Longitudinal monitoring of hypertonia through a multimodal sensing device

Abstract

A device may include a multimodal sensing device comprising one or more sensors for sensing motion data from a subject. A device may include a non-transitory computer-readable storage medium in communication with one or more processors and the multimodal sensing device which cause the one or more processors to perform operations comprising: receiving, from the one or more sensors of the multimodal sensing device, at least one signal comprising motion data of the subject and motion data of the multimodal sensing device, wherein the at least one signal corresponds to a standardized flexion and extension maneuver exerted on the subject by an evaluator wearing the multimodal sensing device, determining a torque score based on the received at least one signal; and determining an effectiveness of a treatment or an estimated progression of disease based on a comparison of the determined torques score and at least one historical torque score.

Description

LONGITUDINAL MONITORING OF HYPERTONIA THROUGH A MULTIMODALSENSING DEVICECROSS REFERENCE TO RELATED APPLICATIONS

[0001] This application claims benefit and priority to U.S. Provisional Application No. 63 / 645,558, filed May 10, 2024, entitled, “Longitudinal Monitoring of Hypertonia Through A Multimodal Sensing Glove,” which is incorporated herein by reference in its entirety.STATEMENT OF GOVERNMENT SUPPORT

[0002] This invention was made with government support under CBET 2054517 awarded by the National Science Foundation. The government has certain rights in the invention.BACKGROUND

[0003] Hypertonia refers to an abnormal increase in muscle tone due to upper motor neuron lesions which can arise from medical conditions such as cerebral palsy, neurodegenerative diseases, stroke, traumatic brain injuries, spinal injuries and the like. Hypertonia can impair motor control and balance, thus reducing quality of life for patients. Treatment of hypertonia and the monitoring treatment progress is limited in precision and consistency because clinical evaluations are based on subject perception ratings such as the Modified Ashworth Scale (MAS) or the Tardieu Scale. The subjectiveness of these ratings hinders efforts to understand the efficacy of treatments provided to those with hypertonia such as medication and therapy.SUMMARY

[0004] In some example embodiments, a multimodal sensing device such as a sensing glove may produce signal data that can be used to monitor certain patient conditions, such as hypertonia. Additionally, the systems and methods described herein can be used for calibration of a multimodal sensing device. In some implementations, the multimodal sensing device can be formed in a sensing glove. The systems and methods described herein can provide an objective tool for the long term monitoring hypertonia including for determining the efficacy of treatment and / or disease progression.In some aspects, the techniques described herein relate to a system including: a multimodal sensing device including one or more sensors for sensing motion data from a subject; a non- transitory computer-readable storage medium in communication with one or more processors and the multimodal sensing device, the non-transitory computer-readable storage medium having instructions stored thereon which, when executed by the one or more processors, cause the one or more processors to perform operations including: receiving, from the one or more sensors of the multimodal sensing device, at least one signal including motion data of the multimodal sensing device, wherein the at least one signal corresponds to a standardized flexion and extension maneuver exerted on the subject by an evaluator wearing the multimodal sensing device; determining a torque score based on the received at least one signal; and determining an effectiveness of a treatment or an estimated progression of disease based on a comparison of the determined torques score and at least one historical torque score.In some aspects, the techniques described herein relate to a system, wherein the motion data of the multimodal sensing device includes a force expended by the evaluator on the subject .In some aspects, the techniques described herein relate to a system, wherein the motion data of the multimodal sensing device includes a maneuver trajectory of the standardized flexion and extension maneuver exerted on the subject by the evaluator.In some aspects, the techniques described herein relate to a system, wherein determining the torque score based on the received at least one signal corresponding to the standardized flexion and extension maneuver includes: determining a subset of the received at least one signal corresponding to the standardized flexion maneuver; determining a subset of the received at least one signal corresponding to the standardized extension maneuver; and determining the torque score based on a force corresponding to the standardized flexion maneuver based on the determined subset of the received at least one signal corresponding to the standardized flexion maneuver, a force corresponding to the standardized extension maneuver based on the determined subset of the received at least one signal corresponding to the standardized extension maneuver, a grip length and angle of treatment.In some aspects, the techniques described herein relate to a system, wherein determining the torque score further includes: determining a heatmap for the median torque score across one or more geographical areas of the sensing glove.In some aspects, the techniques described herein relate to a system, further including: a graphical user interface communicatively coupled to the one or more processors.In some aspects, the techniques described herein relate to a system, wherein the graphical user interface is configured to: receive user input of at least one trajectory corresponding to maneuver or a numerical count of the maneuvers; and display the determined torque score.In some aspects, the techniques described herein relate to a system, wherein the graphical user interface is configured to display one or more visual representations of the determined torque score and the at least one historical torque score.In some aspects, the techniques described herein relate to a system, wherein the graphical user interface is configured to display a calibrated torque score and / or torque signal.In some aspects, the techniques described herein relate to a system, further including: receiving motion data for a calibration activity; and determining a calibration factor based on a torque score for the received motion data for the calibration activity.In some aspects, the techniques described herein relate to a system, wherein applying the determined calibration factor to one or more future signals received from the multimodal device.In some aspects, the techniques described herein relate to a system, wherein the treatment includes a pharmaceutical drug and a dosage, and / or physical therapy.In some aspects, the techniques described herein relate to a system, wherein the at least one historical torque score includes one or more torque scores associated with the subject stored in a database communicatively coupled to the one or more processors.In some aspects, the techniques described herein relate to a system, wherein the multimodal sensing device includes: at least one wearable item; at least one communication pathway configured to communicate with the one or more processors; a sensor array disposed in the at least one wearable item; and an inertial measurement unit including the at least one sensor.In some aspects, the techniques described herein relate to a system, wherein the at least one sensor includes at least one of an accelerometer, a gyroscope, or a magnetometer.In some aspects, the techniques described herein relate to a system, wherein the motion data includes measurements of at least one of the following: at least one magnetic field, linear acceleration, angular acceleration, linear velocity, or angular velocity.In some aspects, the techniques described herein relate to a method including: receiving, at a processor in communication with a multimodal sensing device and from one or more sensors of the multimodal sensing device, at least one signal including motion data of the multimodal sensing device; determining a torque score based on the received at least one signal corresponding to a standardized flexion and extension maneuver exerted on the subject by an evaluator wearing the multimodal sensing device; and determining an effectiveness of a treatment or an estimated progression of disease based on a comparison of the determined torques score and at least one historical torque score.In some aspects, the techniques described herein relate to a method, wherein the motion data of the multimodal sensing device includes a maneuver trajectory of the standardized flexion and extension maneuver exerted on the subject by the evaluator and / or a force expended by the evaluator on the subject.In some aspects, the techniques described herein relate to a method, wherein determining the torque score based on the received at least one signal corresponding to the standardized flexion and extension maneuver includes: determining a subset of the received at least one signal corresponding to a flexion maneuver; determining a subset of the received at least one signal corresponding to an extension maneuver; and determining the torque score based on a force corresponding to the flexion maneuver based on the determined subset of the received at least one signal corresponding to a flexion maneuver, a force corresponding to the extension maneuver based on the determined subset of the received at least one signal corresponding to the extension maneuver, a grip length, or an angle of treatment.In some aspects, the techniques described herein relate to a method including: receiving motion data for a calibration activity; and determining a calibration factor based on a torque score for thereceived motion data for the calibration activity; and applying the determined calibration factor to one or more future signals received from the multimodal device.

[0005] The details of one or more variations of the subject matter described herein are set forth in the accompanying drawings and the description below. Other features and advantages of the subject matter described herein will be apparent from the description and drawings, and from the claims.BRIEF DESCRIPTION OF THE DRAWINGS

[0006] The accompanying drawings, which are incorporated in and constitute a part of this specification, show certain aspects of the subject matter disclosed herein and, together with the description, help explain some of the principles associated with the disclosed implementations.

