Smart footwear and system for identifying body events

Abstract

The present disclosure relates to systems and methods for identifying a physical event of a subject using a footwear garment having a pressure sensor. Examples of physical events that can be identified include a standing event, a weight-bearing event, a motion event, a body posture event, or a safety event.

Description

[0001] Cross-reference to related applications

[0002] This application claims the benefit of Provisional Application No. 63 / 559,280, filed on February 29, 2024, the entire contents of which are incorporated herein by reference in their entirety for all purposes. Technical Field

[0003] This disclosure generally relates to clothing with sensors and associated computer systems. More specifically, this disclosure relates to footwear and computer systems for recognizing bodily events of a user wearing the footwear. Background Technology

[0004] Wearable electronic devices and smart clothing are gaining popularity. These smart garments include sensors and other electronic components that can be used to collect a wide range of information about the user wearing the garment. Examples of such information include physiological data, such as the wearer's pulse rate and oxygen saturation. However, there remains a persistent need for improved smart clothing and associated systems for processing the data collected by smart clothing. In particular, there is a desire for such clothing and systems to accurately identify a wide range of information about the wearer. Summary of the Invention

[0005] The smart clothing, along with the associated computer systems and methods, as illustrated in the published examples, offers several advantages. For instance, it can efficiently and accurately provide useful insights into a wide range of activities and physiological conditions of the subjects.

[0006] One example is a method for operating a computing system including one or more processors to determine bodily events of a subject. Embodiments of this example may include: receiving data from a plurality of sensors, including data from a plurality of pressure sensors mounted on at least one article located at the bottom of the subject's foot, wherein the data includes first pressure data from a first pressure sensor at a anteromedial position of the foot, second pressure data from a second pressure sensor at an anterolateral position of the foot, and third pressure data from a third pressure sensor at a posterior position of the foot; processing the data by the one or more processors to determine bodily events of the subject, including: identifying one or more features associated with the subject's body based on the data from the plurality of sensors, the plurality of sensors including a plurality of pressure sensors mounted on at least one article; and classifying the one or more features as bodily events of the subject, wherein the bodily events are at least one of standing events, weight-bearing events, movement events, body posture events, and safety events.

[0007] In this example embodiment, identifying one or more features may include one or more of the following: (i) determining the sum of two or more of the first pressure data, second pressure data, and third pressure data, for example, to determine the total pressure on the subject's foot; and / or (ii) determining the difference between two or more of the first pressure data, second pressure data, and third pressure data, for example, to determine the pressure distribution on the subject's foot. Identifying one or more features may also include one or more of the following: (i) determining the i and / or ii above at a point in time, for example, static pressure determination; and / or (ii) determining the i and / or ii above over a non-zero time period, for example, dynamic pressure determination or pressure determination over time.

[0008] In any or all of the above embodiments of this example, the method may further include receiving weight data representing the weight of a subject by one or more processors; identifying one or more features including determining weight, optionally including summing first pressure data, second pressure data, and third pressure data; and classifying one or more features including classifying one or more features as standing events when the determined weight represents at least as much as the weight of the subject, optionally including tilting events.

[0009] In any or all of the above embodiments of this example, the method may further include receiving weight data representing the weight of a subject by one or more processors, and identifying one or more features including determining weight, optionally including summing first pressure data, second pressure data, and third pressure data; and classifying one or more features including classifying one or more features as a weight-bearing event when the determined weight represents a weight greater than the weight of the subject.

[0010] In any or all of the above embodiments of this example, receiving data from multiple sensors may include: receiving first article data from multiple pressure sensors mounted to an article, the article being located at the base of a subject's first foot; and receiving second article data from multiple pressure sensors mounted to an article, the article being located at the base of a subject's second foot; identifying one or more features includes identifying changes over time between the first article data and the second article data; and classifying one or more features includes classifying one or more features as motion events based on the identified changes over time between the first article data and the second article data. In embodiments, classifying one or more features may include classifying one or more features as walking events, jogging / running events, cycling events, or dancing events.

[0011] For example, in an embodiment, identifying one or more features may include identifying one or more gait parameters, wherein the one or more gait parameters include one or more of the following: speed or rate, stride length, stride duration, support time / percentage, swing time / percentage, step time, stride width, stride asymmetry, or ground reaction force / pressure.

[0012] In any or all of the above embodiments of this example, identifying one or more features may include determining the difference between two or more of the first pressure data, second pressure data, and third pressure data; and classifying one or more features includes classifying one or more features as body posture events based on the determined difference between two or more of the first pressure data, second pressure data, and third pressure data. In embodiments, classifying one or more features as body posture events may include classifying one or more features as a forward-leaning posture, a backward-leaning posture, or a side-leaning posture.

[0013] In this example embodiment, classifying one or more features may include classifying one or more features as security events.

[0014] In any or all of the above embodiments of this example, receiving data may include receiving data from multiple sensors mounted to at least one of socks, insoles, or footwear articles such as shoes or boots. In embodiments, for example, receiving data may further include receiving data from one or more additional sensors, wherein the one or more additional sensors include one or more of an accelerometer, PPG sensor, light sensor, cardiac sensor, position sensor, bioimpedance sensor, EMG sensor, GPS sensor, environmental sensor, bending sensor, and tension sensor mounted to the article. For example, identifying features and / or classifying features includes identifying features and / or classifying features based on data received from one or more additional sensors.

[0015] In any or all of the above embodiments of this example, data processing may include processing the data through a trained model. In embodiments, for example, data processing may include processing the data through a trained artificial neural network. Embodiments may further include using body event calibration data to train the trained model.

[0016] Another example includes one or more processors configured to perform the methods of any or all of the exemplary embodiments described above.

[0017] Another example is one or more garments, such as socks, insoles, or footwear such as shoes or boots, which are configured to provide sensor data for any or all of the exemplary embodiments described above. Attached Figure Description

[0018] Figure 1 This is a schematic diagram of a system according to an embodiment, which includes smart footwear and a computer system, the smart footwear being shown as socks for illustrative purposes.

[0019] Figure 2 Exemplary pressure data provided by a pressure sensor during a walking event is shown according to an embodiment.

[0020] Figure 3 Exemplary pressure data provided by a pressure sensor during a jogging / running event, according to an embodiment, is shown.

[0021] Figure 4 Exemplary pressure data provided by a pressure sensor during a bicycle riding event is shown according to an embodiment.

[0022] Figure 5 Exemplary pressure data provided by a pressure sensor during walking events, jogging / running events, and cycling events, according to an embodiment, are shown.

[0023] Figures 6A to 6C Exemplary acceleration and pressure data provided by sensors during a walking event are shown according to an embodiment.

[0024] Figures 7A to 7C Exemplary acceleration and pressure data provided by sensors during a jogging / running event, according to an embodiment, are shown.

