Closed loop wearable sensor and method thereof

Abstract

A method and an electronic device for measuring motion and acoustic characteristics of a physiological process of a human subject. The method includes measuring motion and acoustic characteristics of a physiological process of a human subject; sending the first feature-related data to a machine learning service; sending the first feature-related data to a machine learning service; and performing machine learning operation on the feature related data to determine the prediction detection of the machine learning service on the human scratching activity.

Description

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This application claims the benefit of and priority to co-pending U.S. Patent Application No. 18 / 121,640, filed on March 15, 2023, and Provisional Application No. 63 / 321,737, filed on March 20, 2022, the contents of each of which are incorporated by reference into this application as if fully set forth herein. Technical Field

[0003] The embodiments described herein relate generally to systems and methods for sensing physiological parameters and, more particularly, but not exclusively, to systems and methods for detecting certain behaviors based on physiological parameters. Background Art

[0004] More than one in ten people suffer from acute or chronic itch. While itch causes significant morbidity, there is currently no technology that can accurately, objectively, and consistently assess itch by quantifying scratching (or other physiological symptoms or activities) in the user's natural environment. As a result, treatment options remain limited.

[0005] Additionally, existing technologies rely on transmitting sensor-acquired data to a remote location for analysis, which results in lags and delays in analytical performance.

[0006] Therefore, there is a need for systems and methods that overcome the shortcomings associated with the prior art. Summary of the invention

[0007] This summary is provided to introduce some concepts in a simplified form, which are further described in the detailed description below. This summary is not intended to identify or exclude key features or essential features of the claimed subject matter, nor is it intended to be used as an aid in determining the scope of the claimed subject matter.

[0008] According to one aspect, an embodiment relates to an electronic device for measuring motion and acoustic features of a physiological process of a human subject. The electronic device includes an inertial measurement unit (IMU) for detecting motion and acoustic features; a microcontroller unit (MCU) communicatively coupled to the IMU; wherein the IMU is operable to directly mechanically communicate with the subject's skin; wherein the MCU calculates feature-related data based on the motion and acoustic features; wherein the MCU sends the feature-related data to a machine learning service executed on the MCU; and the machine learning service predicts detection of human scratching activity by the human subject by performing a first machine learning operation on the feature-related data.

[0009] In some embodiments, the electronic device further comprises a vibration motor for reminding a human subject. In some embodiments, when a human scratching activity is detected, the MCU turns on the vibration motor.

[0010] In some embodiments, the MCU calculates second feature-related data based on the motion and acoustic features. In some embodiments, the second feature-related data is used to determine the intensity of the detected scratching activity. In some embodiments, the second feature-related data is used to predict that the human subject is in a sleeping state. In some embodiments, if the subject is predicted to be in a sleeping state, a second machine learning operation is used to detect the scratching activity. In some embodiments, the mode of the vibration motion operation is determined based on whether the subject is predicted to be sleeping.

[0011] In some embodiments, the detection of the human body scratching activity includes detecting the start of the scratching activity, which is associated with time 1; detecting the end of the scratching activity, which is associated with time 2. In some embodiments, the MCU calculates the scratching duration by subtracting time 1 from time 2.

[0012] According to another aspect, an embodiment is directed to a method that includes measuring motion and acoustic features of a physiological process of a human subject; calculating first feature-related data based on the motion and acoustic features; sending the first feature-related data to a machine learning service; and determining a predicted detection of a human scratching activity by the machine learning service by performing a machine learning operation on the feature-related data.

[0013] In some embodiments, the second feature-related data is calculated from the motion and acoustic features. In some embodiments, the second feature-related data is used to predict that the human subject is in a sleeping state.

[0014] In some embodiments, the method further includes providing tactile feedback to the human subject upon determining the predicted human scratching activity detection.

[0015] In some embodiments, measuring the motion and acoustic characteristics involves measuring the characteristics via an inertial measurement unit, and determining the detection involves executing a microcontroller unit, wherein the inertial measurement unit and the microcontroller are located on the sensor.