[0007] In the drawings:

[0008] FIG. 1 A illustrates a multimodal sensing device in accordance with some embodiments of the present disclosure;

[0009] FIG. IB provides a second view of the multimodal sensing device of FIG. lA in accordance with some embodiments of the present disclosure;

[0010] FIG. 2 provides a system diagram for a system that monitors and assesses treatment and / or severity of hypertonia in a patient in accordance with some embodiments of the present disclosure;

[0011] FIG. 3 provides a flow diagram of a method for a system that monitors and assesses treatment and / or severity of hypertonia in a patient in accordance with some embodiments of the present disclosure;

[0012] FIG. 4 illustrates examples of the signals received at the processor from the multimodal sensing device in accordance with some embodiments of the present disclosure;

[0013] FIG. 5 illustrates torque scores for a multimodal sensing device in accordance with some embodiments of the present disclosure;

[0014] FIG. 6 illustrates a change in torque scores observed by a multimodal sensing device in accordance with some embodiments of the present disclosure;

[0015] FIG. 7 provides a flow-chart for a method for calibrating a multimodal device in accordance with some embodiments of the present disclosure;

[0016] FIG. 8 provides an example of the calibration process in accordance with some embodiments of the present disclosure;

[0017] FIG. 9 illustrates pre- and post- calibration values in accordance with some embodiments of the present disclosure;

[0018] FIGS. 10- 12 illustrate various examples of a graphical user interface and display in communication with a multimodal sensing device in accordance with some embodiments of the present disclosure;

[0019] FIG. 13 illustrates data received from a multimodal sensor device and a calibration curve for the multimodal sensor device in accordance with some embodiments of the present disclosure;

[0020] FIG. 14 illustrates measurement protocols for a multimodal sensor device in accordance with some embodiments of the present disclosure;

[0021] FIG. 15 illustrates exemplar measurements obtained from a multimodal sensor device in accordance with some embodiments of the present disclosure;

[0022] FIG. 16 illustrates torque scores calculated based on the exemplar measurements obtained from a multimodal sensor device in accordance with some embodiments of the present disclosure;

[0023] FIG. 17 illustrates a comparison of torque scores obtained from a multimodal sensor device in accordance with some embodiments of the present disclosure;

[0024] FIG. 18 illustrates a validation process for a multimodal sensor device in accordance with some embodiments of the present disclosure;

[0025] FIG. 19 illustrates an EMG sensor in communication with the multimodal sensor device in accordance with some embodiments of the present disclosure; and

[0026] FIG. 20 illustrates a computer architecture used in connection with the multimodal sensor device in accordance with some embodiments of the present disclosure.DETAILED DESCRIPTION

[0027] Hypertonia is a condition where a subject may have increased muscle tone. The increased muscle tone may present as stiff muscles in the arms, legs, neck, and the like, that may impair a subjects’ mobility. Hypertonia is commonly seen in conditions such as cerebral palsy, neurodegenerative diseases, stroke, traumatic brain injury, spinal injuries, and the like. Rigidity evaluation, a part of the Unified Parkinson’s Disease Rating Scale, and spasticity evaluation using the Modified Ashworth Scale are both dependent on the resistance felt by the evaluator or treatment provider (e.g., physical therapist, physician, nursing staff, home health aide). These tests rely heavily on the evaluator’s subjective judgment, experience, and perception. Such subjective assessments introduce variability and reduce consistency in outcomes. Disclosed is a multimodal sensing device capable of recording quantitative data such as the evaluator’s movements and the forces applied, can allow for objective, evidence-based assessments, which can provide consistent diagnoses and better-informed treatment plans for patients with motor disorders.

[0028] Implementations of the current disclosure can include a multimodal sensing device that records motions signals corresponding to the evaluator’s movements, the forces applied by an evaluator, and in some implementations, forces experienced by a subject’s muscles. The multimodal sensing device may transmit the recorded motion signals to a processor communicatively coupled to the multimodal sensing device. The processor may further process the recorded data to provide an objective rating of the neuromuscular disorder such as hypertonia. Advantageously, the data recorded by the multimodal sensing device can be processed to allow for objective assessments of disease progression (or regression) in a subject, and / or treatment efficacy using a generated torque score. Additionally, the data recorded by themultimodal sensing device can be used to calibrate torques scores so that they can be evaluated across different treatment providers. Improvements to multimodal sensing devices are discussed.

[0029] FIGS. 1 A and IB illustrate a multimodal sensing device. As shown in FIGS. 1A and IB, in some implementations the multimodal sensing device can be shaped to form a glove that is wearable by a treatment provider or evaluator. FIG. 1 A provides a first view that illustrates the bottom or palm side 110 of the multimodal sensing device 100. FIG. IB provides a second view that illustrates the top, back or dorsal side 120 of the multi-modal sensing device 100. The multimodal sensing device 100 can include at least one communication pathway having a connector 125 that is configured for connection and / or communication with a processing device. In some implementations the processing device can be separate from the multimodal sensing device 100. In some implementations the processing device can be integrated with the multimodal sensing device 100. The multimodal sensing device 100 may include a sensor array. The sensor array may include a plurality of capacitive pressure sensors (e.g. 130, 131, 132, 133, 134, and 135). In some implementations, the capacitive pressure sensors (e.g., 130, 131, 132,133, 134, and 135) can be configured to measure force. Additionally, or alternatively, in some implementations the multi-modal sensing device 100 may include one or more inertial measurement units 140. The inertial measurement units can include inertial motion sensors.

[0030] In some implementations, the multimodal sensing device 100 may may generate motion data based on information sensed by the capacitive pressure sensors (e.g., 130, 131, 132, 133,134, and 135) and inertial measurement units 140. The motion data can be transmitted to one or more processors in communication with the multimodal sensing device 100.

[0031] In some implementations the processing device(s) may be configured to estimate a result (e.g., force, power) based, at least in part, on capacitive pressure sensor data and motion data. For example, for a pressure (p) and a motion (m), the processing device may be configured to compute the following function (f):The function may be related to an effort applied by a medical practitioner employing the hypertonicity measuring device to move a limb of a patient. The processing device may be configured to estimate a power based, at least in part, on the capacitive pressure sensor data and the motion data.

[0032] In some implementations, a force sensor may be communicatively coupled to the at least one processor or to the multimodal sensing device 100. The force sensor may be configured to communicate weight measurements to a microcontroller and / or processor. A motion sensor may be configured to communicate motion measurements to the micro-controller. The microcontroller may be configured to estimate a force applied to the multimodal sensing device 100. The estimate may be based, at least in part, on the weight measurements communicated from a friction sensor. The microcontroller may be configured to estimate an angular velocity applied to a lever of a patient simulation device. The estimate may be based, at least in part, on motion measurements communicated from the motion sensor. The microcontroller may be configured to estimate a power. The power may be based, at least in part, on weight measurements communicated from the friction sensor. The power may be based, at least in part, on motion measurements communicated from the motion sensor. The power (P) may be represented as: P=F*v, where F=m*a, and v is velocity (e.g. an angular velocity) from motion measurements, and where m is mass from weight measurements, and a is acceleration (e.g. standard of gravity).

[0033] The multimodal sensing device can be in the form of a glove that is portable and capable of being worn by an evaluator. The glove may record the applied torque and motion trajectories of the glove as the evaluator performs assessment maneuvers on a subject. In some implementations, the multimodal sensing device can be worn by the evaluator and not the patient. Because the multimodal sensing device is worn by the evaluator and not the patient it may be able to be applied across a greater variety of applications as it is not limited by muscle types and poor fit to different patient sizes.

[0034] Examples of a multimodal sensing device are provided herein. Additional examples are discussed in US Patent No. 11,123,013 entitled “HYPERTONICITY MEASURING DEVICE AND METHOD,” which is hereby incorporated by reference, in its entirety.

