Methods and systems for detecting the progression of a movement disorder

Abstract

A method for detecting the progression of multiple system atrophy (MSA) over a time period using a trained machine learning model for a patient with MSA. The method may include receiving in-clinic sensor data and at-home sensor data for a patient over the time period. The method may include determining respective values indicative of a plurality of different MSA-related activity features for the patient based on the in-clinic sensor data and the at-home sensor data. The method may further include determining, using the trained machine learning model, the progression of MSA over the time period for the patient using the respective values for the plurality of different MSA-related activity features for the patient.

Description

Methods and Systems for Detecting the Progression of a Movement DisorderCROSS-REFERENCE TO RELATED APPLICATIONS

[0001] This application claims the benefit of Provisional U.S. Patent Application No. 63 / 585,458, filed September 26, 2023, and Provisional U.S. Patent Application No. 63 / 636,298, filed April 19, 2024, the entire disclosures of which are hereby incorporated by reference herein in their entirety.BACKGROUND

[0002] Multiple systems atrophy (MSA) is a progressive neurodegenerative disorder characterized by a combination of symptoms that affect both the central nervous system which controls how a person moves, and the autonomic nervous system which controls involuntary functions such as blood pressure or digestion. MSA progresses rapidly and presents first with autonomic symptoms (urinary) following with gait and motor impairment. Gait impairment may include two clinical subtypes: MSA-parkinsonian (MSA-P) and MSA-cerebellar (MSA- C). MSA-P resembles Parkinson’s disease and is characterized by bradykinesia, tremor, spasticity, and loss of coordination. MSA-C more closely resembles progressive ataxias and is characterized by gait and limb ataxia and postural instability. Unlike the subtype’s namesake diseases, MSA progresses rapidly resulting in 50% of the patients requiring a wheelchair within 3 years of diagnosis. Due to this rapid progression and lack of disease modifying treatments, there is an urgent need to develop novel methods to detect treatment efficacy in MSA.

[0003] MSA progression is currently assessed under two scales: Unified Multiple System Atrophy Rating Scale (UMSARS) and MSA health-related quality of life scale (MSA-QoL). Both scales’ limitations at least partially result from the large number of patients required to detect efficacy in interventional trials. Both scales are episodic and rely at least in part on subjective recall. While the UMSARS is currently the most widely used scale for assessing MSA progression, it was developed without patient input, relies nearly entirely on subjective assessments, and shows minimal change over time with a documented ceiling affect. The MSA-QoL was developed from qualitative and quantitative surveys of patients input and is the only validated measure for specifically examining QoL in MSA. However, three fourths of theMSA-QoL measures don’t change within a year, which is the typical length of an interventional trial. In addition, under MSA-QoL, a drug / intervention cannot be reasonably expected to change important factors of the patient’s life, even if altering the course of the disease. Due to the limitations of both UMS ARS and MSA-QoL, there is an urgent need to develop novel methods to provide an objective measure of a change in activity that is indirectly reflective of the health status of the patients of MSA.SUMMARY

[0004] Methods, systems, and computer readable storage mediums (CRMs) comprising instructions may be configured to determine, using the trained machine learning model, the progression of MSA over the time period for a patient using respective values for a plurality of different MSA-related activity features for the patient. Further, methods, systems, and CRMs for training such machine learning models may also be described herein. Additionally, methods, systems, and CRMs for detecting the progression of multiple system atrophy (MSA) over a time period using a trained machine learning model may be described herein.

[0005] A method for detecting the progression of MSA over a time period using a trained machine learning model for a patient with MSA. The method may include receiving in-clinic sensor data and at-home sensor data for a patient (e.g., an individual and / or a patient) over the time period. The method may include determining respective values indicative of a plurality of different MSA-related activity features for the patient based on the in-clinic sensor data and the at-home sensor data. The method may further include determining, using the trained machine learning model, the progression of MSA over the time period for the patient using the respective values for the plurality of different MSA-related activity features for the patient. The machine learning model may be trained to determine the progression of MSA over the time period using respective values for the plurality of different MSA-related activity features for each of a plurality of different patients.

[0006] A computer-readable storage medium may include instructions, that when executed by one or more processors, cause the one or more processors to receive in-clinic sensor data and at-home sensor data for a patient over a time period, determine respective values indicative of a plurality of different multiple system atrophy (MSA)-related activity features for the patient based on the in-clinic sensor data and the at-home sensor data, and determine, using a trained machine learning model, the progression of MSA over the time period for the patient using the respective values for the plurality of different MSA-related activity features for the patient. The machine learning model may have been trained to determine the progression ofMSA over the time period using respective values for the plurality of different MSA-related activity features for each of a plurality of different patients.

[0007] A method for training a machine learning model to detect progression of multiple system atrophy (MSA) over a time period may be described. The method may include receiving in-clinic sensor data and at-home sensor data for a plurality of patients over the time period, determining respective values indicative of a plurality of different MSA-related activity features for each of the plurality of patients based on the in-clinic sensor data and at-home sensor data, and training the machine learning model to determine the progression of MSA over the time period using the respective values for the plurality of different MSA-related activity features for each of the plurality of patients.

[0008] A method for detecting the progression of multiple system atrophy (MSA) over a time period using a trained machine learning model may be described herein. For example, the method may include receiving in-clinic sensor data and at-home sensor data for a patient over the time period, determining respective values indicative of a plurality of different MSA- related activity features for the user based on the in-clinic sensor data and the at-home sensor data, determining, using a trained machine learning model, the progression of MSA over the time period for the user using the respective values for the plurality of different MSA-related activity features for the patient, and orally administering, on a daily basis, emrusolmin based on the progression of MSA over the time period for the user. The machine learning model may have been trained to determine the progression of MSA over the time period using respective values for the plurality of different MSA-related activity features for each of a plurality of different patients

[0009] The progression of MSA over the time period may include at least one of a change in Unified Multiple System Atrophy Rating Scale (UMSARS) Part I score, a change in modified UMSARS Part I score, a change in UMSARS Part II score, or a change in UMSARS Part IV score. The progression of MSA over the period of time may include at least one of a change in combined Unified Multiple System Atrophy Rating Scale (UMSARS) Part I and Part II scores, or a change in combined modified UMSARS Part I and Part II scores.

