the blood, the upregulation of the HIF signal in response to hypoxic conditions, the decay rate of the HIF signal, and / or other factors. The HB set point can be defined as the blood HB concentration at which the activity of endogenous feedback such as upregulation or downregulation of RBC production is minimal.

[0065] Basal EPO synthesis rate (β e ) can be described in the following equation (1.8). For example, the computing device can determine the basal EPO synthesis rate based on serum EPO concentration, EPO synthesis, and / or EPO decay rate. The basal EPO synthesis rate can indicate the rate at which EPO is synthesized in vivo.

[0066] Individualized model parameters may also include additional model parameters. For example, input conditions, limitations, and / or options of the system training strategy may be input into the model. For example, additional model parameters may include, but are not limited to, the expected total length of time that can be used to apply controlled hypoxic conditions, the maximum possible drop in PO2, the earliest and latest possible time points of a hypoxic event before the target time point at which the desired HGB level should be reached, and the like.

[0067] Additionally and / or alternatively, another model parameter can be the time of the performance event (e.g., marathon, competition, race) at which the peak hemoglobin level is desired. Another model parameter can include the method of oxygen reduction. For example, some people may not have access to a hypoxic chamber, but may live near a mountainous area. Alternatively, some people may not have access to a hypoxic chamber and may not have access to a mountainous area. Some people may have access to a hypoxic chamber or a mountainous area, but only with limited frequency (e.g., on weekends). These individualized constraints can be input, so the generated system training strategy takes these constraints into account, which will be described below. In some examples, the altitude of available high-altitude training can also be a model parameter. Another model parameter can be the duration of the peak hemoglobin level. For example, some people may want to maximize their hemoglobin level in a single event that lasts for a day. Others may want to maximize their hemoglobin level over a longer period of time, for example, a bicycle race that lasts for days or even weeks.

[0068] To generate personalized model parameters, the computing device can systematically fit (e.g., assign) a specified set of model parameters to the personalized user data using inputs (e.g., previous hypoxic event data). The computing device then determines the "goodness" of the fit based on the deviation between the simulated and measured hemoglobin / hematocrit concentrations (e.g., the deviation between the hemoglobin / hematocrit concentration output by the personalized erythropoiesis model using the assigned set of model parameters and the actual hemoglobin / hematocrit concentration from the personalized user data). The computing device can define the goodness of fit using a cost function, and can perform the fitting process using one or more parameter fitting algorithms. In some examples, other simulated and measured data element pairs (e.g., erythropoietin serum concentration, iron plasma concentration, etc.) can also be used to determine the goodness of fit.

[0069] In other words, the computing device may assign values ​​to the model parameter set described above (e.g., partial pressure, RBC lifespan, set point for oxygenated hemoglobin concentration in the blood, basal EPO synthesis rate, and / or additional model parameters). Subsequently, the computing device may determine (e.g., obtain / use) previous hypoxic events (e.g., previous PO2 information) and previous hemoglobin / hematocrit concentrations. The computing device may then input the previous hypoxic events and the assigned model parameter set into the mathematical model to determine the simulated hemoglobin / hematocrit concentration. The computing device may compare the simulated hemoglobin / hematocrit concentration with the actual hemoglobin / hematocrit concentration to determine the deviation between the two. Thereafter, the computing device may assign new values ​​to the model parameter set and repeat the process. The computing device may continuously assign new values ​​to the model parameter set and determine the deviation between the simulated hemoglobin / hematocrit concentration and the actual hemoglobin / hematocrit concentration. For example, the computing system 108 may use the previously calculated deviations to determine which new values ​​to try in the next iteration. In other words, the computing device may use one or more parameter fitting algorithms (e.g., optimizing one or more distance measurements between simulated concentrations and actual concentrations) to determine an optimal set of model parameters for a subset of the population user data. The computing device may determine the model parameter set for the virtual user avatar as the optimal set of model parameters. For example, the computing device may compare the distance between the simulated hemoglobin / hematocrit concentration and the actual hemoglobin / hematocrit concentration, and use the model parameter set that produces the minimum distance between the simulated concentration and the actual concentration.

[0070] In some examples, the computing device may use hypoxic events and / or other information (eg, patient characteristics such as gender, height, weight, blood volume, etc.) to determine an individualized erythropoiesis model for the user.

[0071] In some variations, the computing device may select an individualized erythropoiesis model for the user from one of a plurality of virtual patient avatars. For example, each of the plurality of virtual user avatars may be associated with a subset of users (e.g., based on the user's height, weight, gender, etc.). Based on comparing the individualized user data to a subset of the group user data, the computing device may determine a virtual user avatar that is most similar to the user based on the individualized user data (e.g., a virtual user avatar that has the same or similar user characteristics (such as height, weight, gender, hypoxic events, hemoglobin concentration measurements, etc.)). This will be performed in a manner that is consistent with the user's individualized erythropoiesis model. Figure 4 Detailed description is given in further detail in .

[0072] At box 306, the computing device uses the personalized erythropoiesis model to determine one or more predicted hematocrit and / or hemoglobin measurements at one or more future moments. For example, the computing device may input data (e.g., the user's measured hematocrit and / or hemoglobin level) into the personalized erythropoiesis model, and the output of the personalized erythropoiesis model may be one or more predicted hematocrit and / or hemoglobin measurements at one or more future moments. In some cases, the one or more predicted hematocrit and / or hemoglobin measurements may include the expected hematocrit and / or hemoglobin measurements when the performance event occurs. Additionally and / or alternatively, the one or more predicted hematocrit and / or hemoglobin measurements may include one or more interim hematocrit and / or hemoglobin measurements (e.g., checkpoint hematocrit and / or hemoglobin measurements) that indicate a checkpoint to be reached for the user in their goal of reaching the expected hematocrit and / or hemoglobin measurements. For example, the checkpoint hematocrit and / or hemoglobin measurements may be 50% and 75% of the desired hematocrit and / or hemoglobin measurements.