[0016] In some embodiments, the machine learning operation includes at least one of a neural network, a logistic regression classifier, and a random forest classifier. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Non-limiting and non-exhaustive embodiments of the present invention are described with reference to the following figures, wherein like reference numerals refer to like parts throughout the various views unless otherwise specified.

[0018] Figure 1 shows a sensor view according to one embodiment;

[0019] Figure 2 According to an embodiment Figure 1 Schematic diagram of the sensor's electronic board;

[0020] Figure 3 A diagram showing the relationship between force and scratching according to one embodiment;

[0021] Figure 4 shows a graphical representation of different scratching activities detected over a period of time according to one embodiment;

[0022] Figure 5 shows a workflow for analyzing feature-related data according to one embodiment;

[0023] Figure 6 shows a diagram of a Sigmoid function according to one embodiment;

[0024] Figure 7 A random forest classifier according to one embodiment is shown; and

[0025] Figure 8 A flow chart of a method for monitoring user activity according to one embodiment is depicted. DETAILED DESCRIPTION

[0026] Various embodiments are described more fully below with reference to the accompanying drawings, which form a part of this document and show specific exemplary embodiments. However, the concepts of the present disclosure can be implemented in many different forms and should not be interpreted as being limited to the embodiments described herein; on the contrary, the embodiments provided should be part of a thorough and complete disclosure to fully convey the scope of the concepts, techniques, and embodiments of the present disclosure to those skilled in the art. The embodiments may be practiced as methods, systems, or devices. Accordingly, the embodiments may take the form of hardware embodiments, fully software embodiments, or embodiments combining software and hardware aspects. Therefore, the detailed description below should not be understood in a limiting sense.

[0027] References in the specification to "one embodiment" or "an embodiment" mean that a particular feature, structure, or characteristic described in connection with the embodiment is included in at least one embodiment or technique according to the present disclosure. The phrase "in one embodiment" appearing in different places in the specification does not necessarily refer to the same embodiment. The phrase "in some embodiments" appearing in different places in the specification does not necessarily refer to the same embodiment.

[0028] Some portions of the following description are presented in terms of symbolic representations of operations on non-transient signals stored in computer memories. These descriptions and representations are used by those skilled in the art of data processing to most effectively convey the substance of their work to others skilled in the art. Such operations often require physical manipulation of physical quantities. Typically, although not necessarily, these quantities take the form of electrical, magnetic or optical signals capable of being stored, transmitted, combined, compared and otherwise manipulated. Sometimes, primarily for common reasons, it is convenient to refer to these signals as bits, values, elements, symbols, characters, terms, numbers, etc. In addition, it is sometimes convenient to refer to certain arrangements of steps requiring physical manipulation of physical quantities as modules or code devices without loss of generality.

[0029] However, all of these and similar terms are associated with the appropriate physical quantities and are merely convenient labels applied to these quantities. Unless otherwise clearly stated from the discussion below, it should be understood that throughout the description, discussions using terms such as "processing" or "computing" or "calculating" or "determining" or "displaying" refer to the actions and processes of a computer system or similar electronic computing device. A device that operates and transforms data represented by physical (electronic) quantities in a computer system memory or register or other such information storage, transmission or display device. Portions of this disclosure include processes and instructions that may be embodied in software, firmware, or hardware, and when embodied in software, may be downloaded to reside on and operate from different platforms used by various operating systems.

[0030] The present disclosure also relates to an apparatus for performing the operations described herein. The apparatus may be specially constructed for the desired purpose, or it may include a general-purpose computer selectively activated or reconfigured by a computer program stored in the computer. Such a computer program may be stored in a computer-readable storage medium, such as, but not limited to, any type of disk, including a floppy disk, an optical disk, a CD-ROM, a magneto-optical disk, a read-only memory (ROM), a random access memory (RAM), an EPROM, an EEPROM, a magnetic card or an optical card, an application-specific integrated circuit (ASIC), or any type of medium suitable for storing electronic instructions, and each may be coupled to a computer system bus. In addition, the computer referred to in this specification may include a single processor or may be an architecture that uses a multi-processor design for increasing computing power.