[0010] The in-clinic sensor data may include one or more of gait speed, stride length, circumduction, dorsiflexion, foot clearance, path deviation, number of steps, step time, mean cadence, a confidential interval of cadence, stride time variability, sit-to-stand duration, sit-to- stand lean angle, stand-to-sit duration, stand-to-sit lean angle, first step duration, first step range of motion, lateral anticipatory postural adjustment (APA) peak, turning speed, turningduration, turning angle, number of steps in turn arm swing variability, arm swing range of motion, arm swing velocity, arm swing asymmetry, arm swing variability, trunk and lumbar coronal, trunk and lumbar sagittal, trunk and lumbar transverse, trunk variability, sway area, path length, and / or jerk. The in-clinic sensor data may be collected during structured tasks performed by a patient, the structured tasks comprising timed stationary standing, timed walk, and / or timed up and go (TUG). The in-clinic sensor data may be collected using a sensor device system. The sensor system may include one or more wrist-worn sensor devices, one or more ankle- worn sensor devices, a sensor platform device, a waist-worn sensor device, and / or a chest-worn sensor device.

[0011] For example, the in-clinic sensor data may include any combination of Anticipatory Postural Adjustment (APA) - Duration, APA - First Step Duration, APA - First Step Range of Motion, APA - Forward APA Peak, APA - Lateral APA Peak, Gait / Lower Limb - Cadence (Left), Gait / Lower Limb - Cadence (Right), Gait / Lower Limb - Circumduction (Left), Gait / Lower Limb - Circumduction (Right), Gait / Lower Limb - Cycle Validity Flags (Left), Gait / Lower Limb - Cycle Validity Flags (Right), Gait / Lower Limb - Double Support (Left), Gait / Lower Limb - Double Support (Right), Gait / Lower Limb - Elevation at Midswing (Left), Gait / Lower Limb - Elevation at Midswing (Right), Gait / Lower Limb - Foot Strike Angle (Left), Gait / Lower Limb - Foot Strike Angle (Right), Gait / Lower Limb - Gait Cycle Duration (Left), Gait / Lower Limb - Gait Cycle Duration (Right), Gait / Lower Limb - Gait Cycle Validity (Left), Gait / Lower Limb - Gait Cycle Validity (Right), Gait / Lower Limb - Gait Speed (Left), Gait / Lower Limb - Gait Speed (Right), Gait / Lower Limb - Lateral Step Variability (Left), Gait / Lower Limb - Lateral Step Variability (Right), Gait / Lower Limb - N, Gait / Lower Limb - Single Limb Support (Left), Gait / Lower Limb - Single Limb Support (Right), Gait / Lower Limb - Stance (Left), Gait / Lower Limb - Stance (Right), Gait / Lower Limb - Step Duration (Left), Gait / Lower Limb - Step Duration (Right), Gait / Lower Limb - Stride Length (Left), Gait / Lower Limb - Stride Length (Right), Gait / Lower Limb - Swing (Left), Gait / Lower Limb - Swing (Right), Gait / Lower Limb - Terminal Double Support (Left), Gait / Lower Limb - Terminal Double Support (Right), Gait / Lower Limb - Toe Off Angle (Left), Gait / Lower Limb - Toe Off Angle (Right), Gait / Lower Limb - Toe Out Angle (Left), Gait / Lower Limb - Toe Out Angle (Right), Gait / Lumbar - Coronal Range of Motion, Gait / Lumbar - Sagittal Range of Motion, Gait / Lumbar - Transverse Range of Motion, Gait / Trunk - Coronal Range of Motion, Gait / Trunk - Sagittal Range of Motion, Gait / Trunk - Transverse Range of Motion, Gait / Upper Limb - Arm Range of Motion (Left), Gait / Upper Limb - Arm Range of Motion (Right),Gait / Upper Limb - Arm Swing Velocity (Left), Gait / Upper Limb - Arm Swing Velocity (Right), Postural Sway / Acc - 95% Ellipse Axis 1 Radius, Postural Sway / Acc - 95% Ellipse Axis 2 Radius, Postural Sway / Acc - 95% Ellipse Rotation, Postural Sway / Acc - Centroidal Frequency, Postural Sway / Acc - Centroidal Frequency (Coronal), Postural Sway / Acc - Centroidal Frequency (Sagittal), Postural Sway / Acc - Frequency Dispersion, Postural Sway / Acc - Frequency Dispersion (Coronal), Postural Sway / Acc - Frequency Dispersion (Sagittal), Postural Sway / Acc - Jerk, Postural Sway / Acc - Jerk (Coronal), Postural Sway / Acc - Jerk (Sagittal), Postural Sway / Acc - Mean Velocity, Postural Sway / Acc - Mean Velocity (Coronal), Postural Sway / Acc - Mean Velocity (Sagittal), Postural Sway / Acc - Path Length, Postural Sway / Acc - Path Length (Coronal), Postural Sway / Acc - Path Length (Sagittal), Postural Sway / Acc - RMS Sway, Postural Sway / Acc - RMS Sway (Coronal), Postural Sway / Acc - RMS Sway (Sagittal), Postural Sway / Acc - Range, Postural Sway / Acc - Range (Coronal), Postural Sway / Acc - Range (Sagittal), Postural Sway / Acc - Sway Area, Postural Sway / Angles - 95% Ellipse Axis 1 Radius, Postural Sway / Angles - 95% Ellipse Axis 2 Radius, Postural Sway / Angles - 95% Ellipse Rotation, Postural Sway / Angles - Duration, Postural Sway / Angles - RMS Sway, Postural Sway / Angles - RMS Sway (Coronal), Postural Sway / Angles - RMS Sway (Sagittal), Postural Sway / Angles - Sway Area, Turns - Angle, Turns - Duration, Turns - N, Turns - Steps in Turn, Turns - Turn Velocity, Gait / Lower Limb - Gait Cycle Timing (Left), Gait / Lower Limb - Gait Cycle Timing (Right), Gait / Lower Limb - Toe Off Timing (Left), Gait / Lower Limb - Toe Off Timing (Right), Gait / Lower Limb - Midswing Timing (Left), Gait / Lower Limb - Midswing Timing (Right), Gait / Lower Limb - Initial Contact Next Timing (Left), Gait / Lower Limb - Initial Contact Next Timing (Right), Gait / Lower Limb - All Cycles Initial Contact Timing (Left), Gait / Lower Limb - All Cycles Initial Contact Timing (Right), Turns - Turn Timing, Duration, Sit to Stand - Duration, Sit to Stand - Lean Angle, Sit to Stand - N, Sit to Stand - Sit to Stand Timing, Stand to Sit - Stand to Sit Timing, Duration Sway test, and / or Duration Tug test, for example, as described in Table 1.

