Systems and methods for monitoring and treating diabetic foot ulcers

Abstract

In some aspects, a computer-implemented method for collecting and processing sensor data to identify a risk of affecting or causing skin damage is disclosed. The computer-implemented method may include: receiving sensor feature data via a computer network, the sensor feature data representing output of a user sensor configured to be worn on a limb of a user; generating activity classification model output data using the sensor feature data and an activity classification model, the activity classification model output data representing a likelihood that the sensor feature data corresponds to each of a plurality of different activity classifications; determining an activity classification from the activity classification model output data; and transmitting display data representing the activity classification to a computing device configured to present the display data via the computer network.

Description

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This application claims priority to UK Provisional Application Nos. 1907260.2 and 1907254.5 filed on May 23, 2019, and U.S. Provisional Application No. 62 / 889955 filed on August 21, 2019, the disclosures of which are incorporated herein by reference in their entirety. Technical Field

[0003] Embodiments of the present disclosure relate to apparatus, systems, and methods for managing wounds utilizing a user activity monitoring device. Background Art

[0004] Pressure ulcers, which may also be called pressure sores, bedsores or decubitus ulcers, are damage to the skin and underlying tissues caused by prolonged pressure on the skin, soft tissue, muscle or bone above the capillary filling pressure (approximately 32 mmHg).

[0005] One type of pressure sore that develops on the foot is called a diabetic foot ulcer (DFU), which tends to occur with greater frequency and intensity in the diabetic population. Management and treatment of diabetic foot ulcers requires offloading the diabetic foot ulcer through the use of cushioned footwear, such as supportive boots, casts, shoes, etc. While offloading can be effective, it has been found that noncompliance or nonuse of offloading devices can play a significant role in delayed healing of diabetic foot ulcers.

[0006] Prior art methods and systems provide little or no information about an individual's lifestyle and compliance with unloading devices. A deeper understanding of an individual's lifestyle may be important for preventing and healing pressure ulcers. However, due to these limitations, prevention and healing of pressure ulcers using prior art methods and systems may be delayed or worse, worsen, leading to prolonged discomfort, hospitalization, or even surgery. Summary of the invention

[0007] In some aspects, a system is disclosed for collecting and processing sensor data to identify events that are at risk of affecting an existing ulcer on a limb or at risk of causing a new ulcer on a limb. The system may include a memory device and a processor. The memory device may store sensor data collected by a first sensor configured to monitor the movement or orientation of a user and a second sensor configured to monitor the movement or orientation of a load-reducing device. The processor may: determine from a first set of sensor data that a load force is applied to the user's limb that is at risk of affecting an existing ulcer on the limb or at risk of causing a new ulcer on the limb; add a first event to a first list, the first event indicating that the load force is applied to the limb; determine from a second set of sensor data that (i) the load force is not applied to the limb, (ii) the user is engaged in an activity for more than a threshold duration, and (iii) the user is not wearing a load-reducing device; and add a second event to a second list, the second event indicating that (i) the load force is not applied to the limb, (ii) the user is engaged in an activity for more than a threshold duration, and (iii) the user is not wearing a load-reducing device.

[0008] The system described in the previous paragraph may include one or more of the following features: the processor may transmit a subset of the first list and the second list to a computing device via a computer network for presentation on a display. The first sensor may include a first accelerometer and a first magnetometer, the second sensor may include a second accelerometer and a second magnetometer, and the sensor data includes sensor data from the first accelerometer, the second accelerometer, the first magnetometer, and the second magnetometer. The processor may: add a first event to the first list in response to determining that a load force is applied to a limb; and add a second event to the second list in response to determining that (i) the load force is not applied to the limb, (ii) the user participates in the activity for more than a threshold duration, and (iii) the user is not wearing a load-reducing device. The processor may: compare the acceleration value of the first set of sensor data with an acceleration threshold to determine that the load force is applied to the limb; and compare the acceleration value of the second set of sensor data with an acceleration threshold to determine that the load force is not applied to the limb. The activity may include standing, walking, or driving a vehicle.

[0009] In some aspects, a non-transitory computer-readable medium is disclosed, on which an application is stored, which, when executed by a processor of an electronic device, can cause the processor to: receive sensor data collected by a first sensor and a second sensor via a computer network, wherein the first sensor is configured to monitor the movement or orientation of a user and the second sensor is configured to monitor the movement or orientation of a load-reducing device; determine from the first set of sensor data that a load force is applied to the user's limb that risks affecting an existing ulcer on the limb or risks causing a new ulcer on the limb; add a first event to a first list, the first event indicating that the load force is applied to the limb; determine from the second set of sensor data that (i) the load force is not applied to the limb, (ii) the user is engaged in an activity for more than a threshold duration, and (iii) the user is not wearing a load-reducing device; and add a second event to a second list, the second event indicating that (i) the load force is not applied to the limb, (ii) the user is engaged in an activity for more than a threshold duration, and (iii) the user is not wearing a load-reducing device.

[0010] The non-transitory computer-readable medium described in the previous paragraph may include one or more of the following features: when the application is executed by the processor, it may cause the processor to transmit a subset of the first list and the second list to a computing device via a computer network. The first sensor may include a first accelerometer and a first magnetometer, the second sensor may include a second accelerometer and a second magnetometer, and the sensor data includes sensor data from the first accelerometer, the second accelerometer, the first magnetometer, and the second magnetometer. When the application is executed by the processor, it may cause the processor to: compare the acceleration value of the first set of sensor data with an acceleration threshold to determine that a load force is applied to a limb; and compare the acceleration value of the second set of sensor data with an acceleration threshold to determine that a load force is not applied to a limb. The acceleration threshold may be at least 18 m / s 2. The application, when executed by the processor, may cause the processor to add a time to the first list, the time indicating when the limb experienced the load force based on at least the first set of sensor data. The application, when executed by the processor, may cause the processor to add a duration to the first list, the duration indicating how long the limb experienced the load force based on at least the first set of sensor data. The activity may be an activity other than lying with the user's limb elevated. The limb may be a leg. The computer network may be or include a cellular communication network. The first list and the second list may be the same list, and identify multiple events that are at risk of affecting an existing ulcer or at risk of causing a new ulcer, the multiple events including the first event and the second event. The application, when executed by the processor, may cause the processor to compare the acceleration value over the activity duration with an acceleration threshold to determine that the user's participation in the activity exceeds the threshold duration, the activity duration being indicated by the second set of sensor data. The first set of sensor data may correspond to first sensor data collected by the first sensor and the second sensor over a first time period, and the second set of sensor data may correspond to second sensor data collected by the first sensor and the second sensor over a second time period, the second time period not overlapping the first time period. The application, when executed by the processor, may cause the processor to: in response to determining from the first set of sensor data that the load force is applied to the limb, add the first event to the first list; in response to determining from the second set of sensor data that (i) the load force is not applied to the limb, (ii) the user is engaged in an activity for more than the threshold duration, and (iii) the user is not wearing a load-reducing device, add the second event to the second list; determine from a third set of sensor data that a load force that has a risk of affecting an existing ulcer on the limb or a risk of causing a new ulcer on the limb is applied to the limb, the third set of sensor data being different from the first set of sensor data; in response to determining from the third set of sensor data that a load force that has a risk of affecting an existing ulcer on the limb or a risk of causing a new ulcer on the limb is applied to the limb The three sets of sensor data determine that the load force is applied to the limb and add a third event to the first list, the third event being different from the first event; determine from the fourth set of sensor data that (i) the load force is not applied to the limb, (ii) the user is engaged in an activity for more than the threshold duration, and (iii) the user is not wearing a load-reducing device, the fourth set of sensor data being different from the second set of sensor data; and add a fourth event to the second list, the fourth event indicating that (i) the load force is not applied to the limb, (ii) the user is engaged in an activity for more than the threshold duration, and (iii) the user is not wearing a load-reducing device, the fourth event being different from the second event. When the user wears the load-reducing device, the load-reducing device can distribute the applied force to the limb away from an existing ulcer on the limb. The load-reducing device can be a removable cast.

[0011] In some aspects, a method is disclosed. The method may include: receiving sensor data collected by a first sensor and a second sensor via a computer network, the first sensor being configured to monitor the movement or orientation of a user, the second sensor being configured to monitor the movement or orientation of a load-reducing device; determining, by a processor, from the first set of sensor data that a load force that is at risk of affecting an existing ulcer on the limb or at risk of causing a new ulcer on the limb is applied to the user's limb; adding, by the processor, a first event to a first list, the first event indicating that the load force is applied to the limb; determining, by the processor from the second set of sensor data, that (i) the load force is not applied to the limb, (ii) the user is engaged in an activity for a duration, and (iii) the user is not wearing a load-reducing device; and adding, by the processor, a second event to a second list, the second event indicating that (i) the load force is not applied to the limb, (ii) the user is engaged in an activity for a duration, and (iii) the user is not wearing a load-reducing device.

[0012] The method of the preceding paragraph may further comprise transmitting a subset of the first list and the second list to a computing device via a computer network for presentation on a display.

[0013] In some aspects, a system for collecting and processing sensor data to identify risks of affecting or causing skin damage is disclosed. The system may include a computer-readable memory and one or more processors. The system may: obtain sensor feature data including a first set of feature data and a second set of feature data, the first set of feature data representing outputs of a user sensor configured to be worn on a user's limb, the second set of feature data representing outputs of a load-reducing device sensor coupled to a load-reducing device, the load-reducing device configured to be worn on the limb; determine an activity classification using the first set of feature data and an activity classification model; determine a load-reducing usage classification using the first set of feature data, the second set of feature data, and a load-reducing usage model; generate display data representing at least one of the activity classification or the load-reducing usage classification; and transmit the display data to a computing device configured to present the display data.