[0031] The processing and display provided herein are not inherently associated with any particular computer or other device. Various general purpose systems may also be used together with the program according to the teachings of this article, or it may be convenient to construct a more specialized device to perform one or more method steps. The structure of various systems will be discussed below. In addition, any specific programming language sufficient to implement the technology and implementation of the present disclosure may be used. A variety of programming languages ​​may be used to implement the present disclosure discussed herein.

[0032] Furthermore, the language used in the specification is primarily selected for readability and instructional purposes, and may not be selected to describe or limit the subject matter disclosed.Accordingly, this disclosure is intended to illustrate but not to limit the scope of the concepts discussed herein.

[0033] Pruritus, or more informally "itch", is a common symptom of several diseases. These diseases can range from skin diseases to systemic diseases such as, but not limited to, renal failure, urticaria, diabetes, chronic pruritus in the elderly, and malignancies (e.g., cutaneous T-cell lymphoma CTCL).

[0034] Acute and chronic pruritus each affect 8% to 14% of the population and can adversely affect quality of life. The itch-scratch cycle often starts with itching, which leads to scratching, which in turn leads to further itching, inflammation, and infection.

[0035] This scratching behavior can become habitual. Habitual scratching can be particularly troublesome if it occurs during sleep, which is important for normal developmental growth, physical growth, mental health, and general well-being. However, nighttime scratching can disrupt sleep quality and cause, or at least contribute to, the onset of a sleep disorder.

[0036] Existing techniques for assessing itch include administering or otherwise providing patients with subjective surveys that include numeric rating scales. However, correlation between actual scratching activity and observed scratching behavior is poor. Furthermore, the results of these surveys may be of questionable value because different patients may have different tolerance or sensitivity to different levels of itch severity.

[0037] Another existing method for assessing itch involves directly visualizing scratching using video recordings. However, this is time-consuming, impractical, and disruptive to the user.

[0038] The embodiments described herein provide novel systems and methods for assessing itch, detecting scratching activity, and providing therapeutic biofeedback. Figure 1An acoustic electromechanical wearable sensor device 100 is shown according to one embodiment. The sensor 100 can be operably positioned relative to a person's skin to collect physiological data, behavioral data, etc. Specifically, the sensor 100 can capture low-frequency data corresponding to motion and high-frequency data corresponding to various clinical biomarkers.

[0039] The sensor 100 may include a base portion 102, an electronic board 104, and a top portion 106. The base portion 102 may include or be configured as a waterproof silicone portion that can be attached to a user's hand or other body location. In addition, the base portion 102 should be soft and flexible so that it can adapt to the movement of the user's hand and maintain operable contact with the user's hand during movement. The base portion 102 may maintain operable contact with the user due at least in part to an adhesive.

[0040] In some embodiments, the base portion 102 can be attached to the back of the user's hand. For example, the base portion 102 can be located on the metacarpal bones associated with the user's index and middle fingers. In this location, the sensor 100 can detect low-frequency and high-frequency signals generated by the user making scratching or rubbing movements with one or more of their fingers, wrists, hands, etc.

[0041] Upon predictive detection of scratching activity (e.g., determining that the user is engaging in scratching activity or at least is likely to engage in scratching activity), the sensor may provide tactile feedback to the user. This tactile feedback may alert the patient to the scratching activity (including involuntary scratching activity) to encourage the user to stop scratching.

[0042] Electronic board 104 may include or otherwise support several components. These may include, but are not limited to, a microcontroller, a power source 108, an accelerometer 110, a tactile feedback provider 112 (such as a tactile motor), and a wireless charging circuit 114. Thus, sensor 100 provides multiple hardware components in a compact space to minimize form factor and maximize usability and utility of the device while having low power consumption and longer operating time.