[0012] The in-clinic sensor data may be collected during structured tasks performed by a patient, the structured tasks comprising timed stationary standing, timed walk, or timed up and go (TUG). The at-home sensor data may include one or more of number of steps, step time, mean cadence, a confidential interval of cadence, stride time variability, sit-to-stand duration, active time, inactive time, walking duration, stand duration, sitting / lying duration, total sleep duration, sleep efficiency, wake after sleep onset count, wake after sleep onset duration, and / or wake after sleep onset median. The at-home sensor data may be collected in response tounprompted movements and activities performed by the respective patient. The at-home sensor data may be collected using a wrist- worn sensor device. In one example, the at-home sensor data may be collected using different sensor devices than the in-clinic sensor data. For example, he at-home sensor data may be collected using a wrist- worn sensor device, and the in-clinic sensor may be is collected using a sensor device system comprising one or more wrist-worn sensor devices, one or more ankle-worn sensor devices, a sensor platform device, a waist- worn sensor device, and a chest-worn sensor device. For instance, the in-clinic sensor data may include more types of sensor data than the at-home sensor data.

[0013] In some examples, the method may include receiving survey data for the plurality of patients over the time period and training the machine learning model using the survey data. The survey data may be used to identify symptoms meaningful to the patient and / or connections between the MSA-related activity features and the in-clinic sensor data or at-home sensor data. For example, the survey may provide a series of questions related to a meaningfulness of each of the MSA-related activity features, and the method may further include mapping the MSA-related activity features to one or more meaningful symptoms. In one example, the method may further include determining a subset of the MSA-related activity features based on the survey data, and training the machine learning model using the subset of the MSA-related activity features.

[0014] The survey data may be collected in response to a survey prompting a patient to identify symptoms meaningful to the patient or connections between the MSA-related activity features and the in-clinic sensor data or at-home sensor data. For example, the survey may provide a series of questions related to a meaningfulness of each of the MSA-related activity features. In some examples, the method may include mapping the MSA-related activity features to one or more meaningful MSA-related symptoms. Further, in some examples, the method may include determining a subset of the MSA-related activity features based on the survey data, and training the machine learning model using the subset of the MSA-related activity features.

[0015] In some examples, the plurality of MSA-related activity features may include gait features, sit-to-stand features, sleep features, walking initiation features, turning features, arm movement features, trunk and low back range of motion features, and / or postural sway features.

[0016] In some examples, the time period may be 48 weeks. The in-clinic sensor data and the survey data may be collected at a given frequency for each of the plurality of patients overthe time period. For example, the given frequency is every 6 weeks. The at-home sensor data may be collected continuously for each of the plurality of patients over the time period.

[0017] In some examples, the plurality of patients may include at least 100 patients. For example, the plurality of patients may be 160 patients.

[0018] In some examples, a subset of the plurality of patients may take a medication associated with MSA, and the remainder of the plurality of patients are not administered the medication. The medication is a chemical or biologic medication. For example, the medication is a treatment for one or more symptoms of MSA. The medication is administered pursuant to a clinical trial. The medication may comprise emrusolmin, droxidopa, levodopa, ropinirole, pramipexole, clonazepam, vitamin E, propranolol, baclofen, amantadine, buspirone, ampreloxetine, ENT-01, Lu AF82422, AAV2-GDNF, ION464 and / or ATH434.

[0019] In some examples, the method may include administering to the patient a medication associated with MSA. The medication may be a chemical or biologic medication. The medication may be a treatment for one or more symptoms of MSA. The medication may be administered pursuant to a clinical trial. The medication may comprise emrusolmin, droxidopa, levodopa, ropinirole, pramipexole, clonazepam, vitamin E, propranolol, baclofen, amantadine, buspirone, ampreloxetine, ENT-01, Lu AF82422, AAV2-GDNF, ION464 and / or ATH434.

[0020] In some examples, the method may further include segmenting the plurality of patients into a plurality of groups. The method may include training the machine learning model using the respective values for the plurality of different MSA-related activity features for the patients of one of the plurality of groups. The method may include testing the machine learning model using the respective values for the plurality of different MSA-related activity features for the patients of a remainder of the plurality of groups. In one example, each group of the plurality of groups may include the same number of patients.

[0021] In some examples, the method may further include training the machine learning model using covariates associated with each of the plurality of patients. The covariates mayinclude clinical MSA subtype of the patient (e.g., MSA-P or MSA-C), age of the patient, gender of the patient, or neurofilament light chain levels of the patient.

[0022] In some examples, the machine learning model may include a classification and regression model.

[0023] In some examples, the method may include administering (e.g., orally administering) a medication, wherein the medication comprises one or more of emrusolmin, droxidopa, levodopa, ropinirole, pramipexole, clonazepam, vitamin E, propranolol, baclofen, amantadine, buspirone, ampreloxetine, ENT-01, Lu AF82422, AAV2-GDNF, ION464 and / or ATH434.For instance, the medication may be emrusolmin, and about 300 mg / day of emrusolmin may be orally administered to the patient. The patient may have a diagnosis of clinically definite MSA, clinically probably MSA, or clinically possible MSA. In some examples, each of the plurality of patients may have a diagnosis of clinically definite MSA, clinically probably MSA, or clinically possible MSA.BRIEF DESCRIPTION OF THE DRAWINGS

[0024] FIG. 1 is a diagram of an example system for collecting data from various sources and generating a machine learning model capable of determining the progression of multiple systems atrophy (MSA) of a patient with MSA.

[0025] FIG. 2 is a block diagram that illustrates an example of a computing device of the example system shown in FIG. 1.

[0026] FIG. 3 is a flow diagram that illustrates a training process for training a machine learning model to determine the progression of MSA of a patient with MSA.

[0027] FIG. 4 is a flow diagram that illustrates an example process for using a trained machine learning model to determine the progression of MSA of a patient with MSA.