[0073] Subsequently, the computing device may display information associated with the predicted hematocrit and / or hemoglobin measurement (e.g., a checkpoint hematocrit and / or hemoglobin measurement and / or an expected hematocrit and / or hemoglobin measurement). Additionally and / or alternatively, the computing device may employ the individualized erythropoiesis model to generate a system training strategy (e.g., determining one or more first hypoxic events to be performed for the user). For example, after determining the individualized erythropoiesis model at box 304, the computing device may generate a plurality of future PO2 information (e.g., a plurality of time series functions with hypoxic events). The future PO2 information may indicate a system training strategy. In other words, the previous PO2 information obtained at box 302 may indicate a hypoxic event that the user has experienced (e.g., a hypoxic event that the user has experienced in the past month). The future PO2 information may indicate a hypoxic event that the user is to complete in the future (e.g., to be completed within the next two months) to achieve the desired hematocrit and / or hemoglobin measurement when the performance event occurs. For example, the future PO2 information may indicate one or more hypoxic events to be performed for the user, and each hypoxic event may indicate a PO2 level and a time period or time instance. The computing device may input each of the future PO2 information into the personalized erythropoiesis model, and the personalized erythropoiesis model may output a predicted hematocrit and / or hemoglobin measurement value at a future moment. The computing device may then compare the predicted hematocrit and / or hemoglobin measurement value with an expected hematocrit and / or hemoglobin measurement value for the user (e.g., an expected hematocrit and / or hemoglobin measurement value to be obtained when the performance event occurs). Based on the comparison, the computing device may determine the best future PO2 information (e.g., future PO2 information with a predicted hematocrit and / or hemoglobin measurement value that is closest to the expected hematocrit and / or hemoglobin measurement value). Subsequently, the computing device may determine a system training strategy associated with the future PO2 information. For example, the system training strategy may indicate a hypoxic event to be performed by the user in an upcoming time period before the performance event. The computing device may display the system training strategy.

[0074] In some cases, the computing device may determine the system training strategy based on individualized constraints for the user. For example, each user may be unique and may be able to handle constraints differently from other users. For example, a healthy user may be able to be at a higher PO2 level for a longer period of time, while a less healthy user may not be able to do so. The individualized constraints may include any characteristics, features, or indicators associated with the user that can be used to optimize the system training strategy. For example, the individualized constraints may include, but are not limited to, the total time that the user can spend indoors or at an elevated altitude, the time the user trains, the PO2 values ​​(e.g., ranges) that the user can experience, the maximum or minimum PO2 levels that the user can be exposed to, or other types of individualized constraints. The computing device may use individualized constraints to optimize system training. For example, based on a computing device that determines that the optimal future PO2 information is outside the user's individualized constraints (e.g., the optimal future PO2 information indicates a maximum or minimum PO2 level outside the user's maximum or minimum PO2 level indicated by the individualized constraints), the computing device may reject the optimal future PO2 information and select different future PO2 information for the user (e.g., a second or third optimal future PO2 information with an indicator within the user's individualized constraints).

[0075] In some cases, process 300 can end after block 306. In other cases, process 300 can also include blocks 308 and 310, which describe re-adopting the individualized erythropoiesis model. In other words, blocks 308 and 310 include dashed lines to indicate that these blocks are optional in process 300.

[0076] When executed, the computing device obtains one or more new hematocrit and / or hemoglobin measurements for the user and / or additional PO2 information for the user at block 308. The additional PO2 information may indicate one or more hypoxic events that the user has experienced since employing the personalized erythropoietic model at block 306. At block 310, the computing device determines whether to re-employ the personalized model.

[0077] For example, the computing device may determine whether to re-apply the personalized erythropoietic model for the user at a later time. For example, a performance event (e.g., an athletic event or surgery) may occur some amount of time in the future (e.g., 2 months in the future), and the user may seek to achieve a desired hematocrit and / or hemoglobin level at that time. The computing device may initially perform blocks 302-306 to determine future PO2 information and / or predicted hematocrit and / or hemoglobin measurements. Then, after a period of time has passed (e.g., after a month), the computing device may perform blocks 308 and 310 to determine whether to re-apply the personalized erythropoietic model. For example, the user may encounter one or more setbacks that prevent them from performing one or more hypoxic events indicated by the future PO2 information. Thus, after one of the two months until the performance event occurs has passed, the computing device may perform blocks 308 and 310 to determine updated future PO2 information so that the user can still meet the desired hematocrit and / or hemoglobin measurements when the performance event occurs. For example, after a period of time (e.g., one month), the computing device may obtain one or more new hematocrit and / or hemoglobin measurements and / or additional PO2 information for the user. The additional PO2 information may include the user's previous PO2 information and new PO2 information (e.g., new hypoxic events) experienced by the user since the personalized erythropoiesis model was adopted at block 306. The computing device may compare the new hematocrit and / or hemoglobin measurements with one or more predicted hematocrit and / or hemoglobin measurements (e.g., one or more checkpoint hematocrit and / or hemoglobin measurements) from block 306. Based on the comparison, the computing device may determine whether to re-adopt the personalized erythropoiesis model. For example, based on the computing device determining that the new hematocrit and / or hemoglobin measurements are within a certain threshold of the predicted hematocrit and / or hemoglobin measurements (e.g., ±5% from the predicted hematocrit and / or hemoglobin measurements), the computing device may determine not to re-adopt the personalized erythropoiesis model.