[0014] The system described in the previous paragraph may include one or more of the following features: an activity classification model may determine the extent to which a first set of feature data corresponds to each of a plurality of different activity classifications. A load-relief usage model may determine the extent to which a movement or orientation of a user sensor corresponds to a movement or orientation of a load-relief device sensor. The skin injury may be a diabetic foot ulcer or a venous leg ulcer. The user sensor may include a first accelerometer and a first magnetometer, the load-relief device sensor may include a second accelerometer and a second magnetometer, the sensor feature data representing outputs from the first accelerometer, the second accelerometer, the first magnetometer, and the second magnetometer. The system may also be configured to: determine a subsequent activity classification using a third set of feature data and the activity classification model, the third set of feature data representing outputs of the user sensor; and determine from the activity classification and the subsequent activity classification that the user's activity has changed. The system may also be configured to: determine a subsequent load-reducing usage classification using a third set of feature data, a fourth set of feature data, and a load-reducing usage model, wherein the third set of feature data represents the output of the user sensor and the fourth set of feature data represents the output of the load-reducing device sensor; and determine from the load-reducing usage classification and the subsequent load-reducing usage classification that the user's use of the load-reducing device has changed. The system may also be configured to: generate activity model input data using the first set of feature data; generate activity classification model output data using the activity model input data and the activity classification model; and identify from the activity classification model output data a data element having a value that satisfies a classification criterion to determine an activity classification, the activity classification being associated with the data element. The plurality of different activity classifications may include a motion classification, a classification with no motion but with static weight, a classification with no motion but with force change, and a lying classification. The activity classification model may include a decision tree, a neural network, or a support vector machine. The load-reducing device may be a removable cast. The activity classification model may include a first decision tree configured to assign a first set of feature data to one of a plurality of different activity classifications, and the offload usage model may include a second decision tree configured to determine whether the movement or orientation of the user sensor corresponds to the movement or orientation of the offload device sensor.

[0015] In some aspects, a computer-implemented method for collecting and processing sensor data to identify a risk of affecting or causing skin damage is disclosed. The computer-implemented method may include: receiving sensor characteristic data via a computer network, the sensor characteristic data representing output of a user sensor configured to be worn on a user's limb; generating activity classification model output data using the sensor characteristic data and an activity classification model by one or more computer processors; determining an activity classification from the activity classification model output data by one or more computer processors; and transmitting data representing the activity classification to a computing device via the computer network, the computing device being configured to present the display data.

[0016] The computer-implemented method described in the previous paragraph may include one or more of the following features: the activity classification model output data may indicate the likelihood that the sensor feature data corresponds to each of a plurality of different activity classifications. The plurality of different activity classifications may include a first activity classification indicating movement of the limb and a second activity classification indicating no movement of the limb but weight is loaded on the limb. The computer-implemented method may also include identifying, by the one or more computer processors, a risk of affecting or causing a skin injury, the skin injury being a diabetic foot ulcer or a venous leg ulcer. The determining of the activity classification may include determining that the activity classification model output data indicates that the sensor feature data may correspond to a third activity classification indicating no movement of the user's limb but force changes. The determining includes determining that the activity classification model output data indicates that the sensor feature data may correspond to a fourth activity classification indicating that the user is lying down. The computer-implemented method may also include determining, by the one or more computer processors, a subsequent activity classification using second sensor feature data and an activity classification model, the second sensor feature data indicating an output of the user's sensor; and determining, by the one or more computer processors from the activity classification and the subsequent activity classification, that the user's activity has changed. The computer-implemented method may also include: receiving second sensor characteristic data via a computer network, the second sensor characteristic data representing an output of a load-reducing device sensor coupled to a load-reducing device, the load-reducing device configured to be worn on the limb; and determining, by one or more computer processors, a load-reducing usage classification using the sensor characteristic data, the second sensor characteristic data, and a load-reducing usage model, wherein the load-reducing usage model is configured to determine the extent to which the motion or orientation of the user sensor corresponds to the motion or orientation of the load-reducing device sensor. The computer-implemented method may also include: receiving second sensor characteristic data via a computer network, the second sensor characteristic data representing an output of a load-reducing device sensor coupled to a load-reducing device, the load-reducing device configured to be worn on the limb; determining, by one or more computer processors, a subsequent load-reducing usage classification using the second sensor characteristic data and the load-reducing usage model, the second sensor characteristic data representing an output of the load-reducing device sensor; and determining, by one or more computer processors, from the load-reducing usage classification and the subsequent load-reducing usage classification, that a user's use of the load-reducing device has changed. The computer-implemented method may also include loading, by the one or more computer processors, the activity classification model, the activity classification model comprising a decision tree, a neural network, or a support vector machine. The computer-implemented method may further include loading, by the one or more computer processors, the activity classification model, the activity classification model comprising a decision tree configured to assign the sensor feature data to one of the plurality of different activity classifications.

[0017] In some aspects, a non-transitory computer-readable medium having stored thereon an application for collecting and processing sensor data to identify a risk of affecting or causing skin damage is disclosed. The application, when executed by one or more processors of a computing device, can cause the computing device to: receive sensor characteristic data via a computer network, the sensor characteristic data representing output of a user sensor configured to be worn on a user's limb; generate activity classification model output data using the sensor characteristic data and an activity classification model; determine an activity classification from the activity classification model output data; and transmit display data representing the activity classification to a second computing device via the computer network, the second computing device being configured to present the display data.

[0018] The non-transitory computer-readable medium described in the previous paragraph may include one or more of the following features: the activity classification model output data may represent the likelihood that the sensor feature data corresponds to each of a plurality of different activity classifications. The plurality of different activity classifications may include a first activity classification representing movement of the limb and a second activity classification representing no movement of the limb but weight being loaded on the limb. The plurality of different activity classifications may include a third activity classification representing no movement of the limb but a change in force on the limb. The plurality of different activity classifications may include a third activity classification representing that the user is lying down. The activity classification model may include a decision tree, a neural network, or a support vector machine. The activity classification model may include a decision tree configured to assign the sensor feature data to one of a plurality of different activity classifications. The application, when executed by one or more computer processors, may cause the computing device to: determine a subsequent activity classification using second sensor feature data and an activity classification model, the second sensor feature data representing an output of a user sensor; and determine changes that have occurred in the user's activity from the activity classification and the subsequent activity classification. The application, when executed by the one or more computer processors, may cause the computing device to: receive second sensor characteristic data via a computer network, the second sensor characteristic data representing an output of a load-reducing device sensor coupled to a load-reducing device, the load-reducing device being configured to be worn by the limb; and determine a load-reducing usage classification using the sensor characteristic data, the second sensor characteristic data, and a load-reducing usage model, wherein the load-reducing usage model is configured to determine the extent to which the motion or orientation of the user sensor corresponds to the motion or orientation of the load-reducing device sensor. The application, when executed by the one or more computer processors, may cause the computing device to: receive second sensor characteristic data via a computer network, the second sensor characteristic data representing an output of a load-reducing device sensor coupled to a load-reducing device, the load-reducing device being configured to be worn by the limb; determine a subsequent load-reducing usage classification using the second sensor characteristic data and the load-reducing usage model, the second sensor characteristic data representing an output of the load-reducing device sensor; and determine from the load-reducing usage classification and the subsequent load-reducing usage classification that a user's usage of the load-reducing device has changed.

[0019] In some aspects, a system for collecting and processing sensor data to identify a risk of affecting or causing skin damage is disclosed. The system may include a computer-readable memory and one or more processors. The system may: obtain sensor feature data including a first set of feature data and a second set of feature data, the first set of feature data representing an output of a user sensor configured to be worn on a user's limb, the second set of feature data representing an output of a load-reducing device sensor coupled to a load-reducing device, the load-reducing device configured to be worn on the limb; determine an activity classification using the first set of feature data and an activity classification model, wherein the activity classification model may include a first decision tree, the first decision tree is configured to assign the first set of feature data to one of a plurality of different activity classifications; determine a load-reducing use classification using the first set of feature data, the second set of feature data, and a load-reducing use model, wherein the load-reducing use model may include a second decision tree, the second decision tree is configured to determine whether a motion or orientation of the user sensor corresponds to a motion or orientation of the load-reducing device sensor; generate display data representing at least one of the activity classification or the load-reducing use classification; and transmit the display data to a computing device configured to present the display data.

[0020] In some aspects, a system for collecting and processing sensor data to identify risks of affecting or causing skin damage is disclosed. The system may include a computer-readable memory and one or more processors. The system may: obtain sensor feature data including a set of feature data, the set of feature data representing outputs of a user sensor configured to be worn on a user's limb; determine an activity classification using the set of feature data and an activity classification model, wherein the activity classification model may include a decision tree configured to assign the set of feature data to one of a plurality of different activity classifications; generate display data representing the activity classification; and transmit the display data to a computing device configured to present the display data. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Embodiments of the present disclosure will now be described hereinafter, by way of example only, with reference to the accompanying drawings, in which:

[0022] Figure 1 An example user activity monitoring system is shown, which includes an activity monitoring device and a load reduction monitoring device;

[0023] Figure 2 Shows Figure 1 Example components of an activity monitoring device;

[0024] Figure 3 Shown include Figure 1 Example computing environments for activity monitoring devices and load reduction monitoring devices;

[0025] Figure 4 It shows that Figure 3 Example data processing performed within a computing environment of;

[0026] Figure 5 It shows that Figure 3 an event logging process executed within the computing environment of; and

[0027] Figure 6 Shows the available Figure 3 An example computer system for implementing one or more devices or systems within a computing environment. DETAILED DESCRIPTION

[0028] Introduction to User Activity Monitoring

[0029] For various reasons, including wound prevention and monitoring, the user's activity may be desirably monitored by an activity monitoring device. In one example, the user's activity may be monitored when the user may be susceptible to injury or already has an injury, such as a pressure sore. Information about the user's activity collected by the activity monitoring device may be useful in helping prevent or treat a pressure sore. Additionally, information about activity collected by the activity monitoring device may be used to check compliance with a treatment regimen.

[0030] Some aspects disclosed herein relate to wound monitoring or treatment for human or animal bodies. Therefore, any reference to wounds herein may refer to wounds on human or animal bodies, and any reference to body herein may refer to human or animal bodies. The disclosed technology may relate to preventing or minimizing damage to physiological tissue or living tissue, or to the treatment of damaged tissue (e.g., wounds described herein).

[0031] As used herein, the expression "wound" may include damage to living tissue that may be caused by cutting, blows or other impacts, usually skin being cut or broken. Wounds may be chronic or acute injuries. Acute wounds are caused by surgery or trauma. They pass through the healing phase within a predicted time frame. Chronic wounds usually start as acute wounds. When acute wounds do not follow the healing phase, acute wounds can become chronic wounds, resulting in prolonged recovery. It is believed that the transition from acute wounds to chronic wounds can be attributed to the patient being immunocompromised.