[0028] FIG. 5 is a schematic illustrating an example of a study described herein.DETAILED DESCRIPTION

[0029] The present subject matter may be understood more readily by reference to the following detailed description which forms a part of this disclosure. It is to be understood that this disclosure is not limited to the specific methods, conditions or parameters described and / or shown herein, and that the terminology used herein is for the purpose of describing particular embodiments by way of example only and is not intended to be limiting of the claimedinvention.

[0030] Unless otherwise defined herein, scientific and technical terms used in connection with the present application shall have the meanings that are commonly understood by those of ordinary skill in the art. Further, unless otherwise required by context, singular terms shall include pluralities and plural terms shall include the singular.

[0031] As employed above and throughout the disclosure, the following terms and abbreviations, unless otherwise indicated, shall be understood to have the following meanings.

[0032] In the present disclosure the singular forms “a,” “an,” and “the” include the plural reference, and reference to a particular numerical value includes at least that particular value, unless the context clearly indicates otherwise. Thus, for example, a reference to “a compound” is a reference to one or more of such compounds and equivalents thereof known to those skilled in the art, and so forth. When a range of values is expressed, another embodiment includes from the one particular and / or to the other particular value. Similarly, when values are expressed as approximations, by use of the antecedent “about,” it is understood that the particular value forms another embodiment. All ranges are inclusive and combinable.

[0033] As used herein, the terms “composition,” “compound,” “drug,” or “therapeutic” are used interchangeably herein to refer to a compound or compounds or composition of matter which, when administered to a subject induces a desired pharmacological and / or physiologic effect by local and / or systemic action.

[0034] As used herein, the terms “treatment” or “therapy” (as well as different forms thereof) include preventative (e.g., prophylactic), curative, or palliative treatment. As used herein, the term “treating” includes alleviating or reducing at least one adverse or negative effect or symptom of a condition, disease or disorder. This condition, disease or disorder refers to MSA.

[0035] The term “administering” means either directly administering a compound or composition of the present disclosure, or administering a prodrug, derivative or analog which will form an equivalent amount of the active compound or substance within the body.

[0036] The terms “subject,” “individual,” and “patient” are used interchangeably herein, and refer to a human. In some aspects, the human patient is an adult patient, i.e., 18 years of age of older.

[0037] The term “baseline” as used herein refers to the period of time just prior to administration of emrusolmin. The human’s condition prior to administration of emrusolmin isthe human’s “baseline condition.”

[0038] The term “control population” as used herein refers to one or more humans participating in a research study, for example, a clinical trial study, who do not receive the test treatment and whose experience allows discrimination of outcomes caused by the test treatment from outcomes caused by factors other than the test treatment. Those of ordinary skill in the art can identify one or more humans participating in a research study who would be appropriate members of a control population. Guidance for selecting humans for a control population is provided in, for example, European Medicines Agency. (2002). ICH Topic E10: Choice of control group in clinical trials, https: / / www.ema.europa.eu / en / documents / scientific- guideline / ich-e-10-choice-control-group-clinical-trials-step-5 _en.pdf.

[0039] “Pharmaceutically acceptable” refers to those compounds, materials, compositions, and / or dosage forms which are, within the scope of sound medical judgment, suitable for contact with the tissues of human beings without excessive toxicity, irritation, allergic response, or other problem complications commensurate with a reasonable benefit / risk ratio.

[0040] The term “emrusolmin” as used herein refers to 5-(l,3-benzodioxol-5-yl)-3-(3- bromophenyl)-lH-pyrazole and has the following chemical structure.Emrusolmin is also known in the art as anle-138b and TEV-56286.

[0041] Emrusolmin also refers to any tautomers thereof, such as the following tautomer:Methods

[0042] A. Individual Humans

[0043] The present disclosure relates to methods of using emrusolmin in humans diagnosed with Multiple System Atrophy (MSA). In certain embodiments, the disclosure provides methods of stabilizing one or more symptoms of Multiple System Atrophy (MSA) in a humanin need of stabilization thereof. The methods include orally administering, on a daily basis, 300 mg of emrusolmin to the human.

[0044] “Multiple systems atrophy” refers to an adult-onset, fatal neurodegenerative disease characterized by progressive autonomic failure and parkinsonism, cerebellar, or pyramidal features in various combinations. In some embodiments, the MSA is multiple system atrophyparkinsonian type (MSA-P). In other embodiments, the MSA is multiple system atrophy- cerebellar subtype (MSA-C).

[0045] Symptoms of MSA include hypokinetic rigid parkinsonian, gait ataxia, limb ataxia, cerebellar oculomotor impairment, scanning dysarthria, intention tremor, focal dystonia and posture abnormalities. An MSA patient may experience once or more of these symptoms, in any combination. In some embodiments, the MSA symptom is hypokinetic rigid parkinsonian. In other embodiments, the MSA symptom is gait ataxia. In further embodiments, the MSA symptom is limb ataxia. In yet other embodiments, the MSA symptom is a cerebellar oculomotor impairment. In still further embodiments, the MSA symptom is scanning dysarthria. In other embodiments, the MSA symptom is intention tremor. In further embodiments, the MSA symptom is focal dystonia. In still other embodiments, the MSA symptom is a posture abnormality.

[0046] The methods described herein may also include diagnosing the human with MSA. MSA may be diagnosed by one skilled in the art. In some embodiments, the MSA is “clinically possible” or “clinically probably” as determined by the Gilman criteria (2008). See, Gilman, Second consensus statement on the diagnosis of multiple system atrophy. Neurology 2008; 71(9):670-6).

[0047] In some embodiments, the human is diagnosed with MSA-P. In other embodiments, the human is diagnosed with MSA-C.

[0048] FIG. 1 is a diagram of an example system 100 for collecting data from various data sources (e.g., sensors and / or survey data) and generating (e.g., training) a machine learning model capable of determining the progression of multiple system atrophy (MSA) of a patient e.g., a patient or an individual). The patient may be a patient who has received a positive diagnosis of MSA. The system 100 may include a computing device 110, one or more sensor devices 120a-120n, and one or more remote client devices 130a-130c.

[0049] As described in more detail herein, the computing device 110 may generate (e.g., train) and / or use a machine learning model capable of determining an MSA digital activity score (MSA-DAS) based on the data received from the sensor devices 120a- 120n and / or one ormore surveys (e.g., quantitative surveys and / or qualitative surveys). For example, the computing device 110 may be configured to generate a model based on a data-driven development of an activity-based efficacy measure in MSA. In one embodiment, the MSA- DAS represents the progression of MSA in a patient. As such, the computing device 110 may train the machine learning model to allow the model to determine progression of MSA on a patient with MSA, for example, based on the data received from the sensor devices 120a- 120n.