[0078] Based on the computing device determining that the new hematocrit and / or hemoglobin measurement is outside of a certain threshold, the computing device may determine to re-apply the personalized erythropoiesis model. In some cases, by re-applying the personalized erythropoiesis model, the computing device may determine one or more new predicted hematocrit and / or hemoglobin measurement values ​​and / or updated future PO2 information. The new predicted hematocrit and / or hemoglobin measurement values ​​and / or updated future PO2 information may be used to help the user achieve their desired hematocrit and / or hemoglobin measurement values ​​when the performance event occurs.

[0079] In some variations, one or more blocks of process 300 may be repeated. For example, blocks 308 and 310 may be repeated one or more times before reaching the occurrence of a performance event. For example, after a set amount of time (e.g., every week or every few days), the computing device may obtain new hematocrit and / or hemoglobin measurements for the user and additional PO2 information. The computing device may then employ the personalized erythropoietic model as many times as needed. For example, as mentioned above, the user device 110 may include one or more sensors that are configured to continuously obtain PO2 information for the user. Thus, after a certain amount of time has passed, the user device 110 may provide additional PO2 information from the sensors to the computing system 108, and the computing system 108 may re-employ the personalized erythropoietic model.

[0080] In some cases, where the user is iron deficient, the computing device may use iron intake as a parameter for the personalized erythropoiesis model.

[0081] Figure 4 is a flow chart of an exemplary process 400 for generating a model of erythropoiesis according to one or more examples of the present application. The process may be performed by, for example, Figure 1A The personalized erythropoiesis model generation computing system 108 and / or the computing device of the user device 110 depicted in FIG. However, it will be appreciated that any of the following blocks may be performed in any suitable order and that the process 400 may be performed in any suitable environment and by any suitable device or system. Figure 4 The descriptions, diagrams, and processes of are merely exemplary, and process 400 may use other descriptions, diagrams, and processes to generate a model of erythropoiesis.

[0082] For example, in some cases, the individualized user data may be missing certain data points, which may make it insufficient to determine an individualized erythropoietic model. Additionally and / or alternatively, the user may be part of a group of users with similar characteristics or goals (e.g., a team of football players, a group of users about to do a triathlon, or a group of users with similar physiques). In this way, the computing device may generate virtual user avatars for the group of users.

[0083] At block 402, the computing device obtains a mathematical model for generating a virtual user avatar. For example, the computing device may receive the mathematical model (e.g., a mathematical model for oxygen rationing) from another entity (e.g., a server or computing system associated with an enterprise organization). Additionally and / or alternatively, the mathematical model may be stored in a memory (e.g., an external memory or a memory within the computing system 108), and the computing device may retrieve the mathematical model from the memory. Figure 5AOne exemplary mathematical model 500 that may be obtained by a computing device for monitoring a user's hemoglobin or hematocrit level is shown and described in further detail below.

[0084] At box 404, the computing device generates a plurality of virtual user avatars based on the group user data (e.g., historical data and / or simulation data for a plurality of users) and the obtained mathematical model. Each of the virtual user avatars is a personalized model parameter set for the mathematical model. The group user data may be historical data for a plurality of users and / or simulation data for a plurality of users. For example, the group user data may include, but is not limited to, hemoglobin, hematocrit, erythropoietin serum, gender, height, weight, clinical parameters such as oxygen saturation, treatment time, bleeding events, transfusions, hospitalizations, and the like for a plurality of users. The group user data may be within a specified time window (e.g., a certain time period).

[0085] The computing device can generate a virtual user avatar for the obtained mathematical model, wherein the virtual user avatar is a personalized model parameter set. Figure 5A The mathematical model further described in detail in includes a plurality of mathematical equations and user parameters (e.g., variables). The computing device can use group user data to determine user parameters (e.g., variables) for the mathematical model. In other words, each user and / or user group (e.g., a group of users with similar user characteristics (such as weight, height, gender, etc.)) may react slightly differently to a hypoxic event. Therefore, a first user and / or user group performing a hypoxic event may have a different effect (e.g., different hematocrit / hemoglobin concentration levels) than another user and / or user group. The mathematical model can be associated with a first set of variables that are non-user / group specific and a second set of variables that are user / group specific. The second set of variables can be a personalized model parameter set indicated by a virtual user avatar. Therefore, the computing device generates a virtual user avatar by inputting group user data into the mathematical model to determine the personalized model parameter set.

[0086] The set of personalized model parameters associated with each virtual user avatar can be similar to the one described above in Figure 3, except that they are for a group of users rather than a single user. Thus, each virtual user avatar may indicate a personalized model parameter set for a particular group, and the computing device may generate a plurality of these virtual user avatars. To generate a virtual user avatar, the computing device may systematically fit (e.g., assign) a specified model parameter set to the group user data using inputs (e.g., previous hypoxic event data). The computing device then determines the "goodness" of the fit based on the deviation between the simulated and measured hemoglobin / hematocrit concentrations (e.g., the deviation between the hemoglobin / hematocrit concentration output by the erythropoiesis model using the assigned model parameter set and the actual hemoglobin / hematocrit concentration from the group user data). The computing device may define the goodness of fit using a cost function, and may perform the fitting process using one or more parameter fitting algorithms. In some examples, other simulated and measured data element pairs (e.g., erythropoietin serum concentration, iron plasma concentration, etc.) may also be used to determine the goodness of fit.