[0043] The top 106 may be made of or otherwise include a soft, stretchable, waterproof material. The top 106 should also be able to bend to accommodate the movement of the user's hands or other body parts.

[0044] Figure 2 According to an embodiment Figure 1 Schematic diagram 200 of the electronic board 104 . Figure 1 The components shown in Figure 2 The same reference numerals are used in the drawings. Figure 2The electronic board 104 in may include a microcontroller unit (MCU) 202 , an inertial measurement unit (IMU) 204 , a power source 108 , a power management unit ( 206 ), a storage device 210 , and a tactile feedback provider 212 .

[0045] MCU 202 may execute or otherwise include memory 214, interface 216, one or more machine learning models 218, and general purpose input output (GPIO) 220. MCU 202 may be located on-board with sensor 100 to perform desired analysis on sensor 100.

[0046] The IMU 204 may include or otherwise implement one or more accelerometers 110. The accelerometer 110 may measure acceleration along the x-axis and y-axis at 200 Hz. The accelerometer 110 may measure acceleration up to 1600 Hz on the z-axis. Acceleration in these directions may indicate motion (e.g., scratching activity).

[0047] The MCU 202 includes various combinations of resistors, transistors, and diodes to form any number of logic gates to execute the appropriate program. The MCU 202 can store any combination of types of memory 214 (e.g., RAM memory configurations). As described above, the memory 214 can include non-volatile memory (such as flash, EPROM, EEPROM, ROM, and PROM) or volatile memory (such as static or dynamic RAM). The exact configuration / type of on-board memory 214 can of course vary, as long as the instructions for analyzing the data from the IMU 204 can be executed by the MCU 202.

[0048] The machine learning model 218 may execute one or more of the various machine learning models 218 to analyze the data collected and provided by the IMU 204. One or more of the machine learning models 218 may have been trained based on data from a clinical study that monitored scratching activity.

[0049] For example, data on scratching activity events may be collected from various body parts over a period of time. These body parts may include, but are not limited to, the back of the hand, head, forearm, inner elbow, thigh, and calf. Non-scratching movements may also be considered during the training phase and may include activities such as waving, texting, tapping, and movements associated with fidgeting.

[0050] Upon detecting the scratching activity, the GPIO 220 may transmit a signal to the tactile feedback provider 212. The tactile feedback provider 212 may then provide tactile feedback to the user. The tactile feedback provider 112 may include an eccentric rotating mass (ERM) motor 222 to generate vibrations.

[0051] In some embodiments, the motor 222 may be configured to rotate at a speed of 10,000 rpm to generate a vibration feedback of 1.4 G. Regardless of the exact amount of vibration feedback provided, the vibration feedback should be sufficient to alert the user even during sleep.

[0052] The tactile feedback provider 112 can be powered by a voltage regulator 230 controlled by a (GPIO) pin 220 of the MCU 202. For example, the voltage regulator 230 can use pulse width modulation to create feedback of variable intensity, duration, or some combination thereof. Pulse width modulation uses pulses of different widths to produce cycles that can modulate the speed of the motor 222. Using this configuration, the tactile feedback provider 112 can generate several types of tactile profiles. For example, the tactile feedback provider 112 can generate profiles with slight vibrations, strong vibrations, ramp effects, short pulses, etc.

[0053] The power source 108 may be any type of power source that provides power to the components of the board 104. In some embodiments, the power source 108 may be a rechargeable lithium polymer or lithium ion battery.

[0054] Storage device 210 may store data about the user and signals appropriate for the user. For example, a particular user may require a strong and sudden tactile signal to detect a tactile signal. Thus, the tactile profile or type of tactile feedback provided may vary and may depend on the user, the message to be conveyed, etc. For example, the feedback may vary in length, intensity, pattern, or some combination thereof.