[0050] The computing device 110 may comprise a server or combination of servers that may be configured to receive data from various sources, such as the sensor devices 120a- 120n and / or the remote client devices 130a-130c. The computing device 110 may comprise an operational system or subsystem and an analytical system or subsystem. The computing device 110 may include a processor (e.g., one or more processors), and the processor may store information in and / or retrieve information from memory of the computing device 110. The memory may include a non-removable memory and / or a removable memory. The nonremovable memory may include random-access memory (RAM), read-only memory (ROM), a hard disk, or any other type of non-removable memory storage. The removable memory may include a subscriber identity module (SIM) card, a memory stick, a memory card, or any other type of removable memory. The memory may be local memory or remote memory external to the computing system 110. The memory may store instructions which are executable by the processor. Different information may be stored in different locations in the memory.

[0051] The memory may comprise a computer-readable storage media or machine-readable storage media that maintains computer-executable instructions for performing one or more as described herein. For example, the memory may comprise computer-executable instructions or machine-readable instructions that include one or more portions of the procedures described herein. The processor of the computing device 110 may access the instructions from memory for being executed to cause the processor of the computing device 110 to operate as described herein. The memory may comprise computer-executable instructions for executing configuration software. For example, the computer-executable instructions may be executed to perform, in part and / or in their entirety, one or more procedures as described herein. Further, the memory may have stored thereon one or more settings and / or control parameters associated with the computing device 110.

[0052] The computing device 110, the sensor devices 120a- 120n, and one or more remote client devices 130a- 130c may be connected via a network, such as a public and / or private network 140 (e.g., any combination of the Internet, a cloud network, and / or the like). Thenetwork 140 may comprise any suitable network over which computing device 110, one or more sensor devices 120a-120n, and one or more remote client devices 130a-130c may communicate. In some examples, the sensor devices 120a- 120n may communicate with a client device 130a- 130c, and the client device 130a- 130c may communicate data (e.g., altered or unaltered) from the sensor device 120a- 120n to the computing device 110. The network 140 may include a wired and / or wireless communication network. Example wireless communication networks may be comprised of one or more types of RF communication signals using one or more wireless communication protocols, such as a cellular communication protocol, a WIFI communication protocol, and / or another wireless communication protocol.

[0053] The sensor devices 120a- 120n may be wearable by a patient (e.g., the patient with MSA). For example, the sensor devices 120a-120n may be a wearable watch, bracelet, band, shirt, etc. that is worn by the patient, for example, during everyday activities and / or during a clinical setting. For instance, the sensor devices 120a- 120n may comprise one or more at- home sensor devices (z.e., one or more sensor devices worn outside the clinic during everyday activities), and one or more in-clinic sensor devices (z.e., one or more sensor devices worn by a patient during a structured clinical setting). The at-home and in-clinic sensor devices used in each setting may be the same or different. The sensor devices 120a- 120n may include any combination of sensors, such as an accelerometer (e.g., a three-way accelerometer), a light sensor, a gyroscope, and / or a temperature sensor. For example, the at-home sensor device can include an accelerometer, a light sensor, and a temperature sensor, while the in-clinic sensor can include an accelerometer and a gyroscope. In one embodiment, the in-clinic sensor device may have a higher degree of inter- and intra-device reliability than the at-home sensor device. Further, though each may be described as singular, the at-home and / or in-clinic sensor devices can comprise multiple physical components to be simultaneously or sequentially utilized by a patient. In some embodiments, the in-clinic sensor device may be configured to capture a greater variety of types of sensor data as compared to the at-home sensor device. For example, the in-clinic sensor device may have a greater number of sensors than the at-home sensor device. For instance, in some examples, the at-home sensor may include a single component and / or sensor, while the in-clinic sensor may include a plurality of components and / or sensors. For example, the at-home sensor may include a single wearable component, while the in-clinic sensor may include a plurality of wearable components. As such, the at-home sensor is suitable for use at the patient’s home and throughout normal day-to-day activities, while the inclinic sensor may provide more complex or additional sensor information, but may be lesssuitable for use throughout normal day-to-day activities.

[0054] The sensor device 120a- 120n are configured to provide objective data related to health concepts meaningful to MSA patients. Such health concepts include ambulation (e.g., gait and speed, walking volume), postural change, postural sway, physical intensity duration, and sleep quality. These health concepts are both measurable via the sensor device 120a- 120n, and sensitive to change in MSA severity. These health concepts can be embodied by specific health experiences of patients, including walking (e.g., ambulation (e.g., walking volume)); gait and leg agility (e.g., ambulation (e.g., gait and speed)); arising from a chair, standing up, toileting, and balance (e.g., postural change); general physical activities (e.g., physical intensity duration); and excessive daytime sleeping, REM sleep behavior disorder, and sleep apnea (e.g., sleep quality). Collectively, these health concepts and specific health experiences can be referred to herein as activity features.

[0055] The computing system 110 and / or the one or more sensor devices 120a- 120n may be configured to generate sensor data based on the data captured by the one or more sensor devices 120a- 120n that objectively measure activity features. For example, the sensor data may be categorized based on whether it was captured using an in-clinic sensor device (e.g., while the patient is in a clinic and movement in accordance with a physician’s instructions) or captured using an at-home sensor device (e.g., during the patient’s everyday use and activities outside of the clinic environment). For example, the in-clinic sensor data may be collected during structured tasks performed by a patient, where for instance, the structured task may include timed stationary standing (e.g., modified Roberg test), timed walk (directing the patient to walk along a predetermined path for a predetermined period of time), and / or timed up and go (TUG) (directing the patient to stand and walk from an initially seated position, turning, and returning to a seated position). Additionally, each of the structured tasks may be performed utilizing progressively difficult instructions: first, each task may be performed with the patient’s eyes open and feet apart. Then, each task may be performed with the patient’s eyes closed and feet apart. Eastly, each task may then be performed with the patient’s eyes open and feet together.