[0087] In other words, the computing device may determine a subset of the group user data based on certain characteristics (e.g., height, weight, gender, etc.). The computing device may assign values ​​to the model parameter set described above. Subsequently, the computing device may determine previous hypoxic events and hemoglobin / hematocrit concentrations within the subset of the group user data. The computing device may then input the previous hypoxic events and the assigned model parameter set into a mathematical model to determine a simulated hemoglobin / hematocrit concentration. The computing device may compare the simulated hemoglobin / hematocrit concentration with the actual hemoglobin / hematocrit concentration to determine the deviation between the two. Thereafter, the computing device may assign new values ​​to the model parameter set and repeat the process. The computing device may continuously assign new values ​​to the model parameter set and determine the deviation between the simulated hemoglobin / hematocrit concentration and the actual hemoglobin / hematocrit concentration. For example, the computing system 108 may use the previously calculated deviations to determine which new values ​​to try in the next iteration. In other words, the computing device may use one or more parameter fitting algorithms (e.g., optimizing one or more distance measurements between simulated concentrations and actual concentrations) to determine an optimal set of model parameters for a subset of the population user data. The computing device may determine the model parameter set for the virtual user avatar as the optimal set of model parameters. For example, the computing device may compare the distance between the simulated hemoglobin / hematocrit concentration and the actual hemoglobin / hematocrit concentration, and use the model parameter set that produces the minimum distance between the simulated concentration and the actual concentration.

[0088] The computing device may determine a plurality of virtual user avatars based on the subset of the usage population data, and each of the plurality of virtual user avatars may be associated with certain user characteristics (eg, gender, weight, height) and a set of personalized model parameters.

[0089] In some cases, the computing device may perform a pre-calibration function before generating a virtual user avatar. For example, the computing device may determine blood volume based on a subset of the group user data. For example, using the gender, height, weight, and / or other information from the subset of the group user data, the computing device may determine blood volume for the virtual user avatar. Then, using the blood volume, the computing device may determine rates such as formation rate, synthesis rate, and / or other rates for the virtual user avatar. For example, the computing device may determine the HIF signal gain rate (shown as β in equation (1.6) below) h ), EPO synthesis rate (shown as β in equation (1.8) below) e ), the rate of progenitor cell formation (shown as β in equation (1.9) below) prog ). The computing device may use the determined rate, the assigned model parameter set, previous hypoxic events, mathematical models, and / or other data to determine a simulated hemoglobin / hematocrit concentration. Then, based on the simulated hemoglobin / hematocrit concentration, the computing device may determine an optimal set of model parameters to use for the virtual user avatar.

[0090] In some variations, in addition to the simulated hemoglobin / hematocrit concentration, the computing device may also use the mathematical model, previous hypoxic events, and / or other information to determine EPO (shown as variable "e" in equation (1.8)) and / or other simulated data points. The computing device may also obtain EPO and / or other data points from a subset of the population user data. The computing device may compare the simulated data points to the actual data points and use one or more parameter fitting algorithms to determine an optimal set of model parameters for the virtual user avatar.

[0091] At block 406, the computing device uses the plurality of virtual user avatars to determine an individualized erythropoiesis model. Figure 3As mentioned in block 304 of , based on comparing the individualized user data with a subset of the group user data, the computing device may determine a virtual user avatar that is most similar to the user based on the individualized user data (e.g., a virtual user avatar having the same or similar user characteristics (such as height, weight, gender, hypoxic events, hemoglobin concentration measurements, etc.)). The computing device may then use the individualized user data to update the personalized model parameters of the virtual user avatar. For example, the computing device may change one or more of the personalized model parameters based on comparing the obtained hemoglobin concentration of the user with the simulated hemoglobin concentration output by the erythropoiesis model. In some cases, the computing device may use the parameters indicated by the determined virtual user avatar as the personalized erythropoiesis model.

[0092] Figure 5A is a block diagram of an exemplary mathematical model 500 according to one or more examples of the present application. As previously mentioned, the mathematical model 500 can be used to determine / generate multiple virtual user avatars, determine an individualized erythropoiesis model for a user, and / or determine the timing and magnitude of a user's hypoxic event to adjust the user's hematocrit / hemoglobin concentration (e.g., future PO2 information).

[0093] For example, block 502 is exposure to controlled hypoxic conditions. For example, as previously mentioned, computing system 108 may obtain exposure of a user to controlled hypoxic conditions (e.g., PO2 information such as previous and / or additional PO2 information). The PO2 information may be a time series of the user's PO2 levels over a period of time, such as Figure 1B A graphical representation 150 is shown.

[0094] Box 504 of mathematical model 500 is the physiological function of the HIF pathway and erythropoietin (EPO). For example, HIF (hypoxia inducible factor) is a constitutively expressed transcription factor that causes a response to reduced levels of available oxygen by upregulating the expression of various target genes. The HIF transcription complex is a heterodimer of its subunits HIF-α and HIF-β. HIF-α has three subtypes, called HIF-1α, HIF-2α, and HIF-3α. Under normoxic conditions, HIF-α is located in the cytoplasm; HIF-β is located in the nucleus. As O 2 As the concentration decreases, HIF-α (especially the subtype HIF-2α) stabilizes, translocates to the nucleus and binds to HIF-β to form the HIF transcription complex. The HIF complex regulates the transcription of target genes like EPO in the kidney and liver. This is achieved by binding to the hypoxia-responsive elements (HRE: Hypoxia-Responsive Elements) in the corresponding gene promoter.

[0095] The HIF pathway includes multiple biochemicals, such as HIF-α and HIF-β and their dimers. To reduce the complexity of the model, a single composite variable h representing the “HIF signal” is introduced. More precisely, let n cmp represents the amount of HIF transcription complex bound to the EPO promoter region, and represents its average value under normoxic conditions, then h can be defined as

[0096] Therefore, the sign of h depends on n cmp Is the value above or below the steady-state baseline level?