[0025] Figures 8A to 8C Exemplary acceleration and pressure data provided by sensors during a bicycle riding event are shown according to an embodiment.

[0026] Figures 9A to 9C Exemplary acceleration and pressure data provided by sensors during a gait in a walking event, according to an embodiment, are shown.

[0027] Figure 10 This is a schematic diagram of a method for training a model to recognize bodily events according to an embodiment.

[0028] Figure 11 This is a schematic diagram of a subject during a standing or tilting body event, according to an embodiment.

[0029] Figure 12 This is a schematic diagram of a subject during a dancing physical event, according to an embodiment.

[0030] Figure 13 This is a schematic diagram of a subject during a standing weight-bearing physical event according to an embodiment.

[0031] Figure 14 This is a schematic diagram of a subject during a walking tilting body event, according to an embodiment.

[0032] Figure 15 This is a schematic diagram of a subject during a squatting or crouching physical event, according to an embodiment.

[0033] Figure 16 This is a schematic diagram of a subject during a physical event while cycling, according to an embodiment.

[0034] Figure 17 This is a schematic diagram of a subject during a seated physical event according to an embodiment.

[0035] Figure 18 This is a schematic diagram of a subject during a walking event or a jogging / running physical event, according to an embodiment.

[0036] Figure 19 This is a schematic diagram of a subject during an upright standing physical event according to an embodiment.

[0037] Figure 20 This is a schematic diagram of a method for identifying bodily events based on an embodiment of the usage model.

[0038] Figure 21 According to embodiments such as Figure 1 A detailed schematic diagram of the computer system shown. Detailed Implementation

[0039] All cited patent and non-patent literature disclosures are incorporated herein by reference in their entirety.

[0040] The terms “embodiment” or “disclosure” as used herein are not intended to be limiting, but are generally applicable to any embodiment defined in the claims or described herein. These terms are used interchangeably herein.

[0041] Unless otherwise disclosed, the terms “a” and “an” as used herein are intended to cover one or more (i.e., at least one) referenced features.

[0042] Those skilled in the art will more readily understand the features and advantages of this disclosure by reading the following detailed description. It should be understood that, for clarity, certain features of this disclosure described above and below in conjunction with different embodiments may also be provided in combination with a single element. Conversely, for brevity, various features of this disclosure described in conjunction with a single embodiment may also be provided individually or in any sub-combination. Furthermore, unless the context clearly indicates otherwise, references to the singular may also include plural forms (e.g., "a" and "an" may refer to one or more).

[0043] Unless otherwise expressly stated, the numerical values ​​used in the various ranges specified in this application are expressed as approximate values, as the minimum and maximum values ​​within the range are preceded by the word "approximately". In this way, numerical variations slightly above or slightly below the range can be used to obtain substantially the same results as the numerical values ​​within the range. Furthermore, the disclosure of these ranges is intended to represent a continuous range including every value between the minimum and maximum values.

[0044] Description of various embodiments

[0045] Those skilled in the art will readily understand that various aspects of this disclosure can be implemented by any number of methods and apparatuses configured to perform the intended functions. It should also be noted that the accompanying drawings referenced in this disclosure are not necessarily drawn to scale but may be exaggerated to illustrate various aspects of this disclosure, and in this respect, the drawings should not be construed as limiting.

[0046] Figure 1 This is a schematic diagram of a system 10 for identifying and classifying bodily events of a subject according to an embodiment. As shown, system 10 includes smart footwear 12 coupled to a computer system 14. The smart clothing 12 includes an article and a plurality of sensors 18, the article being shown, for example, as socks 16, configured to be worn on a human or other subject (…). Figure 1 Multiple sensors 18 are mounted or otherwise positioned on or near the feet of a sock or subject (not shown in the image) on the body. Although in Figure 1 Socks 16 are shown for illustrative purposes, but other embodiments of the smart garment 12 may include shoes, boots, sandals, or other footwear, or insoles or other components configured to be inserted into or otherwise attached to such footwear. As detailed below, sensor 18 generates sensor data associated with a subject wearing socks 16. Figure 1 As shown, sensor 18 includes multiple pressure sensors, such as, for example, pressure sensor 20A, pressure sensor 20B, and pressure sensor 20C (collectively referred to as pressure sensor 20). In the illustrated embodiment, sensor 18, including pressure sensor 20, is coupled to data transmission device 22 via transmission channel 24. Transmission channel 24 can be a wired channel or a wireless channel. Data transmission device 22 can be a wired data transmission device or a wireless data transmission device, transmitting sensor data (including pressure data from pressure sensor 20) from sensor 18 to computer system 14 via network 26. In some embodiments, electrical conductors (not shown) may couple one or more of the multiple sensors 18 to another or more of the multiple sensors and / or computer system 14. Although in Figure 1The data transmission device 22 is shown as a component of the sock 16, but in other embodiments, the data transmission device may be mounted in other locations on the subject's body (e.g., on other clothing elements), integrated into the associated sensor 18, and / or may be located outside the subject.

[0047] Computer system 14 processes sensor data (including pressure data from pressure sensor 20) to identify and classify certain physical and / or physiological activities of a subject wearing smart clothing 12. Embodiments of computer system 14 may determine certain physical events of the subject. Examples of such physical events include one or more of standing events, weight-bearing events, movement events, body posture events, or safety events. For example, standing events may include tilting (e.g., leaning forward, backward, or sideways) events. For example, movement events may include walking events, jogging / running events, cycling events, or dancing events. For example, body posture events may include sitting events, tilting events, squatting events, or crouching events. Safety events may include overloading and / or potentially harmful body postures. Physical events may also include several different physical events (e.g., tilting while standing).

[0048] As described in more detail below, in embodiments, a physical event can be determined by identifying features associated with the subject's body based on data from sensor 18 (including pressure data from pressure sensor 20) and classifying these features into one or more different types of physical events. In some embodiments, computer system 14 uses a trained machine learning model to process the sensor data to identify and / or classify the subject's physical events. In other embodiments, computer system 14 compares the sensor data with stored physical event data representing physical events to identify and classify physical events, or otherwise determine physical events. The sensor data provided by smart clothing 12 can also be used by computer system 14 for certain setup operations, such as generating calibration data for training models and other data and / or generating stored physical event data for determining physical events. These types of calibration data, along with the trained models, effectively provide a digital representation or model of the associated physical event or other physical or physiological activity of the subject.