[0056] In contrast, the at-home sensor data may be collected in response to unprompted movements and activities performed naturally by the respective patient and unprompted by a clinician. The sensor device used to collect the in-clinic sensor data may be the same as, or different from, the sensor device used to collect the at-home sensor data. In one embodiment, the at-home sensor device comprises a single wrist-worn sensor device, and the in-clinic sensorcomprises a sensor device system comprising one or more wrist- worn sensor devices, one or more ankle- worn sensor devices, a sensor platform device, a waist-worn sensor device, and / or a chest-worn sensor device. In particular, the in-clinic sensor device may comprise two wrist- worn sensor devices, two ankle-worn sensor devices, a sensor platform device, a waist-worn sensor device, and a chest-worn sensor device.

[0057] The in-clinic sensor data may include any combination of gait speed, stride length, circumduction, dorsiflexion, foot clearance, path deviation, number of steps, step time, mean cadence, a confidential interval of cadence, stride time variability, sit-to-stand duration, sit-to- stand lean angle, stand-to-sit duration, stand-to-sit lean angle, first step duration, first step range of motion, lateral anticipatory postural adjustment (APA) peak, turning speed, turning duration, turning angle, number of steps in turn arm swing variability, arm swing range of motion, arm swing velocity, arm swing asymmetry, arm swing variability, trunk and lumbar coronal, trunk and lumbar sagittal, trunk and lumbar transverse, trunk variability, sway area, path length, and / or jerk. A full list of potential in-clinic sensor data is shown below in Table 1. The at-home sensor data may include any combination of (e.g., each on a per day basis), a number of steps, step time (e.g., including number of slow steps and number of fast steps), mean cadence, a confidential interval of cadence, stride time variability, sit-to-stand transitions, active time, inactive time, walking duration, stand duration, sitting / lying duration, sleep onset time, sleep end time, total sleep duration, sleep efficiency, wake after sleep onset count, wake after sleep onset duration, and / or wake after sleep onset median. Each component of the inclinic sensor data and the at-home sensor data can be linked to one or more of the abovedescribed health concepts meaningful to MSA patients.Table 1: Examples of In-Clinic Sensor Data

[0058] In one example, the wearable sensor devices may be utilized to detect sensor data (e.g., activity features) in two settings — in-clinic settings (detected by the one or more in-clinic sensor devices) and real- world settings (z.e., at-home settings as detected by the one or more at-home sensor devices). The acquisition of data utilizing both one or more in-clinic sensor devices in addition to one or more at-home sensor device may provide a plurality of complementary and mitigating benefits. The in-clinic settings may provide a high-context environment to capture specific gait features with greater detail than the specific gait features captured in real-world data. For example, the selected in-clinic settings targeted to MSA-P and MSA-C relevant symptoms may include carefully formulated and implemented clinician-directed tests, as described above. Further, the in-clinic sensor device may comprise more physical sensing components than the at-home sensor device (and thus provide a wider array of data capture and / or increased data accuracy) that would be impractical for patient utilization during routine daily tasks.

[0059] In contrast, sensor data collected via the one or more at-home sensor devices may have the benefit of providing data over a longer period of time that is not limited by the temporal restrictions of a clinical setting. Additionally, the at-home sensor devices can limit in-clinic confounding variables, which include for example recall bias for patient reported outcomes, motivational status, and / or observer effects in performance outcomes. The two settings may inform and complement one another, particularly where the respective data is analyzed by the computing device 110 and utilized to train a machine learning model as described herein. The computing device 110 (e.g., the MSA progression model) may also identify at-home sensor data that is more sensitive to disease progression than the in-clinic sensor data. In some examples, both in-clinic and at-home sensor data may be used to train a machine learning model to build a multi-feature score (e.g., composite score) that represents MSA disease progression in a patient. By measuring activity using two or more sensor types in the two distinct collection settings described above, the likelihood of detecting natural disease progression and treatment effects may be increased, standardized, and rendered more objective as compared to a standard clinical outcome assessment (COA).

[0060] Accordingly, the computing system 110 may receive sensor data (e.g., an activity features) for both in-clinic and at-home activities from one or more in-clinic sensor device and one or more at-home sensor devices. Sensor data derived from in-clinic and at-home sensor devices 120a- 120n may be provided to a trained machine learning model (described further below), which may be used to determine disease progression in patients with MSA. Alternatively, the sensor data may be utilized by the computing device 110 to train a machine learning model (as described further below).

[0061] In various embodiments, the sensor data collected may be analyzed to determine activity features particularly relevant to MSA, such as ambulation e.g., steps, cadence, gait features, etc.) and general movement patterns (e.g., movement duration / frequency, sedentary vs. active time, sleep / wake time, etc.), as described further below. The computing device 110 may determine respective values indicative of a plurality of different activity features (e.g., MSA-related activity features) for the patient based on the in-clinic sensor data and the at- home sensor data. In some examples, the plurality of MSA-related activity features may include one or more of gait features, sit-to-stand features, sleep features, walking initiation features, turning features, arm movement features, trunk and lower back range of motionfeatures, and / or postural sway features. For example, the computing device 110 may determine activity features from the raw sensor data captured by the one or more at-home sensor devices and the one or more in-clinic sensor devices (e.g., accelerometer data, gyroscope data, light data, and temperature data). However, in other examples, the sensor data may itself be in the format representative of an activity feature (e.g., the computing device 110 may receive an activity feature directly from the sensor device itself that has itself performed raw data analysis, as opposed to requiring the computing device 110 to compute a value of the activity feature from raw sensor data).

[0062] The computing system 110 may train a machine learning model to determine the progression of MSA over the time period using respective values for a plurality of different MSA-related activity features captured by one or more in-clinic sensor devices and one or more at-home sensor devices for each of a plurality of different patients. The computing system 110 may then use the trained machine learning model to determine the progression of MSA over the time period for a patient using the respective values for the sensor data and / or plurality of different MSA-related activity features for a patient.

[0063] Additionally, in some examples, the computing system 110 may receive feedback from one or more patient surveys and / or quality of life (QoL) measures that prompt the patient to identify (e.g., evaluate, rank, score, etc.) whether sensor data, activity features, and or general MSA-related symptoms are meaningful to patients. Though it is contemplated that in some embodiments the computing device 110 may utilize all of the sensor data and / or activity features, in other embodiments the computing device 110 may not use all sensor data and / or all activity features to generate (e.g., train) a machine learning model capable of determining the progression of MSA of a patient. In other words, the computing device may limit the categories of sensor data and / or number of activity features to train the machine learning model capable of determining the progression of MSA of the patient. The computing device 110 may determine the content validity of the sensor data. For example, the computing device 110 may determine whether particular sensor data and / or activity feature relates to a symptom that the patient deems as meaningful based on the feedback received via the one or more patient surveys and / or QoL measures. The computing device 110 may be provided with predetermined associations between certain sensor data and / or activity features and particular MSA-related symptoms.