[0032] Chronic wounds may include, for example: venous ulcers (such as those occurring in the legs), which account for the majority of chronic wounds and primarily affect the elderly; diabetic ulcers (e.g., foot or ankle ulcers); peripheral arterial disease; pressure sores or epidermolysis bullosa (EB).

[0033] Examples of other wounds include, but are not limited to, abdominal wounds or other large or open wounds resulting from surgery, trauma, sternotomy, fasciotomy, or other conditions, dehisced wounds, acute wounds, chronic wounds, subacute and dehisced wounds, traumatic wounds, flaps and skin grafts, lacerations, abrasions, contusions, burns, diabetic ulcers, pressure sores, ostomies, surgical wounds, traumatic ulcers, and venous ulcers.

[0034] The wound may include deep tissue injury. Deep tissue injury is a term proposed by the National Pressure Ulcer Advisory Panel (NUPAP) to describe a unique form of pressure ulcer. These ulcers have been described by clinicians over the years using terms such as purple pressure ulcers, ulcers that may worsen, and bruises over bony prominences.

[0035] Wounds can also include tissue that is at risk of becoming a wound as discussed herein. For example, tissue at risk can include tissue on a bony prominence (at risk of deep tissue injury / damage) or pre-operative tissue (e.g., knee tissue) that may be cut (e.g., for joint replacement / surgical alteration / reconstruction).

[0036] Some aspects relate to methods of monitoring or treating wounds using the techniques disclosed herein in combination with one or more of the following: advanced footwear, rotation of the patient, clearance (e.g., clearance of diabetic foot ulcers), treatment of infection, systemic integration, antimicrobial, antibiotics, surgery, removal of tissue, affecting blood flow, physiotherapy, exercise, bathing, nutrition, hydration, nerve stimulation, ultrasound, electrical stimulation, oxygen therapy, microwave therapy, active ozone, antibiotics, antimicrobial, etc.

[0037] Alternatively or additionally, the wound may be treated using topical negative pressure or conventional advanced wound care that is not assisted by the use of applied negative pressure (also referred to as non-negative pressure therapy).

[0038] Although the present disclosure may refer to pressure sores, foot ulcers, etc., the systems and methods disclosed herein may be used to prevent, monitor, or treat any type of skin injury or wound, such as a venous leg ulcer.

[0039] User activity monitoring system

[0040] Figure 1 A user activity monitoring system 100 is shown including an activity monitoring device 120 attached to a body part 110. The activity monitoring device 120 may be attached to the body part 110 using a strap, adhesive, or other coupling mechanism, and may be worn on or supported by the body.

[0041] The body part 110 may be a limb of a user, such as a leg including a knee 112 and a foot 114. As shown, the activity monitoring device 120 may be supported by the body part 110 at a position between the knee 112 and the foot 114, such as near the knee 112. In other aspects, the activity monitoring device 120 may be supported by another part of the body part 110, such as above the knee 112 or elsewhere. The activity monitoring device 120 may monitor and record the activity of the body part 110 (e.g., walking, jumping, sitting, lying, running, squatting or standing), for example, from the position, movement or orientation of the activity monitoring device 120 or one or more other sensors of the activity monitoring device 120. The activity monitoring device 120 may be used, for example, for load monitoring of the load of the foot 114. In some embodiments, multiple body parts may be monitored by the activity monitoring device 120, and different sensors may be used to monitor different body parts. Each sensor of the activity monitoring device 120 or each sensor communicating with the activity monitoring device may be referred to as a user sensor, or sometimes referred to as a user motion sensor.

[0042] A body part 110 is shown wearing and partially covered by a load-reducing device 130, such as a removable cast. The load-reducing device 130 can support the body part 110 and reduce pressure on the foot 114 when the user can stand or participate in other activities. When the load-reducing device is worn, the load-reducing device 130 can distribute the applied forces to the body part 110 away from injury. A load-reducing monitoring device 132 can be attached to the load-reducing device 130. The load-reducing monitoring device 132 can be the same as or similar to the activity monitoring device 120 and monitors and records the activity of the load-reducing device 130. The load-reducing monitoring device 132 can be supported by the load-reducing device 130 using straps, adhesives, or other coupling mechanisms. The load-reducing monitoring device 132 can be attached to an inner or outer surface of the load-reducing device 130. Although Figure 1 1, but the unloading monitoring device 132 can be attached to an unloading device (e.g., a cane or a walker) that is not worn by the user. In addition, the activity monitoring device 120 can be worn regardless of whether the unloading device 130 is wearable. The various sensors of the unloading monitoring device 132 or the various sensors that communicate with the unloading monitoring device can be referred to as unloading device sensors, or sometimes unloading device motion sensors.

[0043] The user activity monitoring system 100 may additionally or alternatively include one or more of the activity monitoring device 120 or the unloading monitoring device 132 at other locations, such as at a location supported by the unloading device 130, another part of the body part 110, another device that is not worn, such as a cane or a walker, or at other locations. One or more additional or alternatives of the activity monitoring device 120 or the unloading monitoring device 132 may be the same as or similar to the activity monitoring device 120 and may monitor and record the activity of the unloading device 130, another part of the body part 110, or the body.

[0044] Figure 2 Example components of an activity monitoring device 120 are shown. The activity monitoring device 120 may include a housing 200, a controller 202, a memory device 204, a user interface 206, a power source 208, one or more sensors 210, and a communication interface 212 configured to, for example, electrically communicate with each other. The power source 208 may provide power to one or more components of the activity monitoring device 120.

[0045] One or more components of the activity monitoring device 120 may be contained in or supported by the housing 200. The housing 200 may be comprised of a top portion and a bottom portion that are sealed together, and the top portion or the bottom portion may be hard or soft. The housing 200 may be flexible and have mechanical structures and design features that provide shoulder pin keyway alignment of components within the housing 200. The housing 200 may support a circuit board within it, and one or more components of the activity monitoring device 120 may be positioned on the circuit board.

[0046] The housing 200 can be prepared by stereolithography (SLA) or polymer jetting photopolymer 3D printing materials or by 3D printing an engineering resin with a Shore hardness of 80A. The housing 200 may include an elastomer, a thermoplastic elastomer, or be constructed by injection molding. The molded portion of the housing 200 may be made of white liquid silicone rubber. Adhesives (e.g., adhesives for attaching plastic and elastomeric materials) can be used to glue the top and bottom portions of the housing 200 together, and a wide range of other adhesives (e.g., cyanoacrylates, silicones, epoxy resins, hydrogels, hydrocolloids, sealant systems) or other techniques (e.g., using double-sided adhesive tape, ultrasonic welding, riveting) can be used.

[0047] Controller 202 may control the operation of one or more other components of activity monitoring device 120 (e.g., memory device 204, user interface 206, power source 208, one or more sensors 210, or communication interface 212) based at least on instructions stored in memory device 204. For example, controller 202 may control monitoring of body weight loading of body part 110 or positioning of body part 110 and record data indicating the loading of body part 110 or positioning of body part 110 to memory device 204.

[0048] The user interface 206 may include one or more output elements, such as a visual feedback device (e.g., a light emitting diode), a tactile feedback device, or an audio device (e.g., a speaker) that provides user output to the user. The one or more output elements may convey status information to the user, such as whether the activity monitoring device 120 is operating successfully or whether communication with another device has been successfully configured. The user interface 206 may include one or more input elements, such as buttons, switches, dials, touch pads, microphones, or touch screens, for receiving user input for configuring the activity monitoring device 120. In some aspects, the user interface 206 may have no more than one user input element, such as a button, for receiving user input to activate and deactivate the activity monitoring device 120 or to perform one or more other functions.

[0049] One or more sensors 210 may be used to detect and monitor movement or orientation of activity monitoring device 120 or other characteristics of activity monitoring device 120 or other characteristics around activity monitoring device 120. One or more sensors 210 may be used to detect and monitor activity of a user of activity monitoring device 120, including, for example, load or positioning of body part 110. One or more sensors 210 may include one or more accelerometers, gyroscopes, magnetometers, impedance sensors, thermistors, pressure sensors, or optical sensors, among other types of sensors. One or more sensors 210 may be positioned near body part 110 by the housing, or may be remote from body part 110 but may be used to monitor characteristics of body part 110.

[0050] Controller 202 and one or more sensors 210 may be used to determine a plurality of features indicating movement or orientation of activity monitoring device 120 over a window of data. For example, controller 202 may use an accelerometer (which senses x-axis, y-axis, and z-axis acceleration) and a magnetometer (which senses x-axis, y-axis, and z-axis magnetic field changes) of one or more sensors 210 to determine one or more or all of the data features listed in Table 1, including for each window of data, e.g., over a time period of 1, 2, 5, 10, 20, or 30 seconds. Data features determined from accelerometer data may include "Acc" in their names, and data features determined from magnetometer data may include "Mag" in their names.

[0051]

[0052]

[0053] Table 1

[0054] The data features in Table 1 can be calculated, for example, according to at least the equations provided in Table 2.

[0055]

[0056] Table 2

[0057] The tGravity_vector data feature in Table 1 may indicate the relative gravity direction of activity monitoring device 120 when body part 110 may be upright, such as when a user of activity monitoring device 120 is standing. The tGravity_vector data feature may be used to account for placement or orientation of housing 200 relative to body part 110 (and potential changes in size or shape of body part 110) without performing a calibration step on activity monitoring device 120 that ensures a specific placement or orientation of housing 200 relative to gravity. The tGravity_vector data feature may be calculated from a plurality of gravity vectors detected by one or more sensors 210 over time, such as periodically (e.g., every 0.2, 0.5, 1, 2, or 5 seconds) from an average of a plurality of detected gravity vectors (e.g., a moving average of the past 2, 5, 10, 20, 50, 200, or 500 vectors), when it may be determined that the user is walking (e.g., the user may be determined to be walking by controller 202 from detecting one or more strides or high accelerations using one or more sensors 210). The value of the tGravity_vector data feature may be reset when the position of activity monitoring device 120 relative to body part 110 changes, such as when activity monitoring device 120 is separated from body part 110 .