[0035] According to some of the various embodiments, the multimodal sensor device or hypertonicity measuring device may comprise at least one wearable item. The at least one wearable item may comprise an article of clothing, a synthetic material, a leather material, combinations thereof, and / or the like. Examples of an article of clothing include a glove, a fingersleeve, a thumb sleeve, a wrist sleeve, a wrap, combinations thereof, and / or the like. Examples of a synthetic material include a fabric, cloth, mesh material, combinations thereof, and / or the like.

[0036] According to some of the various embodiments, the multimodal sensor device may include at least one communication pathway. The at least one communication pathway may be configured to communicate with a processing device. The at least one communication pathway may comprise at least one wireless and / or wired connection. The processing device may comprise a frequency division multiplexing circuit, a multiplexor, an analog to digital converter, an output device, a remote device, combinations thereof, and / or the like. The frequency division multiplexing circuit may employ various modulation schemes. Examples of modulation schemes include Amplitude Modulations (AM), Frequency Modulations (FM), and Phase Modulations (PM). An output device may comprise a display, a plurality of LEDs, a speaker, combinations thereof, and / or the like.

[0037] According to some of the various embodiments, the multimodal sensor device may comprise a sensor array. The sensor array may be disposed to at least one wearable item. The sensor array may comprise a plurality of capacitive pressure sensors. The plurality of capacitive pressure sensors may comprise at least one structured dielectric. The sensor array may be configured to communicate capacitive pressure sensor data to a processing device employing at least one communication pathway. The capacitive pressure sensor data may comprise force measurements. The force measurements may comprise constant force measurements. The force measurements may be communicated from individual capacitive pressure sensors, groups of capacitive pressure sensors, the plurality of capacitive pressure sensors, combinations thereof, and / or the like.

[0038] According to some of the various embodiments, a hypertonicity measuring device may comprise an inertial measurement unit. The inertial measurement unit may be disposed to at least one wearable item. The inertia measurement unit may comprise an accelerometer, a gyroscope, a magnetometer, combinations thereof, and / or the like. The inertial measurement unit may be configured to communicate motion data to a processing device employing at least one communication pathway. The motion data may comprise magnetic field measurements, linear acceleration measurements, angular acceleration measurements, linear velocity measurements,angular velocity measurements, combinations thereof, and / or the like. The motion data may comprise measurements of the inertial measurement unit at rest, under an acceleration, moving at a specific velocity, combinations thereof, and / or the like.

[0039] FIG. 2 provides a system diagram for a system that monitors and assesses treatment and / or severity of hypertonia in a patient. Illustrated is a multimodal sensing device 201 that is communicatively coupled to one or more processing devices 203 which are coupled to a database 205. In some implementations, the multimodal sensing device can include at least one wearable item, at least one communication pathway configured to communicate with the one or more processors, a sensor array disposed in the at least one wearable item, and an inertial measurement unit that includes at least one sensor. Sensors of the inertial measurement unit can include accelerometers, magnetometers, gyroscopes and the like. Sensors in the wearable item can also be configured to measure a force.

[0040] FIG. 3 provides a flow diagram of a method for a system that monitors and assesses treatment and / or severity of hypertonia in a patient. In some implementations, one or more processors can be configured to receive at least one signal including motion data of the multimodal sensing device 301. The motion data can be received from the one or more sensors of the multimodal sensing device such as the inertial motion units, capacitive pressure sensors, force sensors, and the like. In some implementations the signal may correspond to a standardized flexion and extension maneuver that is exerted on the subject by an evaluator that is wearing the multimodal sensing device. A torque score can be determined based on the received signal including the motion data 303. In a next step, the effectiveness of a treatment or an estimated progression of disease can be determined based on a comparison of the determined torque score and at least one historical torque score 305.

[0041] In some implementations the multimodal sensing device can form a sensing glove that is configured to generate sensor data such as motion data. The sensing glove can generate motion data that is transmitted to a processor. In some implementations, the motion data can include force data indicative of the force expended by the evaluator on the subject, which may also indicate the muscle tone of the subject to whom a treatment is being applied to or who is being evaluated. For example, one or more force sensors in the sensing glove can be configured tomeasure the forces exerted by or experienced by the muscles of the subject to whom the maneuver is being applied.

[0042] In some implementations, the motion data can include the maneuver trajectory as recorded by the multimodal sensing device when a standardized flexion and extension maneuver is exerted on the subject by an evaluator. For example, in some implementations, inertial motion units may include sensors such as accelerometers, gyroscopes and magnetometers that can be used to determine a two-dimensional or three-dimensional trajectory taken by the sensing glove when an evaluator or treatment provider was performing a maneuver on a subject. In some implementations, the motion data generated by the multimodal sensing device can include measurements of at least one of a magnetic field, a linear acceleration, an angular acceleration, a linear velocity, and / or angular velocity experienced by the multimodal sensing device, the limb of the subject in contact with the multimodal sensing device, and the like.

[0043] In some implementations, the sensor data generated by the multimodal sensing device can be for a time period corresponding to one or more standardized flexion and extension maneuvers that are performed by an evaluator or treatment provider on the subject by the multimodal sensing device. For example, in some implementations the multimodal sensing device forms a wearable glove with a force sensor and inertial motion unit. An evaluator or treatment provider may wear the multimodal sensing device as a glove and apply maneuvers such as those corresponding to the maneuvers used in the MAS assessment to the subject. The sensing glove can be configured to record signals corresponding to the force, trajectory, and the like, while one or more maneuvers are performed by the evaluator and / or treatment provider.

[0044] FIG. 4 illustrates examples of the signals received at the processor from the multimodal sensing device. As illustrated, signals received at the processor from the multimodal sensing device can include the angular velocity (degrees / second) 401 experienced by the glove during the maneuver as well as the torque (Nm) 403. In some implementations, the processor may receive data or an input indicative of the distance between the arm and the sensing glove so as to be able to determine the torque based on the information from the sensor. Additional data such as a grip length and angle of treatment may be used to determine the torque.

[0045] The angular velocity signal can be processed to identify cycles indicating the start and end times of a maneuver. In some implementations, each “cycle” in the signal trace illustrated in FIG. 4, can correspond to one maneuver. The amplitude of the torque may be indicative of how hard the evaluators have to push or pull against the joint to move during the maneuver, which may also correspond to the severity of the hypertonia. For patients with higher hypertonia, the muscle resists the passive stretch more and the evaluator or treatment provider may have to exert more force (and torque) in order to overcome the muscle resistance.

[0046] In some implementations, the inertial motion unit on the multimodal sensing device may record the acceleration and the moving trajectory of the glove. The recorded signals can be used to visualize the velocity information for the movement. Zero points in the velocity signal may be indicative of switching points between the extension and flexion components of the maneuver as extension and flexion corresponds to a directional change.

[0047] Determining a torque score based on the received one or more signals at the one or more processors may include determining a subset of the received at least one signal corresponding to the standardized flexion maneuver and determining a subset of the received at least one signal corresponding to the standardized extension maneuver. Various methods can be used to identify the zero points in the velocity signal in order to identify portions of the signal corresponding to extension and portions of the signal corresponding to flexion. Accordingly, the torque values for an extension portion of the signal may correspond to the extensor muscles, while the torque values for a flexion portion of the signal may correspond to the flexor muscles. Methods for identifying zero points in the velocity can also help identify the number of cycles in a particular recording, which may be utilized for determining an error margin. Example methods for identifying zero points include, but are not limited to sign change analysis, interpolation, and finding roots of a function.

[0048] Acceleration information determined from the velocity information and / or the motion data can also be used in a calibration process. As will be discussed below, a calibration process may require as input, a force, weight, and acceleration.

[0049] In some implementations, a torque signal can be determined across the time series corresponding to the angular velocity and force measured for the signal multiplied by a distance between the torque.