[0064] The computing device 110 may train the machine learning model based solely on activity features and / or sensor data with high content validity and may not use activity featuresand / or sensor data with low content validity. The activity features and / or sensor data showing high content validity may include the features that map back to the meaningful symptoms and are reported as important to measure by the patients. In one example, the computing device 110 may train the machine learning model using MSA-related activity features with high content validity (e.g., meaningful features). The computing device 110 may generate (e.g., train) the model by providing an output that is highly related to the disease and consequential to patients or providers while capitalizing on the machine learning model's ability to utilize all possible feature to detect patterns unseen by humans and traditional statistical approaches. The generated algorithm e.g., the machine learning model) may then be applied to detect treatment differences.

[0065] As noted above, the computing system 110 may receive feedback from one or more patient surveys and / or QoL measures that identify whether activity features and / or sensor data are meaningful to patients. In one embodiment, a patient survey may require patients to rank a range of activity features on a numerical scale (e.g., 0-4 points). The activity features may be classified by patients as strongly / profoundly meaningful if rated by the patients with a score of three or more in a 0-4-point scale. In contrast, activity features may be classified by patients as not meaningful if rated by the patients with a score of one or less on a 0-4 point scale. The sensor devices 120a- 120n in the system 100 may capture greater than fifty disease-relevant movement features from in-clinic or at-home settings, as described above. The generated machine learning model may identify, leverage, more highly weight, and / or use the activity features and / or sensor data that the surveys indicate are of relatively higher meaningfulness to the patient. To ensure that the computing device 110 and the sensor devices 120a- 120n accurately quantify the meaningful activity features (e.g., and / or sensor data), “fit for purpose” standards may be applied. The “fit for purpose” standards may include standards on content validity, reliability, construct validity, and / or ability to detect change.

[0066] In some examples, the computing device 110 may receive survey data entered from one or more remote client devices 130a-130c. The one or more remote client devices 130a- 130c may include smartphones, laptops, browsers, and the like. In addition, the survey data may include a qualitative survey and / or a quantitative survey.

[0067] In some embodiments, the survey may prompt the patient to rank activity features, as described above, or may prompt the patient to rank symptoms associated with MSA. In another embodiment, the survey may include MSA symptoms (e.g., including activity features) such as muscle stiffness and rigidity, poor balance, frequent falls, slow movement, tremors,vocal changes, and the like. By utilizing MSA symptoms rather than specific activity features, the patient may be presented with more familiar concepts that elicit a more accurate response. When the survey includes MSA symptoms, the computing device 110 may be preprogrammed with associations between the activity features and / or sensor data and the corresponding MSA symptoms. The MSA-related activity features may include one or more of gait features, sit-to- stand features, activity features, sleep features, walking initiation features, turning features, arm movement features, trunk and low back range of motion features, and / or postural sway features. For example, the MSA-related activity features may include one or more of gait speed, stride length, circumduction, dorsiflexion, foot clearance, path deviation, number of steps, step time, mean cadence, a confidential interval of cadence, stride time variability, sit-to- stand duration, sit-to-stand lean angle, stand-to-sit duration, stand-to-sit lean angle, first step duration, first step range of motion, lateral anticipatory postural adjustment (APA) peak, turning speed, turning duration, turning angle, number of steps in turn arm swing variability, arm swing range of motion, arm swing velocity, arm swing asymmetry, arm swing variability, trunk and lumbar coronal, trunk and lumbar sagittal, trunk and lumbar transverse, trunk variability, sway area, path length, and / or jerk. As stated above, the survey may be utilized to train the machine learning model such that the model may identify, leverage, more highly weight, and / or use the activity features and / or sensor data that the survey indicates are of relatively higher meaningfulness to the patient.

[0068] The computing device 110 may utilize the survey to identify connections between MSA-related activity features as represented by the in-clinic sensor data and / or the at-home sensor data. For example, the survey may provide a series of questions related to a meaningfulness of each of the MSA-related activity features.

[0069] The patient’s in-clinic sensor data and at-home sensor data may span (e.g., have been collected over) a certain period of time. In one example, the time period may be 48 weeks. During the certain period of time, the in-clinic data and the survey data may be collected at a given frequency for each of the plurality of patients. For example, the given frequency may be collected at every 6 weeks. Meanwhile, during the certain period of time, the at-home data may be collected continuously for each of the plurality of patients.

[0070] In some examples, the computing device 110 may train the machine learning model using key covariates associated with each of the plurality of patients. The computing device 110 may receive the key covariates from the one or more client devices 130a-130c. The keycovariates may include MSA subtype of the patient (e.g., MSA-C or MSA-P), age of the patient, gender of the patient, and / or neurofilament light chain levels of the patient.

[0071] The computing device 110 may generate a notification indicating a MSA digital activity score (MSA-DAS). In one embodiment, the MSA-DAS represents the progression of MSA of the patient over the period of time. In one example, the notification may indicate whether the patient’s MSA has progressed by more than a threshold amount. In another example, the notification may indicate the degree of change of MSA severity in a patient (e.g., by a numerical or percentage amount). In each case, the degree of change of MSA severity may be represented by a change in score for at least one part of the UMSARS for the patient as determined by the machine learning model. In one embodiment, the degree of change of MSA severity may comprise a change in UMSARS Part I score for a patient. Alternatively, it is contemplated that the degree of change of MSA severity may comprise a change in UMSARS Part II score or UMSARS Part IV score. In another embodiment, the degree of change of MSA severity may comprise a change in modified Part I UMSARS score. In a further embodiment, the degree of change of MSA severity may comprise a change in combined UMSARS Part I and Part II score, or modified UMSARS Part I and Part II score. In a further embodiment, the degree of change of MSA severity may comprise a change in select components of UMSARS Part I and / or UMSARS Part II. A more detailed description of UMSARS is provided herein, for example, with reference to the example study described in reference to FIG. 5.