[0058] Tables 3 and 4 together illustrate example calculations of the tGravity_vector data feature. Table 3 includes 20 example data samples, including for each sample a detected x, y, and z components of a gravity vector, and for each sample an indication of whether the user was walking at the time of the sample.

[0059] sample x y z walk? 1 -0.512 -1.100 0.150 no 2 -0.100 -0.900 0.170 no 3 -0.320 -0.800 0.120 no 4 -0.660 -0.910 0.150 no 5 -0.830 -0.750 0.150 no 6 -0.512 -1.020 0.170 yes 7 0.100 -0.900 0.120 yes 8 -0.320 -0.800 0.150 yes 9 -0.660 -0.910 0.150 yes 10 -0.200 -0.750 0.170 yes 11 -0.200 -1.020 0.120 yes 12 -0.100 -0.800 0.150 yes 13 -0.320 -0.910 0.150 no 14 -0.660 -0.750 0.170 no 15 -0.830 -1.020 0.120 no 16 -0.512 -0.900 0.150 no 17 -0.100 -0.800 0.150 no 18 -0.320 -0.910 0.170 no 19 -0.660 -0.800 0.120 yes 20 -0.830 -0.910 0.150 yes

[0060] Table 3

[0061] Table 4 includes example values ​​of tGravity_vector for the 20 data samples of Table 3. The calculation of tGravity_vector, which may include components GVx, GVy, and GVz, may be initialized with a starting or default vector, such as<x=-0.353、y=-0.928、z=0.121> . The SUM(X), SUM(Y), and SUM(Z) columns can provide a moving sum including a starting or default vector for a sample of the x, y, and z components of a detected gravity vector for each sample, with one indication that the user was walking at the time of sampling. The WCount column can be initialized to one and provide a moving sum for multiple samples during which the user was walking at the time of sampling. The GVx, GVy, and GVz columns can provide a calculated value of tGravity_vector, which can be the value of the SUM(X), SUM(Y), and SUM(Z) columns divided by the corresponding value of the WCount column. It can be seen that the value of tGravity_vector can be updated accordingly when the user may be walking rather than when the user may not be walking.

[0062] sample SUM(x) SUM(y) SUM(z) WCount G JZ G JZ G JZ 1 -0.353 -0.928 0.121 1 -0.353 -0.928 0.121 2 -0.353 -0.928 0.121 1 -0.353 -0.928 0.121 3 -0.353 -0.928 0.121 1 -0.353 -0.928 0.121 4 -0.353 -0.928 0.121 1 -0.353 -0.928 0.121 5 -0.353 -0.928 0.121 1 -0.353 -0.928 0.121 6 -0.865 -1.948 0.291 2 -0.433 -0.974 0.146 7 -0.765 -2.848 0.411 3 -0.255 -0.949 0.137 8 -1.085 -3.648 0.561 4 -0.271 -0.912 0.140 9 -1.745 -4.558 0.711 5 -0.349 -0.912 0.142 10 -1.945 -5.308 0.881 6 -0.324 -0.885 0.147 11 -2.145 -6.328 1.001 7 -0.306 -0.904 0.143 12 -2.245 -7.128 1.151 8 -0.281 -0.891 0.144 13 -2.245 -7.128 1.151 8 -0.281 -0.891 0.144 14 -2.245 -7.128 1.151 8 -0.281 -0.891 0.144 15 -2.245 -7.128 1.151 8 -0.281 -0.891 0.144 16 -2.245 -7.128 1.151 8 -0.281 -0.891 0.144 17 -2.245 -7.128 1.151 8 -0.281 -0.891 0.144 18 -2.245 -7.128 1.151 8 -0.281 -0.891 0.144 19 -2.905 -7.928 1.271 9 -0.323 -0.881 0.141 20 -3.735 -8.838 1.421 10 -0.374 -0.884 0.142

[0063] Table 4

[0064] The controller 202 and the one or more sensors 210 may operate or calculate the data features according to one or more of the data acquisition or processing settings listed in Table 5, for example.

[0065]

[0066] Table 5

[0067] In certain aspects, the settings listed in Table 5 can advantageously allow activity monitoring device 120 to operate for at least 7 days, which can mean that memory device 204 and power supply 208 can continue to operate for at least 7 days without failure (e.g., due to memory overflow or lack of power). In some aspects, activity monitoring device 120 can be designed or configured to last no more than 7 days. For example, activity monitoring device 120 can be operated so that memory device 204 overflows after 10, 12, 14, 16, 18, 20, 22, or 24 days. As another example, activity monitoring device 120 can be operated so that the power supply can be expected to fail after 10, 12, 14, 16, 18, 20, 22, or 24 days. In some embodiments, activity monitoring device 120 can be designed or configured to operate for less than or more than 7 days, such as 5 days, 10 days, 12 days, etc.

[0068] In some aspects, the data features listed in Table 1 may be ideally processed by controller 202 despite controller 202 having processing limitations, such as limited data sorting capabilities. Because activity monitoring device 120 may be portable and designed to have a compact size or shape, controller 202 may be selected to have a compact size or shape, which may therefore have fewer processing features or computing power than some other controllers. In addition, the data features listed in Table 1 may have been selected from additional data features because the data features listed in Table 1 may be used together to accurately (e.g., with a confidence level of approximately 90% or 95%) detect high accelerations experienced by activity monitoring device 120, classify an activity engaged in by a user wearing activity monitoring device 120, or detect use of an unloading device by a user wearing activity monitoring device 120. In some cases, if lower detection accuracy is acceptable, fewer than all of the data features listed in Table 1 (or alternative features) may be used.

[0069] Controller 202 may store the determination of the data characteristics in memory device 204 with or without loss of accuracy (e.g., lossless or lossy). In some aspects, lossless storage methods may advantageously ensure that the data characteristics determination is continuously monitored and recorded once activity monitoring device 120 begins monitoring and recording and throughout use of activity monitoring device 120. Controller 202 may or may not store raw sensor data, such as x-axis accelerometer data or z-axis magnetometer data, to memory device 204. In some aspects, controller 202 may not store raw sensor data to memory device 204, at least limiting the utilization of memory device 204 and allowing the selection of memory device 204 with a lower storage capacity than otherwise available.

[0070] The communication interface 212 may be used to communicate with other devices via radio waves, for example wirelessly. The communication may be based on a communication protocol such as Bluetooth. TM Protocols, such as Bluetooth TM Low energy execution. The communication interface 212 can, for example, communicate and pair with other devices and transmit device usage or sensor data (e.g., [i] sensor-derived data, including determined values ​​of one or more data features listed in Table 1 over a data window, [ii] monitored load or positioning, [iii] alarms, or [iv] changes to a monitoring or treatment regimen performed by the activity monitoring device 120) to other devices. The communication interface 212 can be used to receive data including commands from other devices. Once the communication channel is configured to communicate with another device (e.g., by device pairing), the communication interface 212 can allow communication with the other device (e.g., transmitting data to the other device or processing commands from the other device). In some aspects, the communication interface 212 may be unable to communicate more than 10 meters, 30 meters, or 100 meters away from the communication interface 212.

[0071] Figure 3 Computing environment 300 is shown including activity monitoring device 120 and offload monitoring device 132. Computing environment 300 shows activity monitoring device 120 and offload monitoring device 132 communicating with user operated device 310 and clinician operated device 320, and user operated device 310 and clinician operated device 320 communicating with device management system 340 via network 330.

[0072] The user-operated device 310 may be operated by a user, such as a wearer, of the activity monitoring device 120 and the off-load monitoring device 132. The user-operated device 310 may allow the user to use the user-operated device 310 to collect, process, review, or transmit data collected by the activity monitoring device 120 and the off-load monitoring device 132. On the other hand, the clinician-operated device 320 may be operated by a clinician for a user, such as to supervise, assist, or care for an individual user who uses the activity monitoring device 120 and the off-load monitoring device 132. The clinician-operated device 320 may allow the clinician to use the clinician-operated device 320 to collect, process, review, or transmit data collected by the activity monitoring device 120 and the off-load monitoring device 132.

[0073] The user-operated device 310 and the clinician-operated device 320 may each be a computing device, such as a smartphone, a tablet, or a desktop computer. In some aspects, the user-operated device 310 and the clinician-operated device 320 may receive, send, present, and access data collected by or determined from the activity monitoring device 120 and the offload monitoring device 132, but may not process the data collected by the activity monitoring device 120 and the offload monitoring device 132 to analyze characteristics of the data (e.g., identifying or calculating from the data the duration of an activity in which the user engages while using the offload device).

[0074] The user operating device 310 may include a user application 312, a communication interface 314 (which may, for example, include any one or more features of the communication interface 212), and a user interface 316 (which may, for example, include any one or more features of the user interface 206). The user application 312 may be a program executed by a processor of the user operating device 310. The user application 312 may enable the user operating device 310 to communicate with the activity monitoring device 120, the load reduction monitoring device 132, and the device management system 340 via the communication interface 314. The user application 312 may receive, collect, process, review, or transmit (i) data collected or determined by the activity monitoring device 120 and the offload monitoring device 132, such as determined characteristic data, sensor data, alarms, monitored loads or positions, or changes in monitoring or treatment regimens, (ii) data collected or determined by the user application 312, such as user observations, goal information, or identification of activities in which the user is engaged, or (iii) data collected or determined by the device management system 340, such as the duration of time the user is active, the portion of time the user is using an offload device, or the amount of time the user may be at risk of injury due to the user's activity or non-use of an offload device. Furthermore, the user application 312 may present one or more graphical user interfaces having the user interface 316 to the user, such as on a display or touch screen of the user interface 316.

[0075] The user application 312 may be used to collect information from the user to help understand the types of activities that the user engages in when the user's foot may experience load forces that may have a risk of adversely affecting an existing ulcer on the foot, or a risk of causing a new ulcer on the foot. The user application 312 may receive from the device management system 340 the times at which the user's foot experiences load forces indicated by sensor data of the activity monitoring device 120 that may have a risk of adversely affecting an existing ulcer on the foot, or a risk of causing a new ulcer on the foot, and the user application 312 may determine the geographic location of the user-operated device 310 (e.g., from a global positioning system (GPS) receiver of the user-operated device 310 or a communication received by the user-operated device 310 via the network 330), which may indicate the geographic location of the activity monitoring device 120 at that time. The user application 312 may output the time and geographic location of the user-operated device 310, and a request from the user to identify the type of activity that the user engages in at that time and geographic location. For example, the activity type may include a transportation type, an exercise type, a leisure type, a hobby type, or a work type. The request may list a set of types of activities that the user typically participates in to help select the type of activity that the user participates in.