[0049] The sock 16 has several different portions or regions located adjacent to the subject's foot. For example, sensor 20A is located on the inside of the sock 16 (e.g., the right side of the left sock) and at the front to provide pressure data on the front or inner side of the subject's foot. Sensor 20B is shown located on the outside of the sock 16 (e.g., the left side of the left sock) and at the front to provide pressure data on the front or outer side of the subject's foot. Sensor 20C is shown located at the back of the sock 16 to provide pressure data on the back or heel of the subject's foot. Embodiments of the sock 16 may include additional sensors 18. For example, Figure 1 The illustrated embodiments include motion sensors 30A and 30B, a galvanic sensor 32, an environmental condition sensor 34, an electrocardiogram (ECG or EKG) cardiac sensor 36, a photoplethysmography (PPG) sensor 38, an impedance sensor 40, and a stretch sensor 42. Other embodiments include alternative or additional sensors, such as, for example, implantable sensors, optical sensors, location (e.g., GPS) sensors, electromyography (EMG) sensors, and / or flexure sensors. Sensor 18 may be positioned inside the sock 16 to contact the subject's skin, positioned outside the sock, and / or embedded in the sock's material or other substrate.

[0050] As described above, pressure sensor 20 is positioned to provide data representing pressure applied to different areas of the subject's foot. Motion sensors 30A and 30B (collectively referred to as motion sensors 30) may include an inertial measurement unit (IMU), which may have an accelerometer, gyroscope, and / or magnetometer, and can provide motion data or information such as acceleration data, x-axis, y-axis, and / or z-axis motion data, and direction or heading data. In the illustrated embodiment, motion sensor 30A is positioned below the subject's ankle, and motion sensor 30B is positioned above the ankle. Electrodermal sensor 32 provides data representing skin condition. Environmental condition sensor 34 provides data representing conditions such as temperature or humidity. Electrocardiogram sensor 36 provides data related to the electrical performance or characteristics of the subject's heart. Photoplethysmography sensor 38 provides data representing physiological characteristics such as blood, heart rate, blood oxygen saturation, and blood pressure. Impedance sensor 40 provides information representing the bioimpedance of the subject's skin or other tissues. The stretch sensor 42 (e.g., shown adjacent to the heel and the lower back of the sock 16) can provide data indicating changes in length or bending angle. The sensor 18 can be a commercially available device or other known device.

[0051] although Figure 1Only one article, such as sock 16 (left foot sock), is shown in the illustration, but other embodiments include a second article coupled to computer system 14 via network 26, or a second smart garment, such as smart clothing 12, for example, a right foot sock (not shown). For example, such a second sock or other article may be substantially the same as or similar to sock 16 and include a sensor 20, such as a pressure sensor configured to provide data or information representing pressure at various foot locations associated with the foot, including the forefoot or inner side of the foot, the forefoot or outer side of the foot, and the heel. The sensor data provided by such a second smart garment can improve the efficiency and accuracy of bodily event determination.

[0052] Although sensor 18 is shown at a specific location on sock 16 and a corresponding location on the subject's body, in other embodiments, sensor 18 is positioned at other locations on the sock or on the subject's body. For example, the location of sensor 18 may be determined based on factors such as optimizing the provided signal strength, the correlation between sensor data and events (such as bodily events expected to be identified and classified), and the comfort and fit of sock 16. While one sensor is shown for many of several different types for illustrative purposes, and more than one sensor is shown in the cases of pressure sensor 20 and motion sensor 30, other embodiments include more or fewer sensors 18. As an example, sensor 18 can be attached to sock 16 by means and structures such as pockets, adhesives, stitching, and / or hook and loop fasteners. In embodiments, for example, sensor 18 may be incorporated into a sensor harness such as that described in co-pending U.S. Provisional Application No. 63 / 442,886, filed February 2, 2023, entitled Electronic harness for smart clothing, or co-pending U.S. Application No. 17 / 940,507, filed September 8, 2022, entitled Clothing including electronic devices, both of which are incorporated herein by reference in their entirety for all purposes.

[0053] In embodiments, one or more or all of the sensors 18 are wireless devices configured to transmit associated sensor data to a data transmission device 22 on the sock 16 via transmission channel 24. As an example, embodiments of the system according to this disclosure may include a 0.5-inch diameter FSR 400 series force sensor (e.g., powered by a 3.3V supply voltage) available from Interlink Electronics. A voltage divider circuit may be incorporated between the microprocessor and the sensor, allowing the optimal voltage to be determined via the analog input of the microprocessor. Laminated pressure / force sensors that can be incorporated into these types of systems are generally known and commercially available. For example, these types of devices may vary resistance according to changes in shape and / or pressure. Additionally or alternatively, capacitive force / pressure sensors, also known and commercially available, may be incorporated into the system according to this disclosure.

[0054] Additionally or alternatively, the data transmission device 22 may be located adjacent to the sensor 18, such as, for example, a mobile phone or mobile device. In embodiments, one or more or all of the sensors 18 are wireless devices configured to transmit associated sensor data directly to the computer system 14 via transmission channel 24. In embodiments of the smart garment 12, the data transmission device 22 may include electronic components configured to be coupled to one or more or all of the sensors 18, for example, via a detachable connector plug (not shown). Such electronic components may be configured to be coupled to the sensor 18 to facilitate electrical and mechanical connections of the electronic components and to disconnection of the electronic components from the sensor 18. For example, the electronic components may include a wireless transmitter to transmit sensor data from the sensor 18 to the computer system 14 via network 26. Alternatively or additionally, the electronic components may include electronic memory that stores sensor data from the sensor 18 for download or otherwise transmission to the computer system 14. Exemplary connector plugs and electronic components of these types are disclosed above by reference to U.S. Provisional Application No. 63 / 442,886, which is incorporated herein by reference.

[0055] In embodiments, the data transmission device 22 may also transmit data from the computer system 14 to one or more or all of the sensors 18. In embodiments, the data transmission device 22 may include an electronic module comprising a processor, battery, antenna, and / or memory, and may be configured to provide all or part of the processing capabilities of the computer system 14 (e.g., to perform the methods 100 and 300 described below). The electronic components may be housed in a waterproof housing and removably attached to the sock 16, for example, via one or more pockets, hook-and-loop patches, adhesives, or fasteners. For example, the above-described structure and manner for removably holding a data transmission device such as the data transmission device 22 to the sock 16, including pockets on a strap, is disclosed by reference to U.S. Provisional Application No. 63 / 442,886, incorporated herein by reference. Advantages of removably holding all or part of the data transmission device 22 to the sock 16 include wash isolation and reusability of the sock.

[0056] Although sensor 18 is described above as being mounted to or located on sock 16, or otherwise configured for use in conjunction with sock 16, other embodiments of system 10 include sensors configured for use in conjunction with other structures. For example, auxiliary sensors may be mounted to or located on a detachable ankle or leg strap, or mounted to or located on trousers (not shown) worn or positioned on the subject.