[0097] In healthy patients, HIF signaling is upregulated in the presence of hypoxia via feedback in the kidneys. Here, consider the set point of oxygenated hemoglobin concentration in the blood HIF signaling responds to any deviation from the set point. In addition, since hypoxia prevents HIF-α degradation via reduced proline hydroxylation, they effectively upregulate the amount of available HIF complexes that can bind to the EPO promoter. Therefore, h is upregulated under hypoxic conditions. These contributions are combined in equations (1.6) and (1.7) below.

[0098]

[0099] in,

[0100]

[0101] represents the concentration of oxygenated hemoglobin in the blood measured by the kidneys, which is the hemoglobin concentration c hgb (derived model variables, see equation (D)) and O2 partial pressure In addition, represents the physiological set point of oxygenated hemoglobin concentration in the blood, and β h (θ) represents the formation rate of HIF complex. and φ - The function of type (x) (shown below) is a sigmoid function that describes nonlinear activation and inhibition. The first term Describes the upregulation of HIF signaling in response to hypoxia, which is caused by decreased blood hemoglobin concentration. Function and φ - (x) is defined as follows:

[0102]

[0103] Where x is the input variable, and a, b, and q are shape parameters that determine the offset, height, and steepness of the function. Equation (A) can be called a Hill-type function. Equation (B) can be called a Richards-type equation.

[0104] Returning to equation (1.6), parameter k h represents the decay rate of the HIF signal. In other words, Indicates the rate of change of HIF signaling activity, β h (θ) represents the formation rate of HIF complex, indicates that HIF signaling is upregulated in response to hypoxia, which reduces the concentration of oxygenated hemoglobin in the blood, k h represents the decay rate of the HIF signal, Represents the concentration of oxygenated hemoglobin in the blood as measured by the kidneys, represents the physiological set point of oxygenated hemoglobin concentration in the blood, and h represents the HIF signal. The function S is considered to have the following form:

[0105]

[0106] Among them, the second term describes the S-shaped dependence of the oxygenated hemoglobin fraction in the blood on the inspiratory PO2, which describes the net effect of the monotonically increasing dependence of renal PO2 on inspiratory PO2 and the S-shaped oxygen dissociation curve. Here, represents the inspiratory PO2 required to achieve half-maximal oxygen saturation, r s is an exponent describing the steepness of the S-shaped function.

[0107] The serum EPO concentration is described by the second variable e, which is upregulated by HIF signaling and increases at a rate k. e attenuation,

[0108]

[0109] The first term describes the basal EPO synthesis under steady-state conditions. The second term describes the additional EPO synthesis due to activation of HIF signaling. The third term describes the EPO synthesis at a rate k e In other words, for equation (1.8), represents the rate of change of EPO concentration, β e (θ) represents the basal erythropoietin (EPO) synthesis under steady-state conditions, represents additional EPO synthesis due to HIF signaling activation, and k e e represents the decay rate k e EPO decay.

[0110] Boxes 506 and 508 of mathematical model 500 are the physiological functions of erythrocyte formation in the bone marrow and blood hemoglobin concentration (clinical target variables). In particular, erythrocytes (erythrocytes) are continuously generated through differentiation and proliferation of a hierarchy of stem cells and progenitor cells. Hematopoietic stem cells (HSCs) are the highest level of the hierarchy. They self-renew, while a portion of their progeny successively differentiate into more lineage-specific cell types (megakaryocyte-erythroid progenitor cells, erythroid burst-forming units [BFU-E], erythroid colony-forming units [CFU-E], proerythroblasts, erythroblasts, and reticulocytes). This process is regulated by the HIF pathway, which in turn regulates the production of endogenous EPO. EPO inhibits apoptosis (cell death) of progenitor cells. The last step (formation of erythrocytes) also requires iron.

[0111] The erythropoietic lineage in the bone marrow consists of a variety of different cell types and states. These cell types can be grouped according to whether their cell fate is regulated by EPO or iron (the master regulator in the mathematical model 500). Therefore, the population size n of three composite cell populations was introduced prog 、n pre and n ery , which represent (i) progenitor populations (colony forming unit-erythroid (CFU-E) and proerythroblasts), (ii) immediate precursors of erythroid cells (erythroblasts and reticulocytes), and (iii) erythrocytes, respectively. To avoid huge numbers, all these population sizes are normalized so that they represent the average progeny of a representative unit pool of BFU-E (burst forming unit of erythroid cells).

[0112] In this scheme, apoptosis of the progenitor population is downregulated by EPO. Differentiation of precursor cells into erythroid cells is limited by iron availability; if insufficient iron is available, cells ready to differentiate die immediately, so the total loss of precursor cells due to "attempting" to differentiate is constant. The corresponding kinetics are given by the following equation.

[0113]

[0114] Here, β prog (θ) represents the rate of progenitor cell formation, which depends on patient-specific demographic and physiological parameters θ, including weight, sex, etc. Therefore, β prog Proportional to the product of the number of available BFU-E and their differentiation rate. i→j represents the differentiation flux from population i to j (e.g., population progenitors (prog) to precursors (pre) or pre to erythrocytes (ery)), K i (For example, K prog or K pre ) represents the cell flux due to apoptosis. Factor represents the d due to cell division in the progenitor or precursor compartment, respectively prog or pre In other words, for equation (1.11), Indicates the rate of change of the total amount of red blood cells. represents the differentiation flux from the precursor cell population to the erythroid population, K ery represents the cell flux due to apoptosis.

[0115] The total differentiation flux is given by the following equation.