[0076] User application 312 may include features and interfaces that seek to increase the amount or frequency of user interaction with user application 312. User application 312 may request contextual information (e.g., identification of activities the user engaged in while wearing activity monitoring device 120, or an indication from the user of how the user felt) to assist in tracking or understanding why the user's foot may be experiencing load forces that are at risk of adversely affecting an existing ulcer on the foot or at risk of causing a new ulcer on the foot.

[0077] The clinician operating device 320 may include a clinician application 322, a communication interface 324 (which may, for example, include any one or more features of the communication interface 212), and a user interface 326 (which may, for example, include any one or more features of the user interface 206). The clinician application 322 may be a program executed by a processor of the clinician operating device 320. The clinician application 322 may enable the clinician operating device 320 to communicate with the activity monitoring device 120, the offload monitoring device 132, and the monitoring management system 340 via the communication interface 324. The clinician application 322 may receive, collect, process, review, or transmit (i) data collected or determined by the activity monitoring device 120 and the offload monitoring device 132, such as determined characteristic data, sensor data, alarms, monitored loads or positioning, or changes in monitoring or treatment regimens, (ii) data collected or determined by the user application 312, such as user observations, goal information, or identification of activities in which the user is engaged, or (iii) data collected or determined by the device management system 340, such as the duration of time the user is active, the portion of time the user is using an offload device, or the amount of time the user is at risk of injury due to the user's activity or not using an offload device. The clinician application 322 may present one or more graphical user interfaces having the user interface 326 to the clinician, such as on a display or touch screen of the user interface 326.

[0078] In some aspects, the clinician-operated device 320 may not communicate directly with the activity monitoring device 120 or the offload monitoring device 132. Instead, the clinician-operated device 320 may receive any data collected by or associated with the activity monitoring device 120 and the offload monitoring device 132 from the device management system 340. This design may desirably limit the number of devices (other than the user-operated device 310) that may receive data directly from the activity monitoring device 120 and the offload monitoring device 132, thereby enhancing the security of the data and limiting access to the data.

[0079] Device management system 340 may be a computing device, such as a server, and includes a communication management system 342, a data processing system 344, and a data storage device 346 that may communicate with each other. For example, device management system 340 may be partially or completely comprised of a server infrastructure or cloud architecture, such as using a server managed by Amazon Web Services. TM (AWS), Microsoft TM Azure TM 、Google Cloud Platform TM (GCP) or Oracle TMThe cloud infrastructure provided by Cloud Infrastructure (OCI) The server infrastructure and the cloud infrastructure can comply with HIPAA (Health Insurance Portability and Accountability Act of 1996) requirements and provide data privacy and security protection given the potential sensitivity of data collected, processed, or determined by the device management system 340.

[0080] The communication management system 342 may allow the device management system 340 to communicate with the user-operated device 310 and the clinician-operated device 320 over the network 330. The communication management system 342 may include an application programming interface (API), such as a cloud API, to facilitate its communications.

[0081] The data processing system 344 may collect, process, present, store (e.g., in a data storage device 346), or transmit data collected or determined by the activity monitoring device 120 and the unloading monitoring device 132 (e.g., determined characteristic data, sensor data, alarms, monitored loads or positioning, or changes in monitoring or treatment regimens) and data collected or determined by the user application 312 (e.g., user observations, goal information, or identification of activities in which the user is engaged). For example, the data processing system 344 may process the data collected by the activity monitoring device 120 and the unloading monitoring device 132 to determine activities that the user may have engaged in during various time periods during which the data was collected, and whether the user may have worn an unloading device during various time periods during which the data was collected. The determined possible activities or uses of the unloading device may be shared by the device management system 340 with the user-operated device 310 or the clinician-operated device 320. As another example, data processing system 344 may process data collected by activity monitoring device 120 and offload monitoring device 132 to identify events of interest, such as events that may indicate excessive pressure is being applied to body part 110 , and may share the events of interest with user-operated device 310 or clinician-operated device 320 .

[0082] The network 330 may be a computer network, such as a cellular communication network or may include a cellular communication network. Although the network 330 is shown as a connected network, the network 330 may be subdivided into one or more separate networks that may not communicate directly with each other. For example, the device management system 340 may communicate with the user operating device 310 via a network separate and different from the network used by the device management system 340 to communicate with the clinician operating device 320.

[0083] Although certain data processing in computing environment 300 may be described as being performed by activity monitoring device 120, offload monitoring device 132, user-operated device 310, clinician-operated device 320, or data processing system 344, certain data processing may be transferred to different devices or systems in computing environment 300. For example, user-operated device 310 may be described as not processing data provided by activity monitoring device 120 and offload monitoring device 132, but may rely on data processing system 344 to analyze the data; however, user-operated device 310 may additionally or alternatively use a method or algorithm similar to or different from data processing system 344 to analyze the data.

[0084] Figure 4 An exemplary data processing 400 performed by the data processing system 344 is shown. As shown, activity monitoring data 410 and load shedding monitoring data 420 may be received by the data processing system 344. The activity monitoring data 410 and load shedding monitoring data 420 may then be processed by high acceleration detection 430, activity classification 432, and load shedding usage detection 434 of the data processing system 344. The high acceleration detection 430, activity classification 432, and load shedding usage detection 434 may process various sets of the activity monitoring data 410 and load shedding monitoring data 420 in parallel or serially.

[0085] For example, individual sets may include values ​​of data features determined by activity monitoring data 410 or load shedding monitoring data 420 for individual data windows. Data windows may correspond to time windows that occur in succession, such as windows of 10 milliseconds, 25 milliseconds, 50 milliseconds, or 100 milliseconds, 1 second, 2 seconds, 5 seconds, 10 seconds, 20 seconds, or 30 seconds. For example, if a 30-second data window is used, there may be 2 different windows for one minute of data: a first data window covering 1-30 seconds and a second data window covering 31-60 seconds. Then, the next data window may cover 1-30 seconds of the next minute of data. Data windows may correspond to sliding time windows that advance at regular intervals that are less than the total length of a single data window. For example, if a 30-second sliding data window is used, where the window advances at 1-second intervals, there may be 31 different data windows that occur entirely within one minute of data: the first data window covers 1-30 seconds, the second data window covers 2-31 seconds, the third data window covers 3-32 seconds, and continues until the thirty-first data window covers 31-60 seconds. The next data window may then cover 32-60 seconds of data for one minute, and 1 second of data for the next minute.

[0086] The output of high acceleration detection 430, activity classification 432, and offload use detection 434 may be provided to event detection 436 of data processing system 344. Event detection 436 may monitor single events periodically, for example, perhaps detecting events every 1, 2, 5, 10, 20, 30, 45, or 60 minutes. The output of event detection 436 may be provided to event priority 438 of data processing system 344. The output of event priority 438 may be event data 440, which may be further processed by data processing system 344, stored in data storage device 346, or transmitted to user operated device 310 or clinician operated device 320 (e.g., via an application program interface (API) of data processing system 344) for processing or display.

[0087] High acceleration detection 430 , activity classification 432 , offload usage detection 434 , event detection 436 , and event prioritization 438 may each be algorithms implemented in some aspect by one or more processors of data processing system 344 .

[0088] Activity monitoring data 410 and offload monitoring data 420 may be obtained by communication management system 342 via network 330, such as from a user operating device 310. Activity monitoring data 410 may include data collected or determined by activity monitoring device 120 (e.g., determined characteristic data, sensor data, alarms, monitored loads or positions, or changes in monitoring or treatment regimens), and data collected or determined by user application 312 (e.g., user observations, goal information, identification of activities in which the user is engaged, or other information provided by the user in connection with activity monitoring device 120). Offload monitoring data 420 may include data collected or determined by offload monitoring device 132 (e.g., determined characteristic data, sensor data, alarms, monitored loads or positions, or changes in monitoring or treatment regimens), and data collected or determined by user application 312 (e.g., user observations, goal information, identification of activities in which the user is engaged, or other information provided by the user in connection with activity monitoring device 120 or offload monitoring device 132). High acceleration detection 430 , activity classification 432 , and offload usage detection 434 may be initiated by the device management system 340 upon successful receipt or storage of activity monitoring data 410 , offload monitoring data 420 , or a collection thereof.

[0089] High acceleration detection 430 may process activity monitoring data 410 to determine whether activity monitoring device 120 has experienced high acceleration, such as high instantaneous acceleration or high average acceleration, for a single window of data. High acceleration detection 430 may be performed by comparing one or more acceleration values ​​of activity monitoring data 410 for a single window of data to an acceleration threshold (or thresholds) to determine whether the one or more acceleration values ​​meet the acceleration threshold (or thresholds). The acceleration threshold (or thresholds) may be defined, for example, as a multiple of a standard acceleration due to gravity (sometimes referred to as g0, which has a value of approximately 9.807 m / s 2 ), for example, defined as 1.5, 2, 2.5, 3, 3.5, 4, 4.5, or 5 times the standard acceleration due to gravity. The output of high acceleration detection 430 can be an indication of whether the activity monitoring device 120 experienced high acceleration during data capture over a single data window. The output of high acceleration detection 430 can include, for example, one or more of a sensor ID, a user ID, tBodyAcc_mag, a sensor timestamp, or an evaluation time. For a single data window, the output can be organized into a table, a list, or an array.

[0090] Activity classification 432 may process activity monitoring data 410 to determine an activity engaged in while wearing or supporting activity monitoring device 120. Activity classification 432 may, for example, classify an activity for an individual data window or multiple data windows (e.g., data corresponding to sensor detection data over a 30 or 45 second or 1, 2, 3, 5, or 10 minute period) as one of [i] movement (e.g., walking or running), [ii] no movement but static weight (e.g., standing), [iii] no movement but force changes (e.g., driving a car), or [iv] lying down (e.g., lying on a flat surface). Generally speaking, lying down may be considered to pose no risk for existing injuries (e.g., ulcers on a body part) or causing new injuries due to the lack of or little weight load on the body part, while movement, no movement but static weight, or no movement but force changes may be considered to pose a risk for existing injuries or causing new injuries.