[0057] Figure 2 Exemplary pressure data or information is shown provided by pressure sensors such as pressure sensor 20A, pressure sensor 20B, and pressure sensor 20C on the subject's foot (e.g., located at positions corresponding to the medial front, lateral front, and heel) during walking. For example, Figure 2 The data can represent the normal walking pace on some uneven ground. Figure 3 Exemplary pressure data or information is shown provided by pressure sensors such as pressure sensor 20A, pressure sensor 20B, and pressure sensor 20C on the subject's foot while the subject is jogging / running. Figure 3 The data can represent normal or typical slow running cadence on sidewalks, flat surfaces, uphill surfaces, and downhill surfaces. Figure 4 Exemplary pressure data or information is shown provided by pressure sensors such as pressure sensor 20A, pressure sensor 20B, and pressure sensor 20C on the subject's feet while the subject is riding a bicycle. Figure 5 Exemplary pressure data or information provided by pressure sensors such as pressure sensor 20A, pressure sensor 20B, and pressure sensor 20C on the subject's feet are shown during walking, jogging / running, and cycling. Figures 2 to 5It is evident that the pressure data from sensors 20A, 20B, and 20C differ from each other during each of the walking, jogging / running, and cycling events. Figures 2 to 5 It is also evident that the stress data from sensor 20 differs between walking events, jogging / running events, and cycling events (for example, the stress data during jogging / running events and cycling events differs from the stress data during walking events).

[0058] Figures 6A to 6C Exemplary and corresponding ankle acceleration data from motion sensors such as motion sensor 30A and / or motion sensor 30B during a walking event are shown, as well as pressure data from pressure sensor 20A, pressure sensor 20B and / or pressure sensor 20C. Figure 6A Ankle acceleration is expressed by axis (without considering the specific orientation of the sensor). Figure 6B express Figure 6A The ankle accelerations shown are combinations or averages, such as the sum of squares. Figure 6C This represents the corresponding pressure data from each of sensors 20A, 20B, and 20C. For example, Figures 6A to 6C The data can represent the normal walking pace on certain uneven terrain. Among other things, from... Figures 6A to 6C The data shown clearly shows the stride and the clear signal of flat foot landing.

[0059] Figures 7A to 7C Exemplary and corresponding ankle acceleration data from motion sensors such as motion sensor 30A and / or motion sensor 30B, and pressure data from pressure sensor 20A, pressure sensor 20B, and / or pressure sensor 20C are shown during a jogging / running event. Figure 7A Ankle acceleration is expressed by axis (without considering the specific orientation of the sensor). Figure 7B express Figure 7A The ankle accelerations shown are combinations or averages, such as the sum of squares. Figure 7C This represents the corresponding pressure data from each of sensors 20A, 20B, and 20C. For example, Figures 7A to 7C The data can represent a normal or typical slow running cadence on sidewalks, flat surfaces, uphill surfaces, and downhill surfaces. Among other things, from... Figures 7A to 7C The data shown clearly demonstrates a distinct signal of gait.

[0060] Figures 8A to 8CExemplary and corresponding ankle acceleration data from motion sensors such as motion sensor 30A and / or motion sensor 30B during a cycling event are shown, as well as pressure data from pressure sensor 20A, pressure sensor 20B and / or pressure sensor 20C. Figure 8A Ankle acceleration is expressed by axis (without considering the specific orientation of the sensor). Figure 8B express Figure 8A The ankle accelerations shown are combinations or averages, such as the sum of squares. Figure 8C This represents the corresponding pressure data from each of sensors 20A, 20B, and 20C. For example, Figures 8A to 8C The data can represent a relatively slow pace or effort. Among other things, from Figures 8A to 8C The data shown clearly reveals a clear signal of rhythm.

[0061] Figures 9A to 9C Detailed, exemplary, and corresponding ankle acceleration data from motion sensors such as motion sensor 30A and / or motion sensor 30B, and pressure data from pressure sensor 20A, pressure sensor 20B, and / or pressure sensor 20C are shown during several steps of a jogging / running event. Figure 9A Ankle acceleration is expressed by axis (without considering the specific orientation of the sensor). Figure 9B express Figure 9A The ankle accelerations shown are combinations or averages, such as the sum of squares. Figure 9C This indicates the corresponding pressure data from each of sensors 20A, 20B, and 20C. Among other information, from... Figures 9A to 9C The data shown clearly shows a clear signal of the outer side of the foot landing.

[0062] Certain gait or stride parameters / positions can be determined and used in conjunction with the systems and methods described herein. These gait parameters may include one or more of the following: speed, number of steps, stride length, stride duration, stance time / percentage, swing time / percentage, stride duration, stride width, stride asymmetry, and ground reaction force / pressure. For example, a subject's stride can be characterized by a series of positions of one of the subject's legs and feet during the gait cycle, including heel strike, full foot strike, mid-stance, push-off, acceleration, mid-swing, and deceleration. The gait cycle of this foot can be defined as the distance between the positions of the heel at two consecutive heel strike positions. Stride length can be defined as the distance between the positions of the left and right heels at consecutive heel strike positions (e.g., approximately half the gait cycle). The stride angle of each foot can be defined as the angle between the midline of the subject's foot (e.g., from toes to heel) and the centerline of the path the subject is moving along. The stride width of each foot can be defined as the distance between the centerline of the path and its position on the midline of the foot (e.g., at the toes).

[0063] Embodiments of computer system 14 use one or more models to identify features of bodily events associated with sensor data (provided by sensors 18 of a subject wearing smart clothing 12) and classify the identified features as representing one or more event types (such as specific types of bodily events). In some examples, the one or more models are machine learning models. In some examples, the one or more models are artificial neural networks. Alternatively or additionally, in some examples, the one or more models are statistical models. Based on sensor values ​​corresponding to or characterizing bodily events, these models can be effectively used as digital representations of bodily events. Sensor data received from sensor 18 is applied to and / or compared with models to identify features and classify features as associated with one or more bodily events. In embodiments, for example, computer system 14 includes models for characterizing the following bodily events and associated bodily event features: (1) standing events; (2) weight-bearing events; (3) movement events; (4) body posture events; and / or (5) safety events.

[0064] In an embodiment, computer system 14 combines bodily event features with a model to identify and classify bodily events. Bodily event features may be portions of sensor data from one or more sensors 18 that effectively define a specific type or aspect of a bodily event. For example, the amplitude or intensity of the sensor data may define different bodily events. The relative timing of activities defined by the sensor data may define different characteristics of the bodily event individually (e.g., from a single sensor) or in combination with activities or inactivity defined by other sensors. In an embodiment, one or more bodily event features may be defined by the activities of multiple sensors 18, including, for example, a pressure sensor 20.