[0050] In some implementations, the angular velocity along the axis of maximum motion was determined through applying principal component analysis on the data taken from a 3-axis gyroscope in the inertial motion unit. For the torque measurement, signals from all the force sensor pixels can be summed together at each sampling time point. The relation r -- / / ’ si nF was used to calculate the torque T from the summed force F. The grip length was measured between the patient’s joint and the location where the clinician gripped the patient’s limb. The angle 6 was presumed to be 90° [sin(90°) = 1], since the clinician applied force in the direction perpendicular to the grip length.

[0051] In some implementations, the torque signal (e.g., see FIG. 4) can be used to generate a heat map or other visual mapping of the experienced torque for each of the flexion and extensor muscles at each of the force sensor positions.

[0052] As shown in FIG. 5, in some implementations, the torque can be visually represented in a histogram 500 with torque on the x-axis and count on the y-axis and illustrate the distribution and center of the data. The median torque 503 can be denoted as a solid black line and provide an indication of the torque data’s overall amplitude, and the standard error mean 505. In some embodiments the standard error mean can be calculated as the standard deviation 507 divided by the square root of the maneuver cycle and thus account for the variation between maneuvers, as the margin of error. For example, the standard error mean = standard deviation / sqrt (cycle number). In this manner, the systems and methods and described herein can be configured to manage errors introduced due to human variability.

[0053] In some implementations, when the patient’s data is collected over different visits, the torque histogram, along with the indicative lines, can help determine the muscle tone change. Changes in muscle tone determined by the systems and methods described herein can help assess the effectiveness of treatments and / or the progression of disease.

[0054] As shown in FIG. 6, in some implementations, a torque score can be determined based on the median or average torque experienced at a visit. In some implementations, a determined torque score can be compared to a prior torque score. For example, one or more historical torque scores can be stored in a database communicatively coupled to the at least one processor.

[0055] For example, one or more torque scores for a patient can be compared from a current visit (e.g., visit x) to a prior visit (e.g., visit 0). A delta torque value 601 corresponding to the difference between the median torque at a current visit 603 and the median torque at a prior visit 605 can be used to determine the effectiveness of treatments being applied to a subject and / or the progression of disease in a subject. Shown in FIG. 6 is an example of data collected from one patient at a first visit 0 and a later visit x. A higher torque value may correspond with an indication that an evaluator, therapist or medical provider needed more force to move the patient’s limb which in turn may be indicative of severe muscle tone. On the illustrated histograms, a more severe muscle tone and higher torque value may appear to the right. If the muscle is relaxed or has little resistance, it should take less force to move, which will be indicated by a torque score that is closer to 0. By observing the shift of the torque between a prior visit and a current visit, the patient’s progress can be visualized. The change in torque score can be quantitatively analyzed and by a delta torque measure. If the torque shifts to the left from baseline to recent visit, it indicates a lower tone, which is shown in this example provided in FIG. 6. Accordingly, the equation of delta torque, it should result in a negative delta torque. Delta torque can be expressed as follows: Atorque of Visit(x) = torque median at Visit(x) - torque median at Visit (0).

[0056] In some implementations, the long term monitoring of patient progress can also be based on a measure of the error of margin of the delta torque. For example, the error of margin of the delta torque can be calculated based on the standard error formula. Because the delta toque is the difference between two groups, the formula involves the cycle number (or count) from both visits and the SD from both visits. Accordingly, the statistical significance of the standard error of the means can be determined.

[0057] In some implementations the sensing glove can be configured for us by a single evaluator or therapy provider for a single patient. In some implementations, the sensing glovecan be configured to provide consistent objective evaluations across various evaluators and therapy providers. For example, in some implementations the multimodal sensor device can be configured to provide inter-rater reliability such that different evaluators are able to generate consistent results (e.g., torque scores) when assessing the same patient.

[0058] Inter-rater variability can result from variations in human grip. For example, grip force can be dependent on external visual cues, differences in weights being gripped, differences in object surfaces, difference in speed of maneuvers, and acceleration of movement. Accordingly, these differences can lead to differences in the measured torque for a maneuver.

[0059] In some implementations, a calibration model can be built for each user of the sensing glove. In some implementations, a multivariant model can use calibration data to remove individual variances due to the evaluation or therapy provider.

[0060] FIG. 7 provides a flow-chart for a method for calibrating a multimodal device. As shown in FIG. 7, a processor may receive motion data for a calibration activity 701, determine a calibration factor based on a torque score for the received motion data for the calibration activity 703, and apply the determined calibration factor to one or more future signals received from the multimodal device 705.

[0061] A calibration model for a user can be generated by receiving motion data for a calibration activity and determining one or more calibration factors based on a torque score for the received motion data for the calibration activity. For example, in some implementations the calibration activity may include collecting acceleration, force, and weight data for a user performing one or more maneuvers with a dumbbell or other weight ranging from 0.5 lbs to 6 lbs. As the user is performing the calibration activity, acceleration and force sensor data can be logged. The relationship between the calibration activity and the calibration factors may be expressed as weight = intercept + kl * accel + k2* force. The determined calibration factors can then be used to calibrate output data from a sensing glove for a particular evaluator or treatment provider. In some implementations the determined calibration factors can be used in a multivariant regression model that is applied to data obtained by the sensing glove.

[0062] FIG. 8 provides an example of the calibration process where a regression model may generate one or more weights for users (e.g., kl, k2) 801. The calibration activity may produce acceleration, force, and weight data 803, that can be used to train the multivariant regression model 805 which determines one or more calibration factors or parameters 807 that can be applied to future measurements 809.

[0063] In some implementations, the multivariant regression model 805 may be composed of a multiple variable linear regression model that utilizes a best fit process to minimize distances between the datapoints input that correspond to the force, weight, acceleration in order to determine parameters or weights for the model (e.g., kl, k2 ).

[0064] Calibrated data corresponding to the same subject and various evaluators can be used to determine the effectiveness of the calibration process using kernel density estimation, which evaluates how similar the user results are to one another.

[0065] As illustrated in FIG. 9, one or more determined calibration factors can be applied to signals received from the multimodal device to generate post-calibrated data. As shown in FIG.9, after calibration, data corresponding to three different users may exhibit similar torque values, thus providing inter-rater consistency and addressing the problems of inter-rater variability. FIG.9 illustrates pre- 901 and post- 903 calibration values for three users. As shown, the calibration values lead to overlapping determinations of weight and acceleration across users, thus indicating that the calibration leads to reductions or eliminations in inter-rater variability.

[0066] As shown in FIGS. 10- 12 in some implementations, the multimodal sensing device can be in communication with a graphical user interface and display providing visual representations.

[0067] As shown in FIG. 10, in some implementations the user interface may be configured to receive input such as data files including torque, force, acceleration or weight values corresponding to previous evaluations or treatments. In some implementations the user may also be able to manually input a trajectory or a count of the number of maneuvers that were performed during a time period (e.g., the number of cycles).

[0068] As shown in FIG. 11, in some implementations, the user interface may display torque calculations in a histogram format for two separate time periods in one graphical interface, thus allowing a user to compare the median torque value over time. In some implementations, the torque values for the flexion muscles are viewed with that for their corresponding extensor muscles.

[0069] As shown in FIG. 12, in some implementations, the torque data and angular velocity data can be visually displayed to a user. Additionally, the recalibrated torque value and corresponding calibration factors can be shown.

[0070] Although the systems and methods described herein are with respect to hypertonicity, it is envisioned that they may be applicable to other applications such as Parkinson’s disease evaluation, as well as other motor diseases. As described above, determining an effectiveness of a treatment or an estimated progression of disease can be based on a comparison of the determined torques score and at least one historical torque score. The treatment may involve a pharmaceutical drug and a dosage, and / or physical therapy.EXAMPLESExample #1 : Longitudinal monitoring of hypertonia through a multimodal sensing glove

[0071] In some implementations, the multimodal sensing glove described herein can be used for the longitudinal monitoring of hypertonia.