[0091] Activity classification 432 may be performed using machine learning techniques (e.g., gradient boosting), decision tree-based methods, neural networks, or support vector machines. Activity classification 432 may operate based on an activity classification model 433 (which may also be referred to as an activity model or, in some cases, a trained activity classification module) used by activity classification 432 to identify or distinguish [i] movement, [ii] no movement but with static weight, [iii] no movement but with force change, and [iv] lying down from activity monitoring data 410, such as from feature data determined for one or more or all of the data features in Table 1. In one example, activity classification model 433 may be a decision tree used by activity classification 432 to identify individual data windows or multiple data windows of activity monitoring data 410 that are likely to correspond to one of two or more different activities. As another example, activity classification model 433 may be used by activity classification 432 to determine the extent to which individual data windows or multiple data windows of activity monitoring data 410 correspond to each of two or more different activities, and activity classification 432 may select the different activities that are most likely to be the classified activities for the individual data windows or multiple data windows.

[0092] Activity classification 432 can analyze activity monitoring data 410 using activity classification model 433, which can include a decision tree to determine activity classification. For example, the decision tree can stipulate that if activity monitoring data 410 indicates that the user's legs can be raised, the user may be lying down but not in motion, no motion but static weight, or no motion but force changes. Activity classification 432 can determine whether the legs can be raised from the detected gravity vector and gravity reference vector (e.g., tGravity_vector) of activity monitoring data 410, for example, by determining the angle between the detected gravity vector and the gravity reference vector using formula 1 derived from the dot product of vectors x and y. In one example, activity classification 432 can determine that the legs can be raised when the angle between the detected gravity vector and the gravity reference vector can be at least 30°, 45°, or 60°. The decision tree can additionally or alternatively stipulate that if activity monitoring data 410 indicates that a threshold sway amount is exceeded over time, the user may not be in a state of no motion but static weight.

[0093] Formula 1 θ=arccos(x·y / |x||y|)

[0094] Activity classification 432 can generate activity model input data from activity monitoring data 410. For example, the activity model input data can include feature data of a specific data window, such as a feature vector including data elements representing data features in Table 1, or a subset thereof. Activity classification 432 can analyze the activity model input data using activity classification model 433 to determine the activity classification. Activity classification model 433 can generate activity model output data, which represents the possible activity classification of activity monitoring data 410 based on which the activity model input data is generated. For example, the activity model output data can be a vector, an array, or other set of data elements. Each data element of the set can correspond to each activity classification that the activity classification model 433 is trained to detect. The value of a single element of the activity model output data can represent the possibility that the activity model input data represents an activity within the activity classification corresponding to the single data element. The activity classification corresponding to the activity model output data element that meets one or more classification criteria can be selected as the activity classification of the current data window. Examples of applying classification criteria may include selecting the activity model output data element with the highest value relative to other data elements of the activity model output data, or selecting the activity model output data element with a value that meets a threshold. The exemplary processing of activity monitoring data 410 using activity classification model 433 described herein is illustrative and is not intended to be limiting or exhaustive. Depending on any of a variety of factors, including the structure of activity classification model 433, the structure of activity monitoring data 410, or available computing resources, alternative or additional operations may be used.

[0095] Activity classification 432 can store data about activity classifications determined for a data window (e.g., for each time period or time instance). Activity classification 432 can use two or more activity classifications - determined for two or more data windows - to detect changes in activity classifications and output new data in response to changes in activity classifications. The output of activity classification 432 can, for example, include one or more of the following: activity record ID, user ID, activity start time, activity end time, activity duration, activity intensity, or evaluation time. The maximum activity change frequency (e.g., so that the activity classification can change no more than once every 30 seconds or 1, 2, 5, or 10 minutes) can be set by the user or depend on previously or newly determined activities to limit the sensitivity of activity classification 432 to changes. Activity classification 432 can have a hypertext transfer protocol (HTTP) wrapper and can be accessed by HTTP requests to a URL.

[0096] Load-relief usage detection 434 may process activity monitoring data 410 and load-relief monitoring data 420 to determine whether a load-relief device is worn when wearing or supporting activity monitoring device 120. Load-relief usage detection 434 may determine whether a load-relief device is worn, for example, from a comparison of activity monitoring data 410 with load-relief monitoring data 420, such as a correlation of activity monitoring data 410 and load-relief monitoring data 420 over an individual data window or multiple data windows (e.g., data corresponding to sensor detection data over a 30 or 45 second or 1, 2, 3, 5, or 10 minute time period). Additionally or alternatively, load-relief usage detection 434 may be performed using machine learning techniques (e.g., gradient boosting), decision tree-based methods, neural networks, or support vector machines. Detection may be performed over an individual data window or multiple data windows.

[0097] The load shedding usage detection 434 may operate based on a load shedding usage model 435 used by the load shedding usage detection 434 to identify or distinguish the use or non-use of a load shedding device from the activity monitoring data 410 and the load shedding monitoring data 420, such as feature data determined from one or more or all of the data features in Table 1 generated for the activity monitoring device 120 and the load shedding monitoring device 132. In one example, the load shedding usage model 435 may be a decision tree used by the load shedding usage detection 434 to determine whether an individual data window or multiple data windows of the activity monitoring data 410 or the load shedding monitoring data 420 correspond to the use or non-use of a load shedding device. As another example, the load shedding usage model 435 may be used by the load shedding usage detection 434 to determine the extent to which an individual data window or multiple data windows of the activity monitoring data 410 and the load shedding monitoring data 420 correspond to the use or non-use of a load shedding device, and the load shedding usage detection 434 may select as the use determination of the individual data window or multiple data windows that the load shedding device is more likely to be used or not used.

[0098] The load-reducing usage model 435 may be programmed or trained to determine whether a load-reducing device is likely to be used during a particular data window. For example, the load-reducing usage detection 434 may generate a feature vector for a particular data window of the load-reducing monitoring data 420, the feature vector comprising data elements representing data features in Table 1, or a subset thereof. The feature vector may include data elements representing data features of the activity monitoring data 410 for the same or corresponding data window, one or more data elements representing the output of the activity classification model 433 for the same data window, or one or more data elements representing the activity classification determined by the activity classification 432 for the same data window. The load-reducing usage model 435 may generate load-reducing usage model output data. The load-reducing usage model output data may represent the likelihood of wearing a load-reducing device during the data window, the degree to which the activity monitoring data 410 and the load-reducing monitoring data 420 are related, and the like. The load-reducing usage detection 434 may use the load-reducing usage model output data to determine a load-reducing usage classification that indicates whether a load-reducing device is likely to be worn at a time corresponding to the data window.

[0099] Additionally or alternatively, the load shedding usage detection 434 may analyze the load shedding usage model input data using the load shedding usage model 435 to determine an activity classification for the load shedding monitoring data 420. The activity classification may be compared to the activity classification determined for the activity monitoring data 410 for the same data window. If the activity classification matches or otherwise satisfies one or more relevance criteria, the load shedding usage detection 434 may determine that the load shedding device was likely worn at the time corresponding to the data window. Otherwise, if the activity classification does not match or otherwise fails to satisfy one or more relevance criteria, the load shedding usage detection 434 may determine that the load shedding device was likely not worn at the time corresponding to the data window. The final determination of whether the load shedding device was likely worn at the time corresponding to the data window may be the load shedding usage classification.

[0100] Depending on any of a variety of factors, including the structure of offload usage model 435 , the structure of activity monitoring data 410 or offload monitoring data 420 , or available computing resources, alternative or additional operations for offload usage detection 434 may be used.

[0101] The load shedding usage detection 434 can store data about load shedding usage determinations for a data window (e.g., for various time periods or time instances). The load shedding usage detection 434 can use two or more load shedding usage determinations - for two or more data windows - to detect possible changes in the load shedding usage detection, and output new data in response to the change in the load shedding usage detection. The output of the load shedding usage detection 434 can, for example, include one or more of the following: load shedding record ID, user ID, evaluation time, detection start time, detection end time, or detection duration. The maximum load shedding usage change frequency (e.g., so that the load shedding usage detection can change no more than once every 30 seconds or 1, 2, 5, or 10 minutes) can be set by the user to limit the sensitivity of the load shedding usage detection 434 to changes. The correlation window size can be set by the user to control the number of samples from each of the activity monitoring data 410 and the load shedding monitoring data 420 that can be processed by the load shedding usage detection 434 to determine the use or non-use of the load shedding device.

[0102] Event detection 436 may process outputs from high acceleration detection 430, activity classification 432, and load shedding use detection 434 to identify one or more events. The one or more events may, for example, each be an instance of: (i) high acceleration as determined by high acceleration detection 430; or (ii) low acceleration as determined by high acceleration detection 430, together with an activity classification as determined by activity classification 432 and a load shedding device not being used as determined by load shedding use detection 434. In some cases, event detection 436 may fail to identify an event in response to a determination of lying down as determined by activity classification 432 or a determination of being used as determined by load shedding use detection 434 of a load shedding device being used. The outputs of event detection 436 may, for example, include one or more of: an event record ID, an event start time, an event end time, an event peak time, an event duration, an event impact tag, an event classification, or an event load shedding use percentage. The maximum event change frequency (e.g., so that detected events may change no more than once every 30 seconds or 1, 2, 5, 10, 20, 30, 45, or 60 minutes) may be set by the user to limit the sensitivity of event detection 436 to changes or to limit the number of event changes detected per day. Event detection 436 may store and output new data in response to a change in activity classified by activity classification 432, a change in load shedding device usage determined by load shedding usage detection 434, or the detection of high acceleration by high acceleration detection 430.

[0103] Event priority 438 can organize one or more events identified by event detection 436 to assist in reviewing the one or more events. Event priority 438 can, for example, prioritize any event that is an instance of high acceleration, and a selected number (e.g., 5, 10, or 15) of any events that are instances of low acceleration but have the highest associated acceleration magnitude relative to other events with low acceleration. Event priority 438 can categorize or filter one or more events to help ensure that one or more events that may be of most interest to a user or clinician can be brought to the user's or clinician's attention.