[0065] Figure 10 This is a schematic diagram of a method 100 for training a machine learning model to identify bodily events. According to an embodiment, the model uses bodily event features to classify and describe certain bodily events. As shown, method 100 includes: a process 102 for collecting a calibration dataset or training dataset; a process 104 for filtering the training dataset; a process 106 for contextualizing the training dataset; a process 108 for analyzing the training dataset to determine bodily event features; a process 110 for providing the training dataset and features to the machine learning model for training; a process 112 for generating predicted bodily events; a process 114 for comparing the predicted bodily events with stored, actual, and / or labeled bodily events associated with the training data; a process 116 for adjusting the parameters of the machine learning model; and a process 118 for determining whether the training of the machine learning model has been completed. Although Figure 10 A set of selected procedures for method 100 is shown, but various alternatives, modifications, and variations are possible. For example, some procedures may be extended and / or combined. Other procedures may be inserted. Some procedures may be optional and not required to be performed. According to embodiments, the order of these procedures may be interchanged and other procedures may be substituted. For example, some or all of procedures 102, 104, 106, 108, 110, 112, 114, 116, and 118 may be executed by a computing device or by a processor under the guidance of instructions stored in memory. As an example, some or all of the procedures of method 100 may be executed according to instructions stored in a non-transitory computer-readable medium. It should be understood that although procedures 108 and 110 are ordered and describe training on classic supervised machine learning model types such as KNN or decision trees, this order may be interchanged for artificial neural networks or deep learning methods. While process 108 identifies bodily event features as variables in a classic model, in the case of convolutional neural networks, contextualized data can be presented directly to the classifier to determine features such as those in hidden or pooling layers.

[0066] According to some embodiments, in process 102, one or more calibration datasets or training datasets are collected for each of one or more bodily events. For example, each of the training datasets includes sensor data (including data from pressure sensor 20) relating to bodily event characteristics of the associated bodily event and knowledge of the types of these bodily events (e.g., labeling the training dataset using the types of the associated bodily events). One or more training datasets may be stored in a memory element as stored bodily event data for subsequent comparison with the subject's bodily events.

[0067] The calibration or training data collected at process 102 may be sensor data from sensor 18 when the sensor is in one or more known states, body postures, or body positions corresponding to a bodily event. For example, a known state or body posture may be a static state corresponding to a predetermined position of a subject's body wearing one or more socks 16 during a bodily event and associated physiological data. Alternatively or additionally, a known state may be a dynamic state or a time-varying state corresponding to the subject's body movements during an associated bodily event and associated physiological data.

[0068] For example, each of the subject's left and right feet in multiple gait or stride phases throughout the walking event can be identified. Specifically, the heel strike, full foot strike, mid-stance, push-off, acceleration, mid-swing, and deceleration phases of the right foot can be identified. Sensor data from sensor 18 (including data from each of the pressure sensors 20A, 20B, and 20C from one or both socks 16) can be determined and used as training data during these and other phases of the subject's walking event (e.g., the type of movement event).

[0069] When the subject is in a gait or stride position as described above, the sensor data collected from sensor 18 can be considered as reference position sensor data and can be used for calibration or reference. In embodiments, training data can also be collected from sensor 18 when the subject is in other static positions or postures, such as: (1) when the subject is standing upright and optionally not moving (e.g., standing event); (2) when the subject is holding or otherwise carrying an object (weight-bearing event); (3) when the subject is jogging / running or cycling (other examples of movement events); (4) when the subject (e.g., at the waist) is leaning, such as leaning forward, backward, or sideways, or sitting or lying down (body posture event); and / or (5) a combination of one or more of (1)-(4), such as a weight-bearing event (safety event) occurring simultaneously with a body posture event. In process 102, for each of the body events for which system 10 is configured to identify and classify, training data can also be collected from sensor 18 when the subject is in multiple static positions or postures. For example, for each physical event, calibration data (e.g., the physical event is represented by a series of static postures corresponding to the posture of the actual physical event) can be collected from sensor 18 when the subject is in two or more of a series of positions or postures that the subject typically has during the associated physical event. As another example, gait or stride parameter characteristics, such as those described above, can be determined based on one or more of the subject's socks 16 and used to identify features of the subject's physical events.

[0070] Computer system 14 can use training data collected during process 102 to characterize or “understand” the orientation of sensor 18 relative to the unique geometry of the subject or other subjects providing the training or calibration information. For example, when used for calibration purposes, the training data can be used to adjust for or compensate for a specific or unique fit of socks 16 on the subject. For applications using the absolute position of the subject, the subject may orient themselves in a known direction (such as, for example, north) during process 102. If the subject periodically recalculates to a known orientation, the accuracy of the training data generated during step 102 can be improved, for example, to compensate for drift in sensor 18.

[0071] The training data collected during process 102 may also include the subject's weight. When sensor 18 (including pressure sensor 20) provides information used by computer system 14 to determine that the weight represented by the pressure data is greater than the subject's weight, certain physical events (such as, for example, weight-bearing events and safety events) can be identified.

[0072] As an example relating to gait or stride position as described above, training data can be collected from one or more pressure sensors 20, including pressure sensors 20A-20C, when the subject is in the first phase (such as the gait cycle of a walking event completed during right foot deceleration). Training data can also be collected from one or more pressure sensors 20, including pressure sensors 20A-20C, during one or more subsequent phases of the walking event (such as heel strike, full foot strike, mid-stance, push-off, acceleration, and mid-swing of the left foot). For each of the different types of motion events, the sensor data from sensor 18 (including pressure data from pressure sensors 20) may differ, for example, in amplitude and timing. For example, jogging / running motion events and cycling events can be distinguished based on the different sensor data from pressure sensors 20. As another example, pressure data can be collected from pressure sensors 20 when the subject is instructed to perform one or more different types of physical events.

[0073] Figure 11-19 Subjects in various physical events are shown. For example, Figure 11 The illustration shows a subject in a body posture event such as a tilting event. The pressure data collected by the pressure sensor 20 on each of the two socks 16 may differ because the subject leans more heavily against one foot. The pressure data collected by the pressure sensor 20 on each sock 16 may also differ because the weight applied by the subject to either foot may not be uniformly distributed.

[0074] For example, Figure 12 The experiment illustrates a subject engaged in a movement event, such as dancing. The dynamic pressure data collected by the pressure sensor 20 on each of the two socks will differ and may vary in a time-varying manner to some extent. Similarly, the pressure data collected by the pressure sensor 20 on each sock 16 may vary in a time-varying manner to some extent.

[0075] For example, Figure 13The illustration shows a subject in a standing event and a weight-bearing event. Since the subject's weight is likely to be relatively evenly distributed across both feet, the pressure data collected by the pressure sensor 20 on each of the two socks can be approximately the same. For similar reasons, the pressure data collected by the pressure sensor 20 on each sock 16 can be distributed in a manner representing the upright and standing weight distribution on the relevant feet. The nature of the weight-bearing event can be determined by the weight measured by the pressure sensor 20 being greater than the subject's known weight. In some examples, the anterior-posterior weight distribution and / or lateral weight distribution relative to a centered or standing weight distribution, determined from the pressure sensor data, can indicate that the weight carried by the subject is located in front of, behind, or to the side of the subject (e.g., a forward-leaning, backward-leaning, or lateral weight-bearing event).