[0072] In one example, a double-blind study was used to enable sensitive monitoring of medication effects across 19 participants. The biomechanical measurements from the multimodal sensing glove effectively distinguished patient cohorts receiving a baclofen treatment or a placebo with 95% confidence. Consistent monitoring over a two-month period was demonstrated, closely tracking variations in individual responses to treatment. The biomechanical changes were correlated to neural activities as recorded by electromyography, verifying the medication effects. The multi-modal sensing glove (also referred to as sensing glove, for short) was shown to be a reliable tool for point-of-care settings to facilitate precise evaluation of hypertonia, essential for tailoring individual treatment choices and timelymanagement of chronic symptoms. The multimodal sensing glove was used to objectively monitor the effects of a muscle relaxant, such as an antispasmodic agent an example of which is Baclofen medication, in patients with a neuromuscular disorder, such as hypertonia.

[0073] In this example, the measurements are analyzed to determine their levels of resolution and consistency in monitoring outcomes over time, enabling the differentiation of patients undergoing treatment from those receiving placebos in our double-blind study. In addition to showing the feasibility of longitudinal monitoring, this example validated the sensing glove by comparing its biomechanical measurements with simultaneous EMG signals.

[0074] The complementary results demonstrate the potential of the sensing glove system to serve as a clinically practical standard for assessing hypertonia and enables timely interventions in managing patients’ chronic symptoms and improve their quality of life.Materials and methods

[0075] Patient Recruitment and Dosage information: The patient age ranged from 5 to 38 years old, including 9 males and 10 females. The ethnicity distribution was Hispanic 42.1%, Asian 26.3%, White 26.3%, and African American 5.3%. The patients’ ages did not significantly influence the severity of hypertonia. Rather the size of the affected limbs, independent of age, had a greater influence in the force required to perform the assessment. Among the 19 patients, 16 of them were diagnosed with cerebral palsy, 1 with traumatic brain injury, 1 with Pelizaeus Mertz-bacher disease, and 1 with a pineal germ cell tumor. The patient cohort covered a wide range of age, ethnicities and conditions to prevent bias. For the treatment plan, the starting dosage for all patients was 0.2 mL with 100 pg baclofen nasal solution. If the patient chose to increase their dosage on a daily basis, the baclofen dosage followed the step table in Table 1 and the maximum dose escalation was one step every day until step 16. If they chose not to increase the dosage, they would stay at whatever dosage they deemed fit but were unable to escalate faster than one step every other day. For day 43-46, the washout period, the patient’s dosage decreased to 80%, 60%, 40% and 20% of the maximum dosage. Then for day 47-61, the patient stopped taking baclofen. For the placebo plan, patients took matching saline solution on the same schedule as the treatment plan.Table 1: Dosing Schedule

[0076] For the pressure measurements, each limb was evaluated through at least 10 maneuvers, then the median and the standard error of the mean were computed for comparison. For theEMG signal, the baseline noise level was defined as the average of the last 1 s’s recording, during which the muscles were at rest.

[0077] Clinical treatment exception'. There were some data that were not collected due to missing appointments: one treatment patient’s visit 1; another treatment patient’s visit3; and one control patient’s visit 4.

[0078] Calibration method: The glove calibration process was performed as such: a 2"- diameter syringe was filled with water to adjust the weight corresponding to multiple resistance levels, ranging from 1 to 5 lbs which were the typical for muscles with hypertonia. Then, wearing the glove, the evaluator held the syringe and performed the flexion and extension maneuvers as they would on a patient’s limb. The recorded pressure readings and the corresponding weights were fitted with linear regression. The calibration curve is shown in FIG. 13 for weights 1-5 lbs. The life time of the glove is longer than ten months, proven by the sensor maps in FIG. 13. FIG. 13 provides a heatmap for the determined torque score across one or more geographical areas of the multimodal sensing device.

[0079] S.D. and S.E.M. calculation'. The standard deviation S.D. was calculated with the MATLAB function ‘std’. The S.E.M was calculated with the cycle numbers recorded in each set, using equation S.E.M. = S.D. / A0 5, where the sample number N was the number of maneuver cycles in the recording.

[0080] Atorque error bar calculation: The error bars of Atorque were determined by the standard error formula:where the subscripts denote that the data were collected at Visit(x). For example, Av and No are the number of maneuver cycles in Visit(x) and Visit(O), respectively. The variable S.D.Xis the standard deviation of torque data in Visit(x).

[0081] T-Value calculation'. The t-values in FIG. 17 were calculated with MS Excel using the built-in function found under ‘Data’-’Data Analysis’-‘ / -Test: two-Sample assuming Unequal Variances’.

[0082] Test power calculation. The power in FIG. 17 was calculated post hoc based on the obtained data and sample size with software GPower version 3.1.9.7, under ‘Test family-t tests. Statistical test: Difference between two independent means (two groups)’.

[0083] Surface EMG'. The surface EMG signals were collected using the Delsys Trigno System with Avanti Sensors shown in FIG. 19. The sensors were adhered to participants with double-sided tape. The EMG signal was sampled at 2000Hz and no fdter was applied during the recording process.

[0084] EMG coactivation time calculation'. The EMG signals were fdtered by a 5th order high-pass Butterworth fdter at a cutoff frequency of 10 Hz. Then the signals were normalized with the ‘normalize’ function in MATLAB. To determine the muscle activation state from the EMG signal, the signal’s envelope was compared to the baseline noise level. The baseline noise level was defined as the average of the last 1 s’s recording, during which the muscles were at rest. For the signal envelope, a sliding window of 50 datapoints was used to calculate the rootmean-square envelopes. When the signal envelope was higher than the defined noise level continuously for 30 data points (equivalent to 0.015 s), the period was considered ‘on’, in which the muscle was activated. Otherwise, the period was considered as ‘off’ with no significant muscle activation. The CoD was computed by determining the overlapping duration during which a pair of muscles were both in the ‘on’ activated state, divided by the entire flexion / extension phase duration.1. Results and discussion / .1. Study design

[0085] The abnormal muscle tone of hypertonia, as illustrated in Fig. 14 Section A, is evaluated by using the sensing glove to measure the torque exerted to move the patient’s affected muscles. A higher torque indicates increased muscle tone (namely, resistance to movement) andmore severe hypertonia. The multi-modal sensing glove shown FIG. 14 Section B has 349 forcesensitive resistors (Tekscan Inc.) on the palm side and an inertial measurement unit (IMU from MotionNode Inc.) on the back side. The synchronized sensors simultaneously acquire the muscle tone and the maneuver trajectories, capturing the dynamic characteristics of hypertonic muscles, i.e., velocity-dependent muscle resistance.

[0086] In the evaluation procedure as shown in FIG. 14 Section C a clinician wore the sensor glove to perform standardized flexion and extension maneuvers on patients, as normally done in perception-based clinical evaluations. The patient remained passive in a supine position and voluntary motion was not required from the patient, because the patient might be physically and / or cognitively unable to follow instructions. Throughout the maneuvers, the clinician would stabilize the patient’s joint with one hand, and then the gloved hand would move the patient’s limb through eight to ten cycles, striving to reach a steady level in the cycling measurements. In this study, each patient was evaluated on all four limbs, including the elbows and knees on the left and right sides.