[0104] Figure 5 An event recording process 500 is shown for identifying events and recording event data associated with the events. For convenience, the event recording process 500 is described as being performed by the data processing system 344 and in the context of the computing environment 300, but may alternatively be implemented in other components or systems described herein, or by other computing systems not shown. In certain aspects, the event recording process 500 may advantageously facilitate the detection of events during which an individual experienced a load force that was at risk of adversely affecting an existing injury (e.g., an ulcer on the foot) or at risk of causing a new injury (e.g., a new ulcer on the foot), and the recording of relevant event information. In turn, the event information can be used to understand what may be affecting an existing injury or a new injury in the user's lifestyle and help promote changes in the user's lifestyle.

[0105] At block 510, the event recording process 500 may obtain a sensor data set. The sensor data set may include, for example, activity monitoring data 410 and load shedding monitoring data 420, which may include characteristic data determined by the activity monitoring device 120 or the load shedding monitoring device 132. The sensor data set may be received by the data processing system 344 via the network 330, for example, from the user operating device 310, or from the data storage device 346. The sensor data set may correspond to or indicate movement or orientation of the activity monitoring device 120 or the load shedding monitoring device 132 over a period of time (e.g., 1, 2, 5, 10, 20, 30, 45, or 60 seconds).

[0106] At block 520, the event recording process 500 may determine acceleration, activity duration, and deceleration usage of the sensor data set. For example, the acceleration, activity duration, and deceleration usage may be determined using the high acceleration detection 430, activity classification 432, and deceleration usage detection 434 of the data processing system 344 as described herein.

[0107] At block 530, the event recording process 500 may determine whether the acceleration determined at block 520 is a high acceleration. For example, the event detection 436 may determine whether the output from the high acceleration detection 430 indicates that the acceleration is a high acceleration.

[0108] If the acceleration is high acceleration, the event recording process 500 can transition to box 560 and record an event corresponding to the acceleration. For example, event detection 436 can record the event as event data associated with the event so that the event data can be shared with a user or clinician and reviewed in the future to enable the user or clinician to understand or discuss the event. The event data can be added to the first list. The event data can be an output from event detection 436 (or at least a portion thereof).

[0109] If the acceleration is not high acceleration, the event recording process 500 can be transferred to box 540 and determine whether the activity duration determined at box 520 is an extended duration. For example, event detection 436 can determine whether the activity duration identified by activity classification 432 meets (e.g., continues) a threshold duration. The threshold duration may depend on the activity (e.g., exercise, no exercise but static weight, or no exercise but force change) determined by activity classification 432 to be involved during the activity duration. Because different activities may have different probabilities of negatively affecting existing injuries or causing new injuries over time, a threshold duration can be selected for each activity to correspond to the risk of negative effects of each activity over time. The threshold duration can be set in addition or alternatively based on user input. If the activity duration is not an extended duration, the event recording process 500 can end and no events can be recorded for the sensor data set.

[0110] If the activity duration is an extended duration, the event recording process 500 may transition to block 550 and determine whether the load shedding usage determined at block 520 indicates usage of a load shedding device. For example, event detection 436 may determine whether the output of load shedding usage detection 434 indicates usage of a load shedding device. If the load shedding usage indicates that a load shedding device was used (or at least used and lasted for a threshold duration or percentage of the activity duration), the event recording process 500 may end and no events may be recorded for the sensor data set (or window).

[0111] If the offloading usage indicates that the offloading device was not used (or at least not used for a threshold duration or percentage of the activity duration), the event recording process 500 can transition to box 560 and record an event corresponding to non-high acceleration, extended activity duration, and offloading device non-use. The event data for the event can be recorded in a second list, which can be the same or different from the first list. For example, event detection 436 can record the event as event data associated with the event so that the event data can be shared with a user or clinician and reviewed in the future to enable the user or clinician to understand or discuss the event. The event data can be an output from event detection 436 (or at least a portion thereof).

[0112] The event recording process 500 may be repeated for one or more additional sensor data sets such that the event recording process 500 may be used to record events for the activity monitoring device 120 or the unloading monitoring device 132 .

[0113] In certain aspects, the data processing 400 or event recording process 500 can advantageously process large amounts of activity monitoring data 410 and offloading monitoring data 420 to detect and prioritize hundreds, thousands, or more events for recording or presentation to a clinician or user. Events can indicate a risk of adversely affecting an existing injury or developing a new injury, such as to the user's skin. The events can include [i] high acceleration events or [ii] events corresponding to non-high acceleration, extended activity duration (e.g., the activity is an activity other than lying down), and the offloading device is not used during the duration of the activity. Events prioritized by the data processing 400 or event recording process 500 can ideally support discussions or actions that lead to behavioral changes that limit the risk of adversely affecting existing injuries or developing new injuries.

[0114] Computer system components

[0115] Figure 6 Shows the available Figure 3 A computer system 600 is constructed within the computing environment 300 of the present invention, and one or more devices (eg, a user-operated device 310 or a clinician-operated device 320), systems (eg, a device management system 340), servers, etc. are configured.

[0116] like Figure 6As shown in , the computer system 600 may include (i) a processor (CPU) 610, (ii) an input / output device 620 configured to allow a user to input and output information and interact with the computer system 600, as well as to transmit data and receive data or capture data using one or more sensors such as an image sensor, (iii) a read-only memory device (ROM) 630 or equivalent to provide non-volatile storage of data or programs, (iv) a display 650, such as a computer monitor or other display device, (v) a network connection 640 and a network interface 642, which are configured to allow the computer system 600 to connect to other systems, servers or portable devices, as well as a memory space 660 and a database 690. The database 690 can be further divided or distributed into sub-databases 690A-690N, where the sub-databases store feature or function specific information associated with a specific feature or function. Figure 6 The various components shown in can be incorporated into the computer 670. It should be noted that Figure 6 The various components shown in, including database 690, are typically included as part of computer 660, however, in some aspects, they can be external to computer 670. For example, database 690 can be external to computer 670 and can be a separate database computer system or part of a networked database system. In some cases, computer system 600 can be a computing device such as a desktop computer, a mobile phone, or a server.

[0117] The memory space 660 may include DRAM, SRAM, FLASH, a hard drive, or other memory storage device, such as a media drive 680, configured to store an operating system 662, application programs 664, and data 668, and the memory space 660 may share, distribute, or overlap with the memory storage capacity of the database 690. In some aspects, the memory space 660 may include the database 690, or in some aspects, the database 690 may include the data 668, as shown in the memory space 660. The data stored in the memory space 660 or the database 690 may include information, such as athletic data, pairing program information, data processing routines, or other types of data described herein.

[0118] Other variations and terms

[0119] Many other variations other than those described herein will be apparent from this disclosure. For example, depending on the embodiment, certain actions, events, or functions of any algorithm described herein may be performed in a different order, may be added, merged, or omitted entirely (e.g., not all described actions or events are necessary to practice the algorithm). Furthermore, in some embodiments, actions or events may be performed simultaneously, such as by multithreading, interrupt handling, or multiple processors or processor cores or on other parallel architectures, rather than sequentially. Additionally, different tasks or processes may be performed by different machines or computing systems that may function together.

[0120] One or more user inputs described in the present disclosure may be received using one or more different mechanisms. For example, a user may select a user interface control using one or more input options, such as a mouse, touch screen input, or keyboard input, as well as other user interface input options. The user interface control selected by the user may include one or more of a button, a drop-down box, a selection box, a text box, a check box, a slider control, or other user interface controls.

[0121] The various illustrative logic blocks, modules, and algorithmic steps described in conjunction with the embodiments disclosed herein may be implemented as electronic hardware, computer software, or a combination of the two. In order to clearly illustrate this interchangeability of hardware and software, various illustrative components, blocks, modules, and steps have been generally described above with respect to their functions. Whether such functions are implemented as hardware or software depends on the specific application and the design constraints imposed on the entire system. The described functions may be implemented in different ways for each specific application, but such implementation decisions should not be interpreted as causing deviations from the scope of the present disclosure.

[0122] The various illustrative logic blocks and modules described in conjunction with the embodiments disclosed herein may be implemented or executed by a machine, a microprocessor, a state machine, a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic device, a discrete gate or transistor logic, a discrete hardware component, or any combination thereof to perform the functions described herein. A hardware processor may include a circuit or digital logic circuit configured to process computer executable instructions. On the other hand, the processor includes an FPGA or other programmable device that performs logic operations without processing computer executable instructions. The processor may also be implemented as a combination of computing devices, such as a combination of a DSP and a microprocessor, a plurality of microprocessors, one or more microprocessors combined with a DSP core, or any other such configuration. The computing environment may include any type of computer system, including but not limited to a microprocessor-based computer system, a mainframe computer, a digital signal processor, a portable computing device, a computing engine within a device controller or an appliance, and the like.

[0123] The steps of the methods, processes or algorithms described in conjunction with the embodiments disclosed herein may be directly included in hardware, in a software module stored in one or more storage devices and executed by one or more processors, or in a combination of the two. The software module may reside in a RAM memory, a flash memory, a ROM memory, an EPROM memory, an EEPROM memory, a register, a hard disk, a removable disk, a CD-ROM or any other form of non-temporary computer-readable storage medium, a medium or a physical computer storage device known in the art. An exemplary storage medium may be coupled to a processor so that the processor can read information from the storage medium and write information to the storage medium. In an alternative, the storage medium may be an integral part of the processor. The storage medium may be volatile or non-volatile. The processor and the storage medium may reside in an ASIC.

[0124] Conditional language as used herein, such as "can," "may," "could," "for example," and the like, unless expressly stated otherwise or otherwise understood in the context of use, is generally intended to express that certain aspects include, while other aspects do not include, certain features, elements, or statements. Thus, such conditional language is generally not intended to imply that one or more aspects require features, elements, or statements in any way, or that one or more aspects must include logic for determining whether such features, elements, or statements are included or executed in any particular embodiment with or without author input or prompting. The terms "include," "comprise," "have," and the like are synonymous and are used inclusively in an open-ended manner and do not exclude additional elements, features, actions, operations, and the like. In addition, the term "or" is used in its inclusive sense (rather than in its exclusive sense) so that when used, for example, to connect a list of elements, the term "or" means one, some, or all of the elements in the list. In addition, in addition to having its ordinary meaning, the term "each" as used herein may mean any subset of a set of elements to which the term "each" is applied.