[0076] Additionally or alternatively, the subject's weight may not be relatively evenly distributed across both feet. In these cases, clothing and systems such as socks according to this disclosure may be advantageous. For example, the weight can be determined by summing information from all sensors on each foot, and then summing the information from both feet to correlate with a predetermined initial weight. Calibration may involve using a known weight and then standing on each foot separately. A predetermined weight or a known weight can be determined through calibration.

[0077] Clothing and systems such as socks according to this disclosure can also be configured to determine the degree or amount of weight a subject bears in front of, behind, to the side of, and / or near their body. Similarly, it can determine whether the subject's arms are extended. For example, the clothing and system can determine whether a subject weighing 200 pounds is carrying a weight of 25 pounds with their arms fully extended. As described below, this type of information can be used to provide a safety and / or risk or injury assessment associated with the subject's activities.

[0078] It can also be based on, for example, the above. Figure 13 Safety events are identified using data collected from subjects during the standing and weight-bearing events shown and described. For example, a risk score may be stored by computer system 14. The risk score may represent a range of severity or other importance criteria related to safety and may be characterized by the amount of weight loaded (e.g., relative to the subject's weight and / or other characteristics such as age and sex) and / or the amount of tilt (e.g., along one or more directions, such as front, back, or side) determined based on the weight distribution on one or both feet of the subject from pressure sensor 20.

[0079] For example, Figure 14The illustration depicts a subject walking with the aid of a walking device (e.g., a walking frame or cane) during a combined walking and tilting event. Features of pressure data from pressure sensor 20 during events such as the walking event described above, combined with pressure data features during events such as the tilting event described above, can be used to identify the walking and tilting event. Because the subject supports part of their weight on the walking frame, it can be determined that the subject is using a walking device (such as a walking frame) based on the characteristic that the sum of the weights determined by the pressure sensor data is less than the subject's known weight.

[0080] Walking events involving a subject using a cane can be determined in a similar manner. When a cane is used as a walking aid, since the subject may hold the cane with only one hand, thus using the cane to support the weight on the side of the subject's foot holding the cane, the pressure sensor data representing this walking event will be asymmetrical and periodic.

[0081] For example, Figure 15 The experiment illustrates a subject in a squatting or crouching posture event. Pressure sensor 20 collects relatively static data representing a degree of uneven weight distribution between the two feet, with weight asymmetrically distributed on each foot.

[0082] For example, Figure 16 This demonstrates that the subjects were involved in a cycling activity. (The above is combined with...) Figures 8A to 8C Sensor data representing bicycle riding events from pressure sensor 20 is described. Among other characteristics, the sensor data provided by pressure sensor 20 will be periodic, and can indicate higher pressure toward the front of the foot if the cyclist places the front of their foot on the bicycle pedal.

[0083] For example, Figure 17 The illustration shows a subject in a seated body posture event. Because the subject is seated and the chair is supporting their weight, the sensor data from pressure sensor 20 representing the seated event can be relatively static and can represent a weight less than the subject's weight, and in some examples, can represent a weight significantly less than the subject's weight.

[0084] For example, Figure 18 This shows that the subjects were in the midst of a walking or jogging / running activity. (The above is combined with...) Figures 6A to 6C as well as Figures 7A to 7C Sensor data representing walking events and jogging / running events from pressure sensor 20 are described. Among other characteristics, the sensor data provided by pressure sensor 20 will be periodic and can represent the pressure transferred from back to front on each foot.

[0085] For example, Figure 19The subject is shown in a standing posture event. Sensor data representing the standing event from pressure sensor 20 can be substantially static and relatively evenly distributed on each foot and between the two feet.

[0086] When the sensor is in a dynamic state, calibration data and / or training data can be collected from sensor 18 at process 102. For example, a subject wearing socks 16 may move through a range of postures (which correspond to any one or more bodily events configured to be classified by computer system 14), and sensor data collected from sensor 18 during the movement across this range of postures can be used to characterize the bodily event. Training data for any or all bodily events configured to be identified and classified by computer system 14 can be collected in process 102 in the manner described above.

[0087] Return to reference Figure 10 In process 104, all or part of the training data may be filtered. Conventional or other known filtering methods may be performed in process 104. For example, outlier data may be removed and / or a smoothing function may be applied (e.g., to dynamic state data). Other non-limiting examples of filtering that may be performed at process 104 include noise removal and / or bandpass filtering. Process 104 is optional, and in embodiments, only a portion of the training data (such as, for example, certain types of training data) may be filtered in process 104, or no filtering may be performed on the training data.

[0088] In process 106, all or part of the training data may be contextualized, or converted into contextual data. The data contextualized at process 106 (also referred to as contextual data) may include converting all or part of the training data into other forms that may be helpful to method 100. For example, in process 106, the training data may be converted into values ​​that are more relevant to humans. As another example, training data in the form of acceleration data, or training data in the form of acceleration and gyroscope data, may be converted into joint angles using methods such as Kalman calculations. Process 106 is optional, and in embodiments, only a portion of the training data (such as, for example, certain types of training data) may be contextualized in process 106, or no training data may be contextualized at all.

[0089] In process 108, the context-processed training data is evaluated to determine one or more features associated with a bodily event. For example, the amplitude of pressure data from one or more pressure sensors 20A-20C can be a bodily event feature used by a classical or deep learning machine learning model. Furthermore, the rate of change of pressure data provided by the pressure sensors 20A-20C of one sock 16, or the rate of change of pressure data between the pressure sensors of two socks, can also be a bodily event feature. Process 108 yields variables that may be used by one or more machine learning models to predict the classification of the bodily event. According to some embodiments, one or more bodily event features indicate or define the type of bodily event. For example, the identified bodily event features associated with pressure sensor 20 can represent any of the aforementioned bodily events. For these types of bodily events, at process 110, identified bodily event features associated with other sensors 18 can also be identified, which can be used alone or in combination with identified bodily event features associated with pressure sensor 20.

[0090] At process 110, one or more training datasets are provided to train the machine learning model. In an example where the training data has been filtered and / or contextualized via process 104, the filtered and / or contextualized training data may be provided to the machine learning model in process 108. In process 108, labels or other information for identifying the types of bodily events associated with the training data are also provided to the machine learning model. For example, the machine learning model may be a decision tree network, a logistic regression classifier, an artificial neural network, a convolutional neural network, a recurrent neural network, a modular neural network, or any other suitable type of machine learning model. The machine learning model may analyze each training dataset to determine one or more features associated with the bodily events and process 110.

[0091] In process 112, the machine learning model generates a predicted body event based on the event features identified in process 110. For example, when generating the predicted event, the machine learning model calculates one or more parameters (e.g., weight values ​​associated with each connection layer) related to one or more event features. Regarding the above embodiment, the predicted body event generated in process 112 can be one of the aforementioned body events. Alternatively, the predicted body event generated in process 112 can be a body event of undetermined type, or it can be no body event at all.