[0087] Through the double-blind experiment, it was ensured that the observed changes in patients’ muscle tone were determined without influence from perception biases. The results would establish whether the glove measurements can function as objective, sensitive metrics for assessing treatment effects. This study monitored 19 patients, who were randomly assigned to two groups. One group received the antispasmodic baclofen medication, while the other received a saline placebo delivered in the same manner as the baclofen. The patients were measured five times following the schedule in FIG. 14 Section D. At the initial Visit 0, the clinician recorded each patient’s baseline condition and provided them with doses of either baclofen or placebo for daily consumption over a two-week period. Then, the patients returned biweekly for evaluation and to receive continual dosages. To mitigate the potential for baclofen to induce sudden withdrawal symptoms, the dosage dispensed at Visit 3 was designed to gradually reduce the medication to zero in a washout period. By Visit 4, all the patients were no longer taking any dosage and underwent a final evaluation. It was only after the completion of data analysis that the group assignments were revealed.1.2. Maneuver analysis

[0088] FIG. 15 displays a representative recording of the evaluation maneuvers performed on a patient’s limb. Sensor signals are processed to obtain the angular velocity and torque data. In brief, the angular velocity along the axis of maximum motion was determined through applying principal component analysis on the data taken from the 3-axis gyroscope in the IMU. For the torque measurement, signals from all the force sensor pixels (shown in the inset of FIG. 15) were summed together at each sampling time point. The relation T = / x F = F sin0 was used to calculate the torque r from the summed force F. The grip length / was measured between the patient’s joint and the location where the clinician gripped the patient’s limb, as denoted by the red line in FIG. 14. The angle Q was presumed to be 90° [sin(90°) = 1], since the clinician applied force in the direction perpendicular to the grip length.

[0089] The 10-s recording of angular velocity and torque in FIG. 15 shows 8 maneuver cycles. For these synchronized signals, each cycle can be further segmented to uncover the variations in muscle resistance under flexion and extension. Because the maneuver velocity must drop to zero in order to switch movement directions between flexion and extension, the zero-crossing points within the angular velocity data indicated the endpoints of each flexion / extension segment. Using the corresponding time points where v = 07s, the torque measurements were demarcated into flexion (purple) and extension (orange) segments. This segmented torque data highlighted the asymmetric tone between the extensor and flexor muscles, as clearly evident in the corresponding histograms in FIG. 16. The torque histogram for extension segments presented a higher median value compared to that under flexion. When there was more stiffness in a patient’s muscle, the evaluator had to apply an increased torque to overcome the patient’s muscle resistance. The higher torque median signified worse hypertonia in the muscles engaged in extension. The unequal muscle tone suggests different degrees of damage to the two brain hemispheres and is quite common; specifically among the evaluation trials, with over 10% of the results exhibiting asymmetric muscle tone (see Table 2). However, considering that the study used systemic baclofen, rather than a localized treatment such as botox injections, total torque datasets without segmentation can be studied.

[0090] Table 2 illustrating percentage of patients with asymmetric muscle tone between flexion and extension at each joint. / .3. Comparison of patient cohorts under baclofen treatment and placebo

[0091] As described earlier, the patients were evaluated five times over a two month period. After each visit, the torque histograms were extracted from the recordings, typically encompassing eight to ten maneuver cycles. For the histograms, various statistical measures were determined, such as the median, standard deviation (S.D.), and standard error of the mean (S.E.M.). The dashed black lines in the histograms denote ±1 S.D. of the dataset. In FIG. 16 thedashed lines mark the S.E.M. To quantify the changes in muscle tone between visits, the median values in the histograms were used to calculate a metric defined as Atorque:

[0092] A torque of Visit(x) = torque median at Visit(x) - torque median at Visit(O) (Eqn 2)

[0093] The Atorque represents the change in muscle tone with respect to the initial baseline condition and accounts for cumulative effects from the medication, if any, over the duration of the study. A negative Atorque means a decreased resistance to movement, namely the patient’s muscle tone has reduced to a lower severity compared to their baseline, and conversely for a positive value of Atorque.

[0094] FIG. 17 Panels A and B display opposite shifts in the torque histograms from Visit 0 to Visit 4 for two different patients. Upon revealing the group assignments, it was confirmed that the data of for Panel A of FIG. 17 were taken on a patient in the treatment group receiving baclofen medication, whereas those of Panel B of FIG. 17 were from a patient in the placebo group. These results aligned with the notion that baclofen would have treatment effects of reducing muscle tone, leading to a negative Atorque for the patient receiving the medication. Whereas for the patient on the placebo, the muscle tone did not improve and even slightly worsened over time, as evidenced by the positive Atorque indicating the lack of treatment effect. Meanwhile, the clinical Modified Ashworth Scale (MAS) scores were rated to be the same, at 1.5 out of the scale of 4, for all the visits and showed no clear trend to distinguish the medication effect. The insensitive MAS scores underscored the difficulty in using a perception-based scale to monitor treatment outcomes.

[0095] For our double-blind study which included 11 patients in the treatment group and 8 patients in the placebo group, the Atorque metrics for every patient are summarized in FIG. 17 Panel C where each bar in the same row represents the same patient. While the variance values ranged from 1 to 5 Nm, on average ± 1 Nm was the typical torque resolution of the sensing glove for detecting changes in muscle tone.

[0096] FIG. 17 Panel C reveals key insights when comparing the outcomes between cohorts and within individuals. Regarding the difference between the treatment and placebo cohorts, thepatients under treatment all showed Atorque progressing towards negative values by Visit 3. Subsequently following the washout period, the treatment group still retained some improvement in muscle tone at the final Visit 4. In contrast, the Atorque of patients receiving placebos fluctuated around zero, with some displayed a tendency towards positive values that indicated worsening muscle tone over time, while others bounced between small ± values with no clear pattern.

[0097] To evaluate the difference between the Atorque distributions of the treatment and placebo groups, unpaired t-tests were performed and established that the difference was statistically significant at the confidence level of 95%. In this work the critical t-value was 1.9 as determined from the sample degrees of freedom of 17. Since all the calculated t-values, as shown in FIG. 17 Panel C, exceeded this critical threshold of 1.9, the t-test results demonstrated that the Atorque distribution between the treatment and placebo groups were statistically different across all visits. In addition to t-values, another important statistical parameter was the test power, denoted as (1-13), which was calculated post-hoc with the tool G*Power version 3.1.9.7. The (1-13) value represented the probability of “true positive”, and a power value higher than 0.8 is conventionally considered to be adequate. For the data in FIG. 17 Panel C, all the power values exceeded 0.92, providing reassurance that the study sample size was sufficient in our analyses.

[0098] Thus, two important findings were drawn from FIG. 17 Panel C. First, between the treatment and placebo cohorts, the glove data distinguished the medication efficacy within the studied populations. Second, on an individual level, the glove provided consistent monitoring over time with a resolution down to ±1 Nm, facilitating the sensitive detection of variations in individual responses to treatment. Hence, the measurement results can be used to tailor and optimize personalized medication plans for each patient, beneficial for mitigating side effects and promoting timely interventions, which would be especially valuable for pediatric patients experiencing rapid growth and development.1.4. Comparison of measurements from electromyography and sensing glove

[0099] While the above measurements focused on assessing the biomechanical changes in muscle tone, the abnormality of hypertonic muscles is also correlated with hyperactivities of motor neurons in which the imbalance between excitatory and inhibitory control hindered the muscle from executing fluid coordinated movement and resulting in exaggerated resistance. The electrical signals from motor neurons have been captured by surface EMG for evaluation of motor disorders based on the muscle co-activation characteristics. Therefore, we carried out a double-blind test that incorporated surface EMG as a complementary technique to validate the biomechanical readings of the sensing glove. This test was conducted on two patients to compare treatment and placebo outcomes, following the biweekly evaluation and dosage schedule up to Visit 3, as depicted in FIG. 14, and with simultaneous EMG and sensing glove recordings. The experimental setup is shown in FIG. 18 where EMG sensors were placed on a pair of agonist and antagonist muscles, specifically the biceps and triceps of the patient’s arm. The EMG and sensing glove systems were synchronized to simultaneously capture data as the patient’s arm was moved to flexion and extension positions.