[0116] J prog→pre =ω prog→pre n prog Equation (1.12)

[0117] J pre→ery =ω pre→ery v mat (e)φ iron n pre Equation (1.13)

[0118] Here, ω i→j (For example, ω pre→ery and ω prog→pre ) represents the differentiation rate from population i to j, φ iron is a phenomenological parameter describing the availability of iron in plasma. In addition, v mat (e) represents an epo-dependent regulatory factor that slows down or accelerates maturation as the EPO level changes, which is described in the following equation.

[0119]

[0120] in, Indicates the EPO set point level.

[0121] The cell death flux is given by the following equation.

[0122]

[0123] K pre =ω pre→ery [1-φ iron ]n pre Equation (1.16)

[0124] K ery =k ery n ery +k neo (e)n ery Equation (1.17)

[0125] Among them, k i (For example, k ery and κ pro g) represents the apoptosis rate of cell population i, and is the EC50 of Epo that downregulates apoptosis of progenitor cells, k neo represents the rate of neocellular lysis regulated by the hormone EPO, which is given by the following equation:

[0126]

[0127] Among them, e neo Indicates the threshold for new cell lysis.

[0128] Note that the amount of available iron determines the relative proportion of precursor cells that are triggered to apoptosis or differentiation, respectively; however, the overall “loss” of the precursor cell compartment is independent of iron and is determined by J pre→ery +K pre =ω pre→ery *n pre Given.

[0129] Regarding hemoglobin and observable values, due to the variable n i represents the average progeny of a representative pool of BFU-E (burst-forming units of erythrocytes), so the absolute cell population size is given by Ni=Q prog n i Given, where Q prog =10 9 is the conversion factor. Therefore, the blood hemoglobin concentration is given by the following equation:

[0130] c hgb =λ hgb n ery Equation (D)

[0131] in,

[0132]

[0133] Expressed in typical units, it reads:

[0134]

[0135] The concentration of oxygenated hemoglobin in the blood Depends on blood hemoglobin concentration c hgb and O2 partial pressure

[0136] Figure 5B One or more examples of the present application are described. Figure 5A A block diagram of the blocks within an exemplary mathematical model of . For example, Figure 5B Includes various physiological processes at work. For example, Figure 5B A box 552 for controlled hypoxic events, a box 554 for HIF signaling, a box 556 for erythropoietin, a box 566 for bone marrow including red blood cell boxes 558, erythroid precursor cells 560 and erythroid progenitor cells 562, and a box 564 for plasma iron are shown. Figure 5A These boxes are described in .

[0137] Now go to Fig. 6A , there was a study in Chile that involved members of the Chilean army undergoing intermittent hypoxic exposure at high altitude (3550m). Fig. 6A The schedule of intermittent hypoxic exposure for the subject groups in this study is shown. Fig. 6A From the publication regarding this study, the publication is Heinicke K, Prommer N, Cajigal J, Viola T, Behn C, Schmidt W. Long-term exposure to intermittent hypoxia results in increased hemoglobin mass, reduced plasma volume, and elevated erythropoietin plasma levels in man. Eur J Appl Physiol. 2003 Feb;88(6):535-43. doi:10.1007 / s00421-002-0732-z. Epub 2002 Dec 14. PMID:12560952s, the contents of which are incorporated herein by reference in their entirety. Fig. 6A Soldiers and officers who experienced intermittent hypoxia are shown (SI for the soldier group and OI for the officer group, represented by dotted and solid lines, respectively). Blood samples were collected from SI subjects on days 1, 2, 4, 11 and 14 (hollow arrows), and blood samples were collected from OI subjects on days 1, 2, 4 and 7 (solid arrows). SL soldiers were at sea level and RA residents were at altitude. The SI group consisted of 10 soldiers who experienced several intermittent hypoxia periods. They served alternately for 6 months at an altitude of 3550m (11-day cycle) and at sea level (3-day cycle). The OI group consisted of 9 officers who experienced intermittent hypoxia. They served alternately for approximately 22 years at an altitude of 3550m (3.5-day cycle) and at sea level (3.5-day cycle).

[0138] To validate the above erythropoiesis model, experimental data from the Chilean study were input into the erythropoiesis model and the results from the model are shown in Figure 6B-6K middle. Figure 6B-Figure 6FA graphical representation of the application of the above erythropoiesis model to the hypoxic exposure experienced by soldiers in the study is shown (e.g., Fig. 6A For example, Figure 6B shows the ambient PO2 (model variable Equation 1.7). Figure 6C The HIF signal (model variable h, equation 1.6) is shown over a period of time. Fig.6D Serum EPO concentrations over time are shown (model variable e, Equation 1.8) as well as actual measured serum EPO concentrations (eg, points with vertical lines, where the points represent the mean and the vertical line represents the standard deviation). Fig. 6E shows the blood hemoglobin concentration over a period of time (model variable c hgb , Equation 1.7) and actual measured HGB data points (eg, points with vertical lines, where the points represent the mean and the vertical lines represent the standard deviation). Fig. 6F shows the abundance of a cell population over time (model variable n prog 、n pre 、n ery , equations 1.9-1.11).

[0139] Figure 6G-6K A graphical representation of the application of the above erythropoiesis model to the hypoxic exposure experienced by officers in the study is shown (e.g., Fig. 6A For example, Figure 6G shows the ambient PO2 (model variable Equation 1.7). Figure 6H The HIF signal (model variable h, equation 1.6) is shown over a period of time. Fig.6I Serum EPO concentrations over time are shown (model variable e, Equation 1.8) as well as actual measured serum EPO concentrations (eg, points with vertical lines, where the points represent the mean and the vertical line represents the standard deviation). Figure 6J shows the blood hemoglobin concentration over a period of time (model variable c hgb , Equation 1.7) and actual measured HGB data points (eg, points with vertical lines, where the points represent the mean and the vertical lines represent the standard deviation). Figure 6K shows the abundance of a cell population over time (model variable n prog 、n pre 、n ery , equations 1.9-1.11).