[0055] Similarly, users can select their most preferred tactile feedback to minimize aggravation or alleviate habituation. For example, some users may not notice or respond to low-intensity vibrations from the tactile feedback provider 212. Therefore, some users may prefer to receive higher-intensity vibrations.

[0056] Additionally or alternatively, some users may find high intensity vibrations irritating or disruptive. Accordingly, some users may prefer to receive lower intensity vibrations.

[0057] Data on sleeping patterns, scratching activity, and other user behaviors may have been collected in advance. As part of the training phase, the user wearing the sensor may have been asked to move the force sensor ( Figure 1 Scratching activity is performed on the surface (not shown).

[0058] For example, a user may wear the sensor and perform a scratching motion on a force sensor resistor ("FSR"). The surface of the force sensor may be covered with aluminum foil that makes the coefficient of friction ("COF") of its surface similar to that of the back of a hand ("DH"). The FSR may detect the force applied to its surface during the test scratching activity (i.e., when the user is scratching the surface of the FSR). While the user wears the sensor during the test scratching activity, the sensor's accelerometer may simultaneously collect acceleration data associated with the user's motion.

[0059] The applied force can indicate the intensity level of the scratching activity. That is, the higher the resultant force detected from the scratching activity, the more intense the scratching activity.

[0060] Figure 3 A graph 300 showing the relationship between frequency and force is shown. Graph 300 also shows a Spearman correlation showing the relationship between scratching intensity (measured in force (N)) and the sensor signal. For each trial, the speed of the scratch and the angle between the finger and the FSR surface remain constant. The power of the scratching signal collected by the sensor can then be calculated. As the scratching intensity increases, the signal power between 40 Hz and 60 Hz will also increase.

[0061] Figure 4 A graph 400 of different scratching activities detected over a period of time is presented. As shown in the graph 400, acceleration data of scratching activities of different intensities are detected over a period of time. As the scratching intensity increases, the signal power between 40 Hz and 60 Hz also increases.

[0062] In operation, the IMU 204 can collect data about the user's activities. For example, as described above, the accelerometer 110 can collect acceleration data on various axes to identify behaviors that may indicate scratching activity.

[0063] The machine learning model 218 may execute one or more of the various machine learning models 218 to analyze data from the IMU 204. One or more of the machine learning models 218 may have been trained on data from a clinical study that monitored scratching activity.

[0064] For example, scratching activity training events can be collected from healthy individuals who are at least 18 years old and have no skin allergies to adhesive or current skin stimulation near the expected sensor site to develop a machine learning algorithm. In order to detect these scratching activity events, sensor 100 can be placed on the back of the hand of the participant along the second metacarpal bone of the participant using skin adhesive. Each participant can perform 35 seconds of scratching activity and 65 seconds of non-scratching activity. The participant can perform multiple scratching activities on clothes (for example, on their abdomen, thigh, calf, knee and shoulder), and can perform multiple scratching activities directly on the skin (for example, on their cheek, head, inner elbow, outer elbow, forearm, back of hand and palm).

[0065] The machine learning models 218 may include one or more random forest classifiers, logistic regression, neural networks, etc. The exact type of machine learning models 218 used may vary, as long as they are capable of analyzing data from the IMU 204 to identify scratching activity.

[0066] Figure 5 A workflow 500 for analyzing data from an IMU 204 is shown according to one embodiment. One or more machine learning models 218 may execute the workflow 500 or at least a portion of the workflow 500. The workflow 500 or at least a portion of the workflow 500 may be used to train one or more aspects of a machine learning model 218. The particular model may include a signal processing block 502, a convolutional recurrent neural network ("CRNN") 504, a classification module 506, and a clustering module 508.

[0067] Training data obtained from a subject performing scratching and non-scratching behaviors may first be sent to processing block 502. This may be z-axis acceleration data, and a sliding window module 510 may divide the acceleration data into multiple groups. For example, in some embodiments, the raw time series data of z-axis acceleration may be segmented into frames of 320 milliseconds (512 data points) using a window with 75% overlap between adjacent frames. A fast Fourier transform module 512 may then calculate short-time Fourier transform ("STFT") features.