[0125] Unless expressly stated otherwise, conjunctive language such as the phrase "at least one of X, Y, and Z" is understood in the context as generally used to indicate that an item, term, etc. can be X, Y, or Z. Thus, such conjunctive language is generally not meant to imply that certain embodiments require the presence of at least one X, at least one Y, and at least one Z.

[0126] Language of degree as used herein, such as the terms "about," "approximately," "substantially," and "roughly" as used herein, refers to a value, amount, or feature that is close to a specified value, amount, or feature that still performs a desired function or achieves a desired result. For example, the terms "about," "approximately," "substantially," and "roughly" may refer to an amount that is within less than 10%, less than 5%, less than 1%, less than 0.1%, and less than 0.01% of a specified amount. As another example, in certain embodiments, the terms "substantially parallel" and "roughly parallel" refer to a value, amount, or feature that deviates from exact parallelism by less than or equal to 15 degrees, 10 degrees, 5 degrees, 3 degrees, 1 degree, or 0.1 degrees.

[0127] Although the above detailed description has shown, described and pointed out novel features applicable to various embodiments, it should be understood that various omissions, substitutions and changes in the form and details of the described devices or algorithms may be made without departing from the spirit of the present disclosure. As will be appreciated, certain embodiments described herein may be embodied in a form that does not provide all of the features and benefits set forth herein, as some features may be used or practiced separately from other features.

Claims

1. A system for collecting and processing sensor data to identify a risk of causing skin damage, the system comprising a computer readable memory and one or more computer processors, the system being configured to: obtaining sensor characteristic data including a first set of characteristic data representing outputs of a user sensor configured to be worn on a limb of a user and a second set of characteristic data representing outputs of a load-reducing device sensor coupled to a load-reducing device configured to be worn on the limb; determining an activity classification using the first set of feature data and an activity classification model, the activity classification being one of a plurality of different activity classifications, wherein the activity classification model is configured to determine the extent to which the first set of feature data corresponds to each of the plurality of different activity classifications; determining a load shedding usage classification using the first set of feature data, the second set of feature data, and a load shedding usage model; generating display data representing at least one of the activity classification or the load shedding usage classification; as well as transmitting the display data to a computing device configured to present the display data, Wherein the offload usage model is configured to determine the extent to which the movement and orientation of the user sensor corresponds to the movement and orientation of the offload device sensor.

2. The system of claim 1, wherein the skin lesion comprises a diabetic foot ulcer or a venous leg ulcer.

3. A system according to claim 1 or 2, wherein the user sensor includes a first accelerometer and a first magnetometer, and the unloading device sensor includes a second accelerometer and a second magnetometer, and the sensor characteristic data represents the output from the first accelerometer, the second accelerometer, the first magnetometer and the second magnetometer.

4. The system according to claim 1 or 2, further configured to: determining a subsequent activity classification using a third set of feature data and the activity classification model, the third set of feature data representing a subsequent output of the user sensor; and It is determined from the activity classification and the subsequent activity classification that an activity of the user has changed.

5. The system according to claim 1 or 2, further configured to: using the third set of feature data, the fourth set of feature data and the load shedding usage model to determine a subsequent load shedding usage classification, wherein the third set of characteristic data represents a subsequent output of the user sensor, and the fourth set of characteristic data represents a subsequent output of the unloading device sensor; and A determination is made from the load shedding usage classification and the subsequent load shedding usage classification that usage of the load shedding device by the user has changed.

6. The system according to claim 1 or 2, further configured to: generating activity model input data using the first set of feature data; generating activity classification model output data using the activity model input data and the activity classification model; and A data element is identified from the activity classification model output data, the data element having a value that satisfies classification criteria to determine the activity classification associated with the data element.

7. A system according to claim 1 or 2, wherein the multiple different activity classifications include classification of movement of the limb, classification of no movement of the limb but weight loading on the limb, classification of no movement of the user's limb but forceful changes, and classification of the user lying down.

8. The system according to claim 1 or 2, wherein the activity classification model comprises a decision tree, a neural network or a support vector machine.

9. A system according to claim 1 or 2, wherein the activity classification model includes a first decision tree, which is configured to assign the first set of feature data to one of a plurality of different activity classifications, and the load shedding usage model includes a second decision tree, which is configured to determine whether the movement and orientation of the user sensor corresponds to the movement and orientation of the load shedding device sensor.

10. The system of claim 1 or 2, wherein the load relief device comprises removable plaster.

11. A computer-implemented method for collecting and processing sensor data to identify a risk of causing skin damage, the computer-implemented method comprising: receiving, via a computer network, sensor characteristic data representing outputs of a user sensor configured to be worn on a limb of a user; generating, by one or more computer processors, activity classification model output data using the sensor feature data and an activity classification model, the activity classification model output data representing a likelihood that the sensor feature data corresponds to each of a plurality of different activity classifications, The various activity categories include: a first activity classification, the first activity classification representing movement of the limb; and a second activity classification, the second activity classification indicating that the limb is not moving and weight is loaded on the limb; determining, by the one or more computer processors, an activity classification from the activity classification model output data; transmitting, via the computer network, display data representing the activity classification to a computing device configured to present the display data; receiving, via the computer network, second sensor characteristic data representing an output of a load-reducing device sensor coupled to a load-reducing device, the load-reducing device being configured to be worn on the limb; and determining, by the one or more computer processors, a load shedding usage classification using the sensor characteristic data, the second sensor characteristic data, and a load shedding usage model, Wherein the offload usage model is configured to determine the extent to which the movement and orientation of the user sensor corresponds to the movement and orientation of the offload device sensor.

12. The computer-implemented method of claim 11, further comprising identifying, by the one or more computer processors, a risk of causing a skin lesion, the skin lesion being a diabetic foot ulcer or a venous leg ulcer.

13. A computer-implemented method according to claim 11 or 12, wherein the determining the activity classification comprises determining that the activity classification model output data indicates that the sensor feature data corresponds to a third activity classification, and the third activity classification represents no movement but forceful changes in the user's limbs.

14. The computer-implemented method of claim 11 or 12, wherein said determining an activity classification comprises determining that said activity classification model output data indicates that said sensor feature data corresponds to a fourth activity classification representing that said user is lying down.

15. The computer-implemented method of claim 11 or 12, further comprising: determining, by the one or more computer processors, a subsequent activity classification using third sensor characteristic data and the activity classification model, the third sensor characteristic data representing a subsequent output of the user sensor; and A determination is made, by the one or more computer processors, from the activity classification and the subsequent activity classification that an activity of the user has changed.

16. The computer-implemented method of claim 11 or 12, further comprising: receiving, via the computer network, fourth sensor characteristic data representing a subsequent output of a load-reducing device sensor coupled to a load-reducing device, the load-reducing device being configured to be worn on the limb; determining, by the one or more computer processors, a subsequent load shedding usage classification using the fourth sensor characteristic data and the load shedding usage model; as well as A determination is made, by the one or more computer processors, from the load shedding usage classification and the subsequent load shedding usage classification that usage of the load shedding device by the user has changed.

17. The computer-implemented method of claim 11 or 12, further comprising loading, by the one or more computer processors, the activity classification model, the activity classification model comprising a decision tree, a neural network, or a support vector machine.

18. The computer-implemented method of claim 11 or 12, further comprising loading, by the one or more computer processors, the activity classification model comprising a decision tree configured to assign the sensor feature data to one of the plurality of different activity classifications.

19. A non-transitory computer readable medium having stored thereon an application for collecting and processing sensor data to identify a risk of causing skin damage, the application, when executed by one or more computer processors of a computing device, causing the computing device to: receiving, via a computer network, sensor characteristic data representing outputs of a user sensor configured to be worn on a limb of a user; generating activity classification model output data using the sensor feature data and the activity classification model, the activity classification model output data representing the likelihood that the sensor feature data corresponds to each of a plurality of different activity classifications, The various activity categories include: a first activity classification, the first activity classification representing movement of the limb; and a second activity classification, the second activity classification indicating that the limb is not moving and weight is loaded on the limb; determining an activity classification from the activity classification model output data; transmitting, via the computer network, display data representing the activity classification to a second computing device configured to present the display data, receiving, via the computer network, second sensor characteristic data representing an output of a load-reducing device sensor coupled to a load-reducing device, the load-reducing device being configured to be worn by the limb; and determining a load shedding usage classification using the sensor characteristic data, the second sensor characteristic data, and a load shedding usage model, Wherein the offload usage model is configured to determine the extent to which the movement and orientation of the user sensor corresponds to the movement and orientation of the offload device sensor.

20. The non-transitory computer-readable medium of claim 19, wherein the plurality of different activity classifications includes a third activity classification that indicates no movement of the limb but forceful changes in the limb.

21. The non-transitory computer-readable medium of claim 19 or 20, wherein the plurality of different activity classifications includes a fourth activity classification indicating that the user is lying down.

22. The non-transitory computer-readable medium of claim 19 or 20, wherein the activity classification model comprises a decision tree, a neural network, or a support vector machine.

23. The non-transitory computer-readable medium of claim 19 or 20, wherein the activity classification model comprises a decision tree configured to assign the sensor feature data to one of the plurality of different activity classifications.

24. The non-transitory computer-readable medium of claim 19 or 20, wherein the application, when executed by the one or more computer processors, causes the computing device to: determining a subsequent activity classification using third sensor characteristic data and the activity classification model, the third sensor characteristic data representing a subsequent output of the user sensor; and It is determined from the activity classification and the subsequent activity classification that an activity of the user has changed.

25. The non-transitory computer-readable medium of claim 19 or 20, wherein the application, when executed by the one or more computer processors, causes the computing device to: receiving, via the computer network, fourth sensor characteristic data representing a subsequent output of a load-reducing device sensor coupled to a load-reducing device configured to be worn by the limb; determining a subsequent load shedding usage classification using the fourth sensor characteristic data and the load shedding usage model; and A determination is made from the load shedding usage classification and the subsequent load shedding usage classification that usage of the load shedding device by the user has changed.

Images