[0140] It should be understood that the various machine-implemented operations described herein may occur via execution by one or more respective processors of processor-executable instructions stored on a tangible, non-transitory computer-readable medium, such as a random access memory (RAM), a read-only memory (ROM), a programmable read-only memory (PROM), and / or another electronic memory mechanism. Thus, for example, the operations performed by any device described herein may be performed according to instructions stored on the device and / or an application installed on the device, and via software and / or hardware of the device.

[0141] All references, including publications, patent applications, and patents, cited herein are hereby incorporated by reference to the same extent as if each reference were individually and specifically indicated to be incorporated by reference and were set forth in its entirety herein.

[0142] Although the present invention has been described and illustrated in detail in the accompanying drawings and the foregoing description, such description and illustration should be considered as illustrative or exemplary rather than restrictive. It should be understood that those of ordinary skill in the art can make changes and modifications within the scope of the appended claims. In particular, the present application covers other embodiments with any combination of features from the different embodiments described above and below.

[0143] The terms used in the claims should be interpreted as having the broadest reasonable interpretation consistent with the foregoing description. For example, the use of the article "one" or "an" when introducing an element should not be interpreted as excluding multiple elements. Similarly, the narration of "or" should be interpreted as inclusive, so that the narration of "A or B" does not exclude "A and B", unless it is clear from the context or the foregoing description that only one of A and B is referred to. In addition, the narration of "at least one of A, B and C" should be interpreted as one or more of a group of elements consisting of A, B and C, and should not be interpreted as requiring at least one of each of the listed elements A, B and C, regardless of whether A, B and C are related as categories or otherwise. In addition, the narration of "A, B and / or C" or "at least one of A, B or C" should be interpreted as including any single entity from the listed elements, such as A, any subset of the listed elements, such as A and B, or the entire list of elements A, B and C.

[0144] Unless otherwise specified herein, the description of the range of values ​​herein is intended only to be used as a shorthand method for individually referring to each individual value falling within the range, and each individual value is incorporated into this specification as if it were individually described herein. Unless otherwise specified herein or clearly contradictory to the context, all methods described herein can be performed in any suitable order. The use of any and all examples or exemplary language (e.g., "such as") provided herein is only intended to better illustrate the present invention, without limiting the scope of the present invention, unless otherwise stated. Any language in the specification should not be interpreted as indicating that any unclaimed element is essential to the practice of the present invention.

Claims

1. A computing system comprising: A user device comprising: one or more sensors configured to obtain a partial pressure of oxygen (PO2) level; one or more first processors; and A first non-transitory computer-readable medium having first processor-executable instructions stored thereon, wherein the first processor-executable instructions, when executed by the one or more first processors, facilitate: obtaining a PO2 level over a period of time using the one or more sensors; generating previous PO2 information for a user based on the PO2 levels obtained using the one or more sensors; and providing the previous PO2 information to a personalized erythropoiesis model generation computing platform; and The personalized erythropoiesis model generation computing platform comprises: one or more second processors; and a second non-transitory computer-readable medium having second processor-executable instructions stored thereon, wherein the second processor-executable instructions, when executed by the one or more second processors, facilitate: obtaining the previous PO2 information from the user device; obtaining individualized user data for the user, wherein the individualized user data indicates one or more previous hematocrit and / or hemoglobin measurements for the user; determining an individualized erythropoiesis model for the user based on the one or more previous hematocrit and / or hemoglobin measurements and the previous PO2 information, wherein the individualized erythropoiesis model indicates an individualized set of model parameters for the user; employing the individualized erythropoiesis model to determine one or more predicted hematocrit and / or hemoglobin measurements at one or more future time points; and One or more actions are performed based on the one or more predicted hematocrit and / or hemoglobin measurements.

2. The computing system of claim 1, wherein: Determining the personalized erythropoiesis model for the user includes: The previous PO2 information and the one or more previous hematocrit and / or hemoglobin measurements are input into a mathematical model for oxygen rationing to determine the personalized erythropoiesis model.

3. The computing system of claim 1, wherein: The previous PO2 information indicates one or more hypoxic events experienced by the user during a previous time period, wherein each of the one or more hypoxic events indicates a PO2 level that is lower than a normoxic level.

4. The computing system of claim 1, wherein: Using the individualized erythropoiesis model to determine the one or more predicted hematocrit and / or hemoglobin measurements comprises: generating a plurality of future PO2 information, wherein each of the plurality of future PO2 information indicates one or more hypoxic events to be performed for the user in a future time period; inputting the plurality of future PO2 information into the personalized erythropoiesis model to determine a plurality of predicted hematocrit and / or hemoglobin measurements; and Optimal future PO2 information is determined from the plurality of future PO2 information based on comparing the plurality of predicted hematocrit and / or hemoglobin measurement values ​​to desired hematocrit and / or hemoglobin measurement values ​​associated with the user.

5. The computing system of claim 4, wherein: The second processor executable instructions, when executed by the one or more second processors, further facilitate: determining a system training strategy for the user based on the optimal future PO2 information, and Wherein, executing the one or more actions includes displaying the system training strategy on a display device associated with the personalized erythropoiesis model generation computing platform.

6. The computing system of claim 1, wherein: The first processor executable instructions, when executed by the one or more first processors, further facilitate: generating additional PO2 information for the user based on the PO2 level obtained using the one or more sensors; and The additional PO2 information is provided to the personalized erythropoiesis model generation computing platform.

7. The computing system of claim 6, wherein: The second processor executable instructions, when executed by the one or more second processors, further facilitate: After adopting the individualized erythropoiesis model, determining whether to re-adopt the individualized erythropoiesis model; and The individualized erythropoiesis model is re-adapted using the additional PO2 information from the user device.