[0068] The CRNN 504 may analyze the temporal pattern of the signal through one or more 1-D convolutional layers 514, linear layers 516, and gated recurrent unit ("GRU") 518 layers to receive and process these features.

[0069] The output of the CRNN 504 can then be provided to a fully connected neural network 520 for classification. For example, the predicted output can be clustered using a density-based spatial clustering of applications with noise (DBSCAN) algorithm. For example, a sigmoid function 522 can classify an activity as a scratching activity or a non-scratching activity based on the associated feature data.

[0070] Figure 6 A graphical representation of a sigmoid function 600 according to one embodiment is shown. The sigmoid function 600 can represent the probability that a particular activity indicates a scratching activity based on feature-related data on the x-axis. The x-axis can correspond to acceleration data obtained from the IMU 204, for example.

[0071] A user or administrator (eg, medical staff) may specify the decision boundary 602 so that features that place activities to the right of the boundary 602 have a higher probability of being scratching activities than non-scratching activities. Therefore, the classification module 506 may classify the activity as scratching activities.

[0072] In other words, sigmoid function 600 represents the probability that a particular activity represents a scratching activity as determined by data from IMU 204. Activity having features to the right of boundary 602 may be classified as a scratching activity, while activity having features to the left of boundary 602 may be classified as a non-scratching activity.

[0073] Workflow 500 can be used to classify activities at least substantially in real time. S-shaped function 522 can then classify the activity as scratching or non-scratching based on the output from the neural network. Clustering module 506 can then cluster feature-related data by, for example, executing a DBSCAN algorithm.

[0074] As an aspect, the embodiments described herein allow for in-sensor computation of a CRNN model using edge computing. Based on the developed CRNN model, the sensor 100 can use a fixed-size queue of 512 data points (320 milliseconds, 1600Hz sampling rate) to replicate frames of the CRNN model. Once the queue is full, the data contained in the queue can be passed to the signal processing block to calculate the STFT. The STFT can then be passed to the CRNN model. For each new input data point, the oldest data point is deleted and the new data point is added to the queue via a first-in, first-out (FIFO) scheme. The calculation of scratch detection can then be repeated for every 128 new data points.

[0075] As another example, embodiments herein may use a random forest-based machine learning model to classify an activity as a scratching activity. Figure 7 A random forest classifier 700 is shown receiving feature correlation data from the IMU 204 .

[0076] The random forest classifier 700 may include multiple trees, each of which is configured to detect whether certain features are present or otherwise associated with the identified activity. For example, tree 1 may first consider whether an activity such as acceleration above a certain threshold on the z-axis is detected (e.g., whether the user may be sleeping, whether the acceleration has occurred a threshold number of times, etc.). If so, the classifier 700 may traverse a certain branch of the tree.

[0077] Certain nodes of the tree are dimmed to indicate the presence or absence of certain features. Figure 7 Three trees are shown in , but any number of trees can be used to consider any number of features.

[0078] Each tree can output a classification decision about whether the activity is indicative of scratching. These predictions are based on the presence or absence of certain features.

[0079] The classification decisions of one or more trees may be combined in step 704. The classification decision is then provided in step 706.

[0080] The above examples and illustrations of machine learning models are exemplary. Other types of machine learning models may be used, such as support vector machines, models based on logistic regression, etc.

[0081] Return to reference Figure 2 When a scratching activity is detected, the GPIO 220 may send a signal to the tactile feedback provider 212 to provide tactile feedback to the user. For example, the motor 222 may cause the sensor 100 to vibrate in a certain form, intensity, duration, etc.

[0082] The user may detect or feel the vibration, which may be associated with the scratching activity. For example, the vibration may alert the user that they are scratching themselves and may encourage the user to stop scratching.