[0100] To interpret the EMG data, prior studies have introduced a metric of co-activation duration (CoD), where CoD was computed by determining the overlapping duration during which a pair of muscles were both active, divided by the entire flexion / extension phase duration. The muscles being examined should be an agonist-antagonist pair, such as biceps-triceps on the upper arm or quadriceps-hamstring on the upper leg. In a healthy person, the EMG of an agonist-antagonist muscle pair showed alternating activation of each muscle. In contrast, the EMG of a patient with hypertonia displayed frequent co-activation as illustrated in FIG. 18, such that the bursts of neuron firings were concurrent in both muscles and would interfere with each other.

[0101] The muscles were considered active when the EMG amplitude exceeded the noise level of the baseline for at least 15 ms. The EMG magnitudes were not correlated to the applied torque and were notoriously prone to issues with environmental noises and reproducibility due to variations in electrode locations and skin conductivity. However, the EMG CoD showed better repeatability than magnitudes and therefore were selected for comparison with the torquemeasurements from the sensing glove. At times, the phase duration was not clear-cut based solely on EMG data; but the simultaneous velocity measurements from the glove could be used to better define segmentation of flexion and extension phases as done in FIG. 18. Since the EMG and the glove data were collected simultaneously, the agonist and antagonist muscles should activate alternatively corresponding to the extension and flexion direction maneuver. The angle of the glove and the time at which the direction switched were recorded by the IMU. The phase durations obtained from the IMU data were then employed to reduce the un-certainties in the denominators when calculating the CoD.

[0102] Panels C and D of FIG. 18 compare the data collected on each patient’s right elbow, with one patient under treatment and the other under placebo, respectively. During the initial Visit 0, both patients showed a CoD >50%, manifesting abnormal co-activation. For the patient under baclofen treatment in FIG. 18 both the EMG and glove measurements showed the same trend, with the CoD and median torque dropped as treatment progressed. By Visit 3, the CoD was only 10%, and the median torque was also lower than the initial resistance in Visit 0. These measurements were in agreement and indicated improvement of hypertonia symptoms under treatment. Whereas for the patient under placebo, both the CoD and median torque fluctuated but did not show improvement between Visits 0 and 3. Overall, the trends in the EMG and biomechanical measurements were generally consistent and mutually validated the medication effects.

[0103] The EMG and the sensing glove provided complementary views of mechanical and neural aspects in hypertonia. Nonetheless, we note that the torque data offered a more direct mechanical characterization of muscle resistance. On its own, the easy-to-use sensing glove can serve a stand-alone standard for point-of-care settings to facilitate precise evaluation of treatment outcomes.Conclusion

[0104] This example demonstrates precise longitudinal monitoring of baclofen medication efficacy in treating hypertonia through a portable multi-modal sensing glove. In our doubleblind study, the biomechanical measurements yielded results that effectively distinguishedbetween patient cohorts and closely tracked changes within individual patients. Moreover, this study validated the biomechanical glove system with neural EMG, establishing the glove metrics as a reliable standard for hypertonia evaluations even in noisy point-of-care settings.

[0105] The highly sensitive measurements readily showed treatment out-comes during biweekly evaluations, eliminating the need for long waiting periods. This capability would be critical for identifying potential problems during treatment; for instance, a study of intrathecal baclofen treatment determined that 19.7% of the cases encountered medication delivery problems to the target sites. Objective assessments through the sensing glove can indicate when the efficacy of a treatment was low and promote timely intervention. In turn, this added measure can lead to savings on medication costs, minimize delays in managing chronic symptoms, and ultimately enhance the quality of life for patients.

[0106] In some implementations, aspects of the current subject matter can be configured to be implemented in a system 2000, as shown in FIG. 20. The system 2000 can include a processor 2010, a memory 2020, a storage device 2030, and an input / output device 2040. Each of the components, such as 2010, 2020, 2030 and 2040, can be interconnected using a system bus 2050. The processor 2010 can be configured to process instructions for execution within the system 2000. In some implementations, the processor 2010 can be a single-threaded processor. In alternate implementations, the processor 2010 can be a multi -threaded processor. The processor 2010 can be further configured to process instructions stored in the memory 2020 or on the storage device 2030, including receiving or sending information through the input / output device 2040. The memory 2020 can store information within the system 2000. In some implementations, the memory 2020 can be a computer-readable medium. In some implementations, the computer-readable medium includes a non-transitory computer- readable storage medium in communication with one or more processors and the multimodal sensing device and having instructions stored thereon. The instructions may cause the one or more processors to perform operations. In alternate implementations, the memory 2020 can be a volatile memory unit. In yet some implementations, the memory 2020 can be a non-volatile memory unit. The storage device 2030 can be capable of providing mass storage for the system 2000. In some implementations, the storage device 2030 can be acomputer-readable medium. In alternate implementations, the storage device 2030 can be a floppy disk device, a hard disk device, an optical disk device, a tape device, non-volatile solid- state memory, or any other type of storage device. The input / output device 2040 can be configured to provide input / output operations for the system 2000. In some implementations, the input / output device 2040 can include a keyboard and / or pointing device. In alternate implementations, the input / output device 2040 can include a display unit for displaying graphical user interfaces.

[0107] Aspects of the systems and methods disclosed herein can be embodied in various forms including, for example, a data processor, such as a computer that also includes a database, digital electronic circuitry, firmware, software, or in combinations of them. Moreover, the above-noted features and other aspects and principles of the present disclosed implementations can be implemented in various environments. Such environments and related applications can be specially constructed for performing the various processes and operations according to the disclosed implementations or they can include a general -purpose computer or computing platform selectively activated or reconfigured by code to provide the necessary functionality. The processes disclosed herein are not inherently related to any computer, network, architecture, environment, or other apparatus, and can be implemented by a suitable combination of hardware, software, and / or firmware. For example, various general-purpose machines can be used with programs written in accordance with teachings of the disclosed implementations, or it can be more convenient to construct a specialized apparatus or system to perform the required methods and techniques.

[0108] Aspects of the systems and methods disclosed herein can be implemented as a computer program product, i.e., a computer program tangibly embodied in an information carrier, e.g., in a machine-readable storage device or in a propagated signal, for execution by, or to control the operation of, data processing apparatus, e.g., a programmable processor, a computer, or multiple computers. A computer program can be written in any form of programming language, including compiled or interpreted languages, and it can be deployed in any form, including as a stand-alone program or as a module, component, subroutine, or other unit suitable for use in a computing environment. A computer program can be deployed to be executed on one computer or onmultiple computers at one site or distributed across multiple sites and interconnected by a communication network. These computer programs, which can also be referred to programs, software, software applications, applications, components, or code, include machine instructions for a programmable processor, and can be implemented in a high-level procedural and / or object- oriented programming language, and / or in assembly / machine language. As used herein, the term “machine-readable medium” refers to any computer program product, apparatus and / or device, such as for example magnetic discs, optical disks, memory, and Programmable Logic Devices (PLDs), used to provide machine instructions and / or data to a programmable processor, including a machine-readable medium that receives machine instructions as a machine-readable signal. The term “machine-readable signal” refers to any signal used to provide machine instructions and / or data to a programmable processor. The machine-readable medium can store such machine instructions non-transitorily, such as for example as would a non-transient solid- state memory or a magnetic hard drive or any equivalent storage medium. The machine-readable medium can alternatively or additionally store such machine instructions in a transient manner, such as for example as would a processor cache or other random-access memory associated with one or more physical processor cores.

[0109] As used herein, the term “user” can refer to any entity including a person or a computer.