[0092] In process 114, the predicted physical events are compared with the actual types or physical events to determine the accuracy of the predicted physical events. In some embodiments, a loss function or cost function using the set of training data is used to determine the accuracy.

[0093] In process 116, the machine learning model adjusts one or more parameters related to the bodily event features based on this comparison. For example, one or more parameters may be adjusted to reduce (e.g., minimize) the loss function or cost function.

[0094] In process 118, it is determined whether training has been completed. For example, training for a training dataset can be completed when the loss function or cost function of that training dataset is sufficiently reduced (e.g., minimized). As an example, training of a machine learning model is complete when training achieves an acceptable level of accuracy between the predicted labels and known labels for one or more datasets. In an embodiment, if process 118 determines that training of the machine learning model has not yet been completed, method 100 iteratively returns to process 108 until training is determined to be complete. In an embodiment, if process 118 determines that training of the machine learning model has been completed, method 100 stops, as shown in process 120. The trained machine learning model effectively possesses existing knowledge about which bodily event features are expected to or helpful in identifying and classifying bodily events.

[0095] While the method 100 described above trains a machine learning model for the purpose of identifying multiple different types of body events, in other embodiments, each of one or more machine learning models may be trained for the purpose of identifying one of multiple different types of body events, or more than one but fewer than all body events. In other words, the machine learning model may be trained by method 100 for the purpose of identifying and classifying one or more types of body events.

[0096] Figure 20 This is a schematic diagram of an exemplary method 300 for identifying and classifying or determining bodily events according to an embodiment. As shown, method 300 includes: a process 302 for providing sensor data to one or more models, a process 304 for processing the sensor data by the models, and a process 306 for determining or identifying bodily events by the one or more models. In some embodiments, the model is a statistical model. Alternatively or additionally, one or more models or all of the models may be machine learning models, such as, for example, those combined as described above. Figure 10The described method 100 trains an artificial neural network. Although method 300 is described as using a selected set of procedures, many alternatives, modifications, and variations are possible. For example, some procedures may be extended and / or combined. Other procedures may be inserted. According to embodiments, the order of these procedures may be interchanged and other procedures may be substituted. For example, some or all of the procedures of method 300 are executed by a computing device or by a processor under the guidance of instructions stored in memory. As an example, some or all of the procedures of the method are executed according to instructions stored in a non-transitory computer-readable medium.

[0097] In process 302, sensor data from smart garment 12 is provided to the model. For example, sensor data is collected from sensor 18 (including pressure sensor 20). In some embodiments, the sensor data includes at least sensor data from pressure sensor 20A, pressure sensor 20B, and pressure sensor 20C.

[0098] In process 304, the sensor data is processed in a manner similar to processes 104, 106 and 108, that is, process 304 filters and contextualizes the sensor data and generates features from the sensor data.

[0099] In process 306, the model determines, or identifies and classifies, bodily events based on the processed sensor data. For example, in an embodiment, the model identifies or classifies the sensor data as representing at least one of the bodily events described herein.

[0100] Other embodiments use other methods to identify and classify bodily events based on sensor data. For example, during calibration operations of the type described above, bodily event features or other bodily event data associated with different types of bodily events can be identified (e.g., through a trained machine model or human labeling) and stored in association with the bodily events. In practice, each of the different types of bodily events can be defined, characterized, or represented by a set of one or more bodily event features specific to the bodily event. Sensor data received from sensors 18 on the smart garment 12 (including sensor data from pressure sensors 20, such as pressure sensors 20A, 20B, and 20C) can then be compared with the stored bodily event features to identify and classify the bodily events represented by the sensor data. Feature extraction, pattern recognition, and other processing methods can be used in conjunction with such methods for identifying and classifying bodily events.

[0101] Figure 21This is a block diagram illustrating the physical components (e.g., hardware) of an exemplary computer system 14 according to an embodiment. In a basic configuration, the computer system 14 may include at least one processing unit 802 and system memory 804. Depending on the configuration and type of the computer system 14, the system memory 804 may include, but is not limited to, volatile memory (e.g., random access memory), non-volatile memory (e.g., read-only memory), flash memory, or any combination of these memories. The system memory 804 may include an operating system 805 and one or more program modules 806, such as sensing and processing units 820.

[0102] For example, operating system 805 may be adapted to control the operation of computer system 14. Furthermore, embodiments of this disclosure may be implemented in conjunction with graphics libraries, other operating systems, or any other applications, and are not limited to any particular application or system. Basic configuration in Figure 21 The components within the dashed lines 808 are illustrated. Computer system 14 may have additional features or functions. For example, computer system 14 may also include additional data storage devices (removable and / or non-removable). Figure 21 In the diagram, such additional storage is illustrated by removable storage device 809 and non-removable storage device 810.

[0103] As described above, several program modules and data files may be stored in system memory 804. When program module 806 (e.g., sensing and processing unit 820) is executed on processing unit 802, it may perform processes including but not limited to those described herein, such as processing sensor data from sensor 18 and processing methods 100 and 300.

[0104] Furthermore, embodiments of this disclosure can be implemented in circuits including discrete electronic components, packaged or integrated electronic chips containing logic gates, circuits utilizing microprocessors, or on a single chip containing electronic components or a microprocessor. For example, embodiments of this disclosure can be implemented via a system-on-a-chip (SoC), wherein... Figure 21The various components or components shown may be integrated onto a single integrated circuit. Such a SoC device may include one or more processing units, graphics units, communication units, system virtualization units, and various application functions, all of which are integrated (or “programmed”) onto a chip substrate as a single integrated circuit. When operating via the SoC, the client switching protocol capabilities described herein can be operated via specific application logic integrated with other components of the computer system 14 on the single integrated circuit (chip). Embodiments of this disclosure may also be implemented using other techniques capable of performing logical operations such as, for example, AND, OR, and NOT operations, including but not limited to mechanical, optical, fluidic, and quantum technologies. Furthermore, embodiments of this disclosure may be implemented in a general-purpose computer or any other circuit or system.

[0105] Computer system 14 may also have one or more input devices 812, such as visual image sensors, audio sensors, sound or voice input devices, touch or swipe input devices, etc. It may also include one or more output devices 814, such as displays, speakers, etc. The foregoing devices are merely examples, and other devices may be used. Computer system 14 may include one or more communication connections 816, thereby allowing communication with other computing devices 850 (e.g., computing device 128 and / or computing device 130). Examples of suitable communication connections 816 include, but are not limited to, radio frequency (RF) transmitters, receivers, and / or transceiver circuitry; universal serial buses (USB), parallel ports, and / or serial ports.