[0083] As described above, the user may be sleeping while wearing the sensor 100. During sleep, the user may scratch themselves subconsciously. Nevertheless, the sensor can provide tactile feedback that encourages the user to stop the scratching activity, although at least partially subconscious.

[0084] Figure 8 Depicted is a flow diagram of a method 800 for monitoring user activity according to one embodiment. Figure 1 The sensor 100 or components thereof may perform one or more steps of the method 800 .

[0085] Step 802 includes measuring motion and acoustic characteristics of a physiological process of a human subject. The human subject (referred to as a "user") may be diagnosed with a medical disease with itching as a symptom. Therefore, the user may be inclined or wish to scratch one or more parts of their body. As previously mentioned, scratching may aggravate the symptoms or condition.

[0086] The user also does not need to be diagnosed. Instead, the user may want to inhibit their scratching activity on their own, even without a diagnosis from a medical professional.

[0087] The user may place a sensor such as sensor 100 at a location on their body (e.g., on the back of their hand between the second and third metacarpal bones). Alternatively, a medical professional may place the sensor on the user.

[0088] The sensor may continuously measure motion and acoustic characteristics of a human subject's physiological processes (e.g., during sleep, during a medical examination, etc.) For example, a user may want to prevent scratching activity during sleep.

[0089] Step 804 includes calculating first feature-related data based on the motion and acoustic features. Figure 2 The MCU 202 of the MCU may analyze the received raw data regarding motion and acoustics associated with the user.

[0090] Step 806 includes sending the first feature-related data to a machine learning service. The machine learning service may include any type of model previously discussed, as well as any other type of model now available or developed later.

[0091] Step 808 includes determining, by the machine learning service, detection of a human scratching activity by performing a machine learning operation on the feature-related data. Step 808 may include executing one or more machine learning models to determine whether the feature-related data indicates or at least suggests that the user is performing a scratching activity. For example, the machine learning model may rely on previously collected data associated with a known scratching activity.

[0092] Once scratching activity is detected, step 810 includes providing tactile feedback to the user to encourage them to stop the scratching activity - whether they are sleeping or awake. Thus, embodiments herein provide a therapeutic biofeedback tool for itch.

[0093] Applying low-power, high-fidelity, AI-based analytics to epidermal sensors with skin-friendly adhesives, such as sensor 100 for extended wear, maximizes clinical potential and relevance. Integration of machine learning analytics with tactile feedback on sensor 100 provides a closed-loop modality to quantify scratching events and provide real-time accurate feedback to improve patient outcomes. This technology can serve as a platform to triage and provide feedback to patients with a variety of skin conditions and complications, and help guide and evaluate the efficacy of treatment plans.

[0094] The methods, systems, and devices discussed above are examples. Various configurations may omit, replace, or add various processes or components as appropriate. For example, in alternative configurations, the methods may be performed in an order different from that described, and various steps may be added, omitted, or combined. In addition, features described with respect to certain configurations may be combined in various other configurations. Different aspects and elements of the configurations may be combined in a similar manner. In addition, technology is developing, and therefore, many elements are examples and do not limit the scope of the present disclosure or claims.

[0095] For example, the block diagrams and / or operating instructions of the methods, systems and computer program products according to the embodiments of the present disclosure are described above. The functions / actions marked in the boxes may not occur in the order shown in any flow chart. For example, two boxes shown in succession can actually be executed substantially simultaneously, or these boxes can sometimes be executed in reverse order, depending on the functions / actions involved. Additionally or alternatively, not all blocks shown in any flow chart need to be executed and / or implemented. For example, if a given flow chart has five blocks containing functions / actions, it may be the case that only three of the five blocks are executed and / or implemented. In this example, any three boxes in the five boxes can be executed and / or implemented.

[0096] A statement that a value exceeds (or is greater than) a first threshold is equivalent to a statement that the value reaches or exceeds a second threshold that is slightly greater than the first threshold (e.g., in the resolution of the relevant system, the second threshold is one value higher than the first threshold). A statement that a value is less than (or is at) a first threshold is equivalent to a statement that the value is less than or equal to a second threshold that is slightly less than the first threshold (e.g., in the resolution of the relevant system, the second threshold is one value lower than the first threshold).