8. The computing system of claim 1, wherein: The first processor executable instructions, when executed by the one or more first processors, further facilitate: derive a mathematical model for oxygen rationing; obtaining group user data associated with a plurality of users; and generating a plurality of virtual user avatars based on the mathematical model and the group user data, Wherein, determining the individualized erythropoiesis model for the user is further based on the multiple virtual user avatars.

9. The computing system of claim 1, wherein: Determining the personalized erythropoiesis model for the user is based on the following equation: in, Indicates the rate of change of hypoxia-inducible factor (HIF) signaling activity, β h (θ) represents the formation rate of HIF complex, indicates that HIF signaling is upregulated in response to hypoxia, which reduces blood hemoglobin concentration, k h represents the decay rate of the HIF signal, Represents the concentration of oxygenated hemoglobin in the blood as measured by the kidneys, represents the physiological set point of oxygenated hemoglobin concentration in the blood, and h represents the HIF signal.

10. The computing system of claim 1, wherein: Determining the personalized erythropoiesis model for the user is based on the following equation: in, represents the rate of change of EPO concentration, β e (θ) represents the basal erythropoietin (EPO) synthesis under steady-state conditions, represents additional EPO synthesis due to HIF signaling activation, and k e e represents the decay rate k e EPO decay.

11. The computing system of claim 1, wherein: Determining the personalized erythropoiesis model for the user is based on the following equation: in, Indicates the rate of change of the total amount of red blood cells. represents the differentiation flux from the precursor cell population to the erythroid population, and K ery represents the cell flux due to apoptosis.

12. A method comprising: obtaining individualized user data for a user, wherein the individualized user data indicates one or more previous hematocrit and / or hemoglobin measurements for the user and previous partial pressure of oxygen (PO2) information for the user; determining an individualized erythropoiesis model for the user based on the one or more previous hematocrit and / or hemoglobin measurements and the previous PO2 information, wherein the individualized erythropoiesis model indicates an individualized set of model parameters for the user; employing the individualized erythropoiesis model to determine one or more predicted hematocrit and / or hemoglobin measurements at one or more future time points; and One or more actions are performed based on the one or more predicted hematocrit and / or hemoglobin measurements.

13. The method according to claim 12, wherein: Determining the personalized erythropoiesis model for the user includes: The previous PO2 information and the one or more previous hematocrit and / or hemoglobin measurements are input into a mathematical model for oxygen rationing to determine the personalized erythropoiesis model.

14. The method according to claim 12, wherein: The previous PO2 information indicates one or more hypoxic events experienced by the user during a previous time period, wherein each of the one or more hypoxic events indicates a PO2 level that is lower than a normoxic level.

15. The method according to claim 12, wherein: Using the individualized erythropoiesis model to determine the one or more predicted hematocrit and / or hemoglobin measurements comprises: generating a plurality of future PO2 information, wherein each of the plurality of future PO2 information indicates one or more hypoxic events to be performed for the user in a future time period; inputting the plurality of future PO2 information into the personalized erythropoiesis model to determine a plurality of predicted hematocrit and / or hemoglobin measurements; and Optimal future PO2 information is determined from the plurality of future PO2 information based on comparing the plurality of predicted hematocrit and / or hemoglobin measurement values ​​to desired hematocrit and / or hemoglobin measurement values ​​associated with the user.

16. The method according to claim 15, wherein: The method further comprises: determining a system training strategy for the user based on the optimal future PO2 information, and Wherein, executing the one or more actions includes displaying the system training strategy on a display device associated with the personalized erythropoiesis model generation computing platform.

17. The method according to claim 12, wherein: Determining the personalized erythropoiesis model for the user is based on the following equation: in, Indicates the rate of change of hypoxia-inducible factor (HIF) signaling activity, β h (θ) represents the formation rate of HIF complex, indicates that HIF signaling is upregulated in response to hypoxia, which reduces blood hemoglobin concentration, k h represents the decay rate of the HIF signal, Represents the concentration of oxygenated hemoglobin in the blood as measured by the kidneys, represents the physiological set point of oxygenated hemoglobin concentration in the blood, and h represents the HIF signal.

18. The method according to claim 12, wherein: Determining the personalized erythropoiesis model for the user is based on the following equation: in, represents the rate of change of EPO concentration, β e (θ) represents the basal erythropoietin (EPO) synthesis under steady-state conditions, represents additional EPO synthesis due to HIF signaling activation, and k e e represents the decay rate k e EPO decay.

19. The method according to claim 12, wherein: Determining the personalized erythropoiesis model for the user is based on the following equation: in, Indicates the rate of change of the total amount of red blood cells. represents the differentiation flux from the precursor cell population to the erythroid population, and K ery represents the cell flux due to apoptosis.

20. A computing platform for generating a personalized erythropoiesis model, comprising: one or more processors; as well as A non-transitory computer-readable medium having processor-executable instructions stored thereon, wherein the processor-executable instructions, when executed by the one or more processors, facilitate: obtaining individualized user data for a user, wherein the individualized user data indicates one or more previous hematocrit and / or hemoglobin measurements for the user and previous partial pressure of oxygen (PO2) information for the user; determining an individualized erythropoiesis model for the user based on the one or more previous hematocrit and / or hemoglobin measurements and the previous PO2 information, wherein the individualized erythropoiesis model indicates an individualized set of model parameters for the user; employing the individualized erythropoiesis model to determine one or more predicted hematocrit and / or hemoglobin measurements at one or more future time points; and One or more actions are performed based on the one or more predicted hematocrit and / or hemoglobin measurements.

Images