[0110] Although ordinal numbers such as first, second, and the like can, in some situations, relate to an order; as used in this document ordinal numbers do not necessarily imply an order. For example, ordinal numbers can be merely used to distinguish one item from another. For example, to distinguish a first event from a second event, but need not imply any chronological ordering or a fixed reference system (such that a first event in one paragraph of the description can be different from a first event in another paragraph of the description).[0U1] To provide for interaction with a user, the subject matter described herein can be implemented on a computer having a display device, such as for example a cathode ray tube (CRT) or a liquid crystal display (LCD) monitor for displaying information to the user and a keyboard and a pointing device, such as for example a mouse or a trackball, by which the user can provide input to the computer. Other kinds of devices can be used to provide for interaction with a user as well. For example, feedback provided to the user can be any form of sensoryfeedback, such as for example visual feedback, auditory feedback, or tactile feedback; and input from the user can be received in any form, including, but not limited to, acoustic, speech, or tactile input.

[0112] The implementations set forth in the foregoing description do not represent all implementations consistent with the subject matter described herein. Instead, they are merely some examples consistent with aspects related to the described subject matter. Although a few variations have been described in detail above, other modifications or additions are possible. Further features and / or variations can be provided in addition to those set forth herein. For example, the implementations described above can be directed to various combinations and subcombinations of the disclosed features and / or combinations and sub-combinations of several further features disclosed above. In addition, the logic flows depicted in the accompanying figures and / or described herein do not necessarily require the order shown, or sequential order, to achieve desirable results. Other implementations can be within the scope of the following claims. In the descriptions above and in the claims, phrases such as “at least one of’ or “one or more of’ may occur followed by a conjunctive list of elements or features. The term “and / or” may also occur in a list of two or more elements or features. Unless otherwise implicitly or explicitly contradicted by the context in which it is used, such a phrase is intended to mean any of the listed elements or features individually or any of the recited elements or features in combination with any of the other recited elements or features. For example, the phrases “at least one of A and B;” “one or more of A and B;” and “A and / or B” are each intended to mean “A alone, B alone, or A and B together.” A similar interpretation is also intended for lists including three or more items. For example, the phrases “at least one of A, B, and C;” “one or more of A, B, and C;” and “A, B, and / or C” are each intended to mean “A alone, B alone, C alone, A and B together, A and C together, B and C together, or A and B and C together.” Use of the term “based on,” above and in the claims is intended to mean, “based at least in part on,” such that an unrecited feature or element is also permissible.

[0113] The subject matter described herein can be embodied in systems, apparatus, methods, and / or articles depending on the desired configuration. The implementations set forth in the foregoing description do not represent all implementations consistent with the subject matterdescribed herein. Instead, they are merely some examples consistent with aspects related to the described subject matter. Although a few variations have been described in detail above, other modifications or additions are possible. In particular, further features and / or variations can be provided in addition to those set forth herein. For example, the implementations described above can be directed to various combinations and subcombinations of the disclosed features and / or combinations and subcombinations of several further features disclosed above. In addition, the logic flows depicted in the accompanying figures and / or described herein do not necessarily require the particular order shown, or sequential order, to achieve desirable results. For example, the logic flows may include different and / or additional operations than shown without departing from the scope of the present disclosure. One or more operations of the logic flows may be repeated and / or omitted without departing from the scope of the present disclosure. Other implementations may be within the scope of the following claims.

Claims

CLAIMS1. A system comprising: a multimodal sensing device comprising one or more sensors for sensing motion data from a subject; a non-transitory computer-readable storage medium in communication with one or more processors and the multimodal sensing device, the non-transitory computer-readable storage medium having instructions stored thereon which, when executed by the one or more processors, cause the one or more processors to perform operations comprising: receiving, from the one or more sensors of the multimodal sensing device, at least one signal comprising motion data of the multimodal sensing device, wherein the at least one signal corresponds to a standardized flexion and extension maneuver exerted on the subject by an evaluator wearing the multimodal sensing device; determining a torque score based on the received at least one signal; and determining an effectiveness of a treatment or an estimated progression of disease based on a comparison of the determined torques score and at least one historical torque score.

2. The system of claim 1, wherein the motion data of the multimodal sensing device comprises a force expended by the evaluator on the subject .

3. The system of claim 1, wherein the motion data of the multimodal sensing device comprises a maneuver trajectory of the standardized flexion and extension maneuver exerted on the subject by the evaluator.

4. The system of claim 1, wherein determining the torque score based on the received at least one signal corresponding to the standardized flexion and extension maneuver comprises: determining a subset of the received at least one signal corresponding to the standardized flexion maneuver; determining a subset of the received at least one signal corresponding to the standardized extension maneuver; anddetermining the torque score based on a force corresponding to the standardized flexion maneuver based on the determined subset of the received at least one signal corresponding to the standardized flexion maneuver, a force corresponding to the standardized extension maneuver based on the determined subset of the received at least one signal corresponding to the standardized extension maneuver, a grip length and angle of treatment.

5. The system of claim 4, wherein determining the torque score further comprises: determining a heatmap for the determined torque score across one or more geographical areas of the multimodal sensing device.

6. The system of claim 1, further comprising: a graphical user interface communicatively coupled to the one or more processors.

7. The system of claim 6, wherein the graphical user interface is configured to: receive user input of at least one trajectory corresponding to maneuver or a count of the maneuvers; and display the determined torque score.

8. The system of claim 6, wherein the graphical user interface is configured to display one or more visual representations of the determined torque score and the at least one historical torque score.

9. The system of claim 6, wherein the graphical user interface is configured to display a calibrated torque score and / or torque signal.

10. The system of claim 1, further comprising: receiving motion data for a calibration activity; and determining a calibration factor based on a torque score for the received motion data for the calibration activity.

11. The system of claim 10, wherein applying the determined calibration factor to one or more future signals received from the multimodal device.

12. The system of claim 1, wherein the treatment comprises a pharmaceutical drug and a dosage, and / or physical therapy.

13. The system of claim 1, wherein the at least one historical torque score comprises one or more torque scores associated with the subject stored in a database communicatively coupled to the one or more processors.

14. The system of claim 1, wherein the multimodal sensing device comprises: at least one wearable item; at least one communication pathway configured to communicate with the one or more processors; a sensor array disposed in the at least one wearable item; and an inertial measurement unit comprising the at least one sensor.

15. The system of claim 1, wherein the at least one sensor comprises at least one of an accelerometer, a gyroscope, or a magnetometer.

16. The system of claim 1, wherein the motion data comprises measurements of at least one of at least one magnetic field, linear acceleration, angular acceleration, linear velocity, or angular velocity.

17. A method comprising: receiving, at a processor in communication with a multimodal sensing device and from one or more sensors of the multimodal sensing device, at least one signal comprising motion data of the multimodal sensing device; determining a torque score based on the received at least one signal corresponding to a standardized flexion and extension maneuver exerted on a subject by an evaluator wearing the multimodal sensing device; and determining an effectiveness of a treatment or an estimated progression of disease based on a comparison of the determined torques score and at least one historical torque score.

18. The method of claim 17, wherein the motion data of the multimodal sensing device comprises a maneuver trajectory of the standardized flexion and extension maneuver exerted on the subject by the evaluator and / or a force expended by the evaluator on the subject.

19. The method of claim 17, wherein determining the torque score based on the received at least one signal corresponding to the standardized flexion and extension maneuver comprises: determining a subset of the received at least one signal corresponding to a flexion maneuver; determining a subset of the received at least one signal corresponding to an extension maneuver; and determining the torque score based on a force corresponding to the flexion maneuver based on the determined subset of the received at least one signal corresponding to a flexion maneuver, a force corresponding to the extension maneuver based on the determined subset of the received at least one signal corresponding to the extension maneuver, a grip length, or an angle of treatment.

20. The method of claim 17 comprising: receiving motion data for a calibration activity; and determining a calibration factor based on a torque score for the received motion data for the calibration activity; and applying the determined calibration factor to one or more future signals received from the multimodal sensing device.

Images