[0097] Specific details are given in the description to provide a comprehensive understanding of example configurations (including implementations). However, the configuration can be implemented without these specific details. For example, well-known circuits, processes, algorithms, structures and techniques have been shown without unnecessary details to avoid obscuring the configuration. This description only provides example configurations and does not limit the scope, applicability or configuration of the claims. On the contrary, the above description of the configuration will provide a description of implementing the technology for those skilled in the art. Various changes can be made to the functions and arrangements of the elements without departing from the spirit or scope of the present disclosure.

[0098] Several example configurations have been described, and various modifications, alternative constructions, and equivalents may be used without departing from the spirit of the invention. For example, the above elements may be components of a larger system, where other rules may take precedence over or otherwise modify the application of various embodiments or techniques of the present disclosure. In addition, many steps may be taken before, during, or after considering the above elements.

[0099] Having provided the description and illustrations of the present application, those skilled in the art may envision changes, modifications and alternative embodiments that are incorporated into the general inventive concept discussed in the present application without departing from the scope of the claims.

Claims

1. An electronic device for measuring motion and acoustic characteristics of a physiological process of a human subject, comprising: An inertial measurement unit (IMU) for detecting motion and acoustic signatures; a microcontroller unit (MCU), the microcontroller unit (MCU) being communicatively coupled to the IMU; in, The IMU is operable to be in direct mechanical communication with the subject's skin; wherein the MCU calculates feature-related data based on the motion and acoustic features; The MCU sends the feature-related data to a machine learning service executed on the MCU; and The machine learning service predicts detection of human scratching activity by a human subject by performing a first machine learning operation on feature-related data.

2. The device according to claim 1, wherein: The electronic device also includes a vibration motor for alerting the human subject.

3. The device according to claim 2, wherein: When human scratching activity is detected, the MCU turns on the vibration motor.

4. The device according to claim 1, wherein: The MCU calculates second feature related data according to the motion and acoustic features.

5. The device according to claim 4, wherein: The second feature-related data is used to predict that the human subject is in a sleeping state.

6. The device according to claim 5, wherein: The second machine learning operation is used to detect scratching activity when the subject is predicted to be sleeping.

7. The device according to claim 5, wherein: A pattern of vibration motor operation is determined based on whether the subject is predicted to be in a sleeping state.

8. The device according to claim 1, wherein: Detecting a human scratching activity includes detecting a start of the scratching activity, associating time 1 with the start of the scratching activity, detecting an end of the scratching activity, and associating time 2 with the end of the scratching activity.

9. The device according to claim 8, wherein: The MCU calculates the scratching duration by subtracting time 1 from time 2.

10. The device according to claim 4, wherein: The second feature related data is used to determine the intensity of the detected scratching activity.

11. A method comprising: measuring motion and acoustic signatures of physiological processes in human subjects; calculating first feature-related data based on the motion and acoustic features; Sending the first feature-related data to a machine learning service; as well as Determine a predicted detection of a human scratching activity by the machine learning service by performing a machine learning operation on the feature-related data.

12. The method according to claim 11, wherein: Second feature related data is calculated based on the motion and acoustic features.

13. The method according to claim 12, wherein: The second feature-related data is used to predict that the human subject is sleeping.

14. The method according to claim 11, further comprising: Tactile feedback is provided to the human subject upon detection of human scratching activity.

15. The method according to claim 11, wherein: Measuring the motion and acoustic characteristics includes measuring the characteristics via an inertial measurement unit, and determining the predicted detection involves executing a microcontroller unit, wherein the inertial measurement unit and the microcontroller are located on the sensor.

16. The method according to claim 11, wherein: The machine learning operation includes at least one of a neural network, a logistic regression classifier, and a random forest classifier.

Images