[0106] As used herein, the term computer-readable medium may include computer storage media. Computer storage media may include volatile and non-volatile, removable and non-removable media implemented by any method or technique for storing information such as computer-readable instructions, data structures, or program modules. System memory 804, removable storage device 809, and non-removable storage device 810 are examples of computer storage media (e.g., memory storage). Computer storage media may include RAM, ROM, electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, optical storage devices, magnetic data storage devices, or any other article of manufacture that can be used to store information and is accessible by computer system 14. Any such computer storage media may be part of computer system 14. Computer storage media does not include carrier waves or other propagated or modulated data signals.

[0107] Communication media can be embodied in computer-readable instructions, data structures, program modules, or other data in modulated data signals (such as carrier waves or other transmission mechanisms), and include any information transmission medium. The term "modulated data signal" describes a signal having one or more characteristics that are set or altered in a manner that encodes information into the signal. By way of example and not limitation, communication media can include wired media such as wired networks or direct wired connections, and wireless media such as acoustic, radio frequency (RF), infrared, and other wireless media.

[0108] Other aspects of the smart clothing, systems, methods, and techniques disclosed herein can be used in conjunction with the systems and methods described in U.S. Provisional Application No. 63 / 539,182 (“182 Application”), filed September 19, 2023, entitled “System and Method for Physical Demand Characterization”, which is incorporated herein by reference in its entirety for all purposes. As an example, the smart clothing, systems, and methods of this disclosure can be used to generate an assessment of the physical demands of the work and its tasks disclosed in “182 Application”. The capabilities of the systems described in “182 Application” can be enhanced by incorporating information of the type described herein that can be generated by the smart clothing, and inferences that can be drawn from that information. Additionally or alternatively, the capabilities of the techniques described in this disclosure can be enhanced by combining the features described in “182 Application” with the clothing, systems, and methods of this disclosure.

[0109] The invention described herein has been generally and specifically illustrated above. It will be apparent to those skilled in the art that various modifications and variations can be made to the embodiments without departing from the scope of this disclosure. Therefore, this disclosure is intended to cover various modifications and variations of the invention, provided that such modifications and variations are within the scope of the appended claims and their equivalents.

Claims

1. A method for operating a computing system including one or more processors to determine bodily events of a subject, the method comprising: The processor receives data from multiple sensors, including data from multiple pressure sensors mounted on at least one article located at the bottom of the subject's foot, wherein the data includes first pressure data from a first pressure sensor located at the anteromedial position of the foot, second pressure data from a second pressure sensor located at the anterolateral position of the foot, and third pressure data from a third pressure sensor located at the posterior position of the foot. The data is processed by the one or more processors to determine the subject's bodily events, including: Based on the data from the plurality of sensors, one or more features associated with the subject's body are identified, the plurality of sensors including the plurality of pressure sensors mounted to the at least one article; and The one or more features are classified as physical events of the subject, wherein the physical events are at least one of standing events, weight-bearing events, movement events, body posture events, and safety events.

2. The method according to claim 1, characterized in that, Identifying the one or more features includes one or more of the following: (i) Determine the sum of two or more of the first pressure data, the second pressure data, and the third pressure data, for example, to determine the total pressure on the subject's foot; and / or (ii) Determine the difference between two or more of the first pressure data, the second pressure data, and the third pressure data, for example, to determine the pressure distribution on the subject's foot.

3. The method according to claim 2, characterized in that, Identifying the one or more features includes one or more of the following: i. Determine the above i and / or ii at a certain time point, e.g., static pressure determination; and / or ii. Determine i and / or ii above within a non-zero time period, such as dynamic pressure determination or pressure determination that varies over time.

4. The method according to any one of claims 1-3, characterized in that: The method further includes receiving weight data representing the weight of the subject by the one or more processors; Identifying one or more features includes determining weight, and optionally includes summing the first pressure data, the second pressure data, and the third pressure data; as well as Classifying the one or more features includes classifying the one or more features as standing events when the determined weight represents at least as much as the subject's weight, optionally including tilting events.

5. The method according to any one of claims 1-4, characterized in that: The method further includes receiving weight data representing the weight of the subject by the one or more processors; and Identifying one or more features includes determining weight, and optionally includes summing the first pressure data, the second pressure data, and the third pressure data; as well as Classifying the one or more features includes classifying the one or more features as a weight-bearing event when the determined weight represents a weight greater than the subject's weight.

6. The method according to any one of claims 1-5, characterized in that: Data received from multiple sensors includes: Multiple pressure sensors attached to the bottom of the first article receive data from the first article; and Data from the second article is received from multiple pressure sensors of the article installed at the bottom of the subject's second foot; Identifying the one or more features includes: identifying the changes over time between first product data and second product data; and Classifying the one or more features includes classifying the one or more features as motion events based on the identified changes over time between the first product data and the second product data.

7. The method according to claim 6, characterized in that, Classifying the one or more features includes classifying the one or more features into walking events, jogging / running events, cycling events, or dancing events.

8. The method according to any one of claims 6-7, characterized in that, Identifying the one or more features includes identifying one or more gait parameters, wherein the one or more gait parameters include one or more of the following: speed or rate, stride length, stride duration, support time / percentage, swing time / percentage, step duration, stride width, stride asymmetry, or ground reaction force / pressure.

9. The method according to any one of claims 1-8, characterized in that: Identifying one or more features includes determining the difference between two or more of the first pressure data, the second pressure data, and the third pressure data; and Classifying the one or more features includes classifying the one or more features as body posture events based on a determined difference between two or more of the first pressure data, the second pressure data, and the third pressure data.

10. The method according to claim 9, characterized in that, Classifying the one or more features as body posture events includes classifying the one or more features as forward-leaning posture, backward-leaning posture, or side-leaning posture.

11. The method according to claim 1, characterized in that, Classifying the one or more features includes classifying the one or more features as security events.

12. The method according to any one of claims 1-11, characterized in that, Receiving the data includes receiving the data from at least one of the sensors installed in socks, insoles, or footwear such as shoes or boots.

13. The method according to claim 12, characterized in that, Receiving the data also includes receiving data from one or more additional sensors, wherein the one or more additional sensors include one or more of an accelerometer, PPG sensor, light sensor, cardiac sensor, position sensor, bioimpedance sensor, EMG sensor, GPS sensor, environmental sensor, bending sensor, and tensile sensor mounted on the article.

14. The method according to claim 13, characterized in that, Identifying features and / or classifying features includes: identifying features and / or classifying features based on data received from the one or more additional sensors.

15. The method according to any one of claims 1-14, characterized in that, Processing the data includes processing the data through a trained model.

16. The method according to claim 15, characterized in that, Processing the data includes processing the data through a trained artificial neural network.

17. The method according to any one of claims 15-16, further comprising using body event calibration data to train the trained model.

18. A computer system comprising one or more processors configured to perform the method of any one of claims 1-17.

19. One or more garments, such as socks, insoles, or footwear such as shoes or boots, said garments being configured to provide sensor data according to any one of claims 1-14.

Images