[0072] Notably, the degree of change in MSA severity as determined by the machine learning model of the present disclosure may not comprise a change in UMSARS Part III.

[0073] In one embodiment, the machine learning model is utilized to determine progression of MSA in a patient over a period of time, where the patient is not undergoing treatment for MSA. In a specific embodiment, the machine learning model may be utilized to determine progression of MSA in a patient as a measure of efficacy of a medication for the treatment of MSA, or a medication for treatment of one or more symptoms of MSA, over a period of time. For example, the machine learning model may be utilized to determine medication efficacy in a clinical trial with two patient groups — a control group and a group receiving the treatment. In another embodiment, the machine learning model may be utilized to determine treatment efficacy during a particular patient’s treatment regimen, such as when compared to a pretreatment time period or treatment of the patient with a different medication.

[0074] For example, the medication may be a chemical or biologic medication. In one embodiment, the medication is a treatment for one or more symptoms of MSA. In anotherembodiment, the medication is administered pursuant to a clinical trial. In a particular embodiment, the medication may be emrusolmin. Emrusolmin is described further in U.S. Application No. 14 / 978,964 filed Dec. 22, 2015 and issued as U.S. Patent No. 10,071,966, the disclosure of which is incorporated by reference in its entirety. In an embodiment, the medication is emrusolmin, and each patient is administered about 300 mg / day of emrusolmin. In one embodiment, each patient is administered about 300 mg / day of emrusolmin in a single dose. In other embodiments, the medication may comprise one or more of droxidopa, levodopa, dopamine agonists such as ropinirole and pramipexole, clonazepam, vitamin E, propranolol, baclofen, amantadine, buspirone, ampreloxetine, ENT-01, Lu AF82422, AAV2- GDNF, ION464 and ATH434.

[0075] The term “stabilizing” as used herein with reference to MSA symptoms refers to preventing the worsening of the one or more MSA symptom in a human. Thus, although the MSA symptoms may be still present, the methods herein prevent the symptoms from worsening. Stabilization may be assessed by a person skilled in the art using, for example, the Unified Multiple System Atrophy Rating Scale (UMSARS) score, or a modification thereof. In some aspects, stabilization may be determined as a change from baseline in the human’s UMSARS score, or modification thereof.

[0076] The disclosure also relates to methods of limiting the clinical progression of MSA, for example, limiting the clinical progression of one or more symptoms of MSA, in a human in need of limitation thereof. The methods include orally administering, on a daily basis, 300 mg of emrusolmin to the human. The term “limiting” as used herein with reference to MSA symptoms refers to limiting the progression of one or more MSA symptoms or in the human or the human developing one or more new MSA symptoms. The limiting may be determined by a person of skill in the art, for example, by assessing the change from baseline in the human’s UMSARS score, or a modification thereof. In some embodiments, limiting the clinical progression comprises reducing or stopping the worsening and / or progression of one or more MSA symptoms.

[0077] The disclosure further provides methods of slowing the clinical progression of MSA in a human in need of slowing thereof. The methods include orally administering, on a daily basis, 300 mg of emrusolmin to the human. As used herein, the term “slowing the clinical progression of MSA” refers to reducing the deterioration experienced by a MSA patient. Slowing may be determined by a person of skill in the art, e.g., by assessing changes in UMSARS score. In some aspects, the slowing may be assessed as a comparison to thedeterioration experienced by a control MSA patient not receiving emrusolmin over the same period of time. For example, in certain of these aspects, the slowing may be assessed by an average less worsening of the UMSARS at week 48 compared to baseline, e.g., comparing the UMSARS scores of treated and non-treated groups. In some embodiments, slowing the clinical progression comprises slowing the worsening of MSA symptoms, as determined by those skilled in the art.

[0078] The disclosure also provides methods of stabilizing the clinical progression, limiting the clinical progression, and / or slowing the clinical progression of MSA in a human in need thereof. The methods include orally administering, on a daily basis, 300 mg of emrusolmin to the human. The clinical progression can be determined by those skilled in the art, for example, by assessing changes in the human’s total UMSARS score of part I and part II combined, or a modification thereof, over time. For example, clinical progression can be determined by assessing a change in UMSARS from baseline, over time. In some embodiments, stabilizing the clinical progression comprises reducing or stopping the worsening of MSA symptoms, as determined by those skilled in the art.

[0079] The disclosure further provides methods of stabilizing the clinical progression, limiting the clinical progression, and / or slowing the clinical progression of MSA in a human in need thereof. The methods include orally administering, on a daily basis, 300 mg of emrusolmin to the human. In some aspects, those skilled in the art can determine clinical progression as a change in the human’s CGI-S score over time, for example as a change from baseline over time. In some embodiments, stabilizing, limiting, or slowing the clinical progression comprises reducing or stopping the worsening of MSA symptoms, as determined by those skilled in the art.

[0080] The disclosure also provides methods of stabilizing the clinical progression, limiting the clinical progression, and / or slowing the clinical progression of MSA in a human in need thereof. The methods include orally administering, on a daily basis, 300 mg of emrusolmin to the human. Clinical progression may be determined as a change in the human’s MSA-QoL score over time, for example, as a change from baseline over time. In some embodiments, stabilizing, limiting, or slowing the clinical progression comprises reducing or stopping the worsening of MSA symptoms, as determined by those skilled in the art.

[0081] The disclosure further provides methods of improving progress-free survival in a population of humans with MSA. The methods include orally administering, on a daily basis, 300 mg emrusolmin to the human. The term “progress-free survival” refers to the time fromrandom assignment in a clinical trial to disease progression or death from any cause. In some embodiments, the improvement in progress-free survival is relative to a population of humans with MSA that is not receiving treatment with emrusolmin.

[0082] The disclosure also provides methods of improving overall survival in a population of humans with MSA. The methods include orally administering, on a daily basis, 300 mg emrusolmin to the human. In some embodiments, the improvement in overall survival is relative to a population of humans with MSA that is not receiving treatment with emrusolmin.

[0083] The disclosure further provides methods of treating a human having MSA and in need of treatment thereof. The methods include a treatment period comprising orally administering to the human, once daily, for a period of treatment, about 300 mg of emrusolmin. In some embodiments, the administration results in a reduced rate of progression of the human’s MSA symptoms during the treatment period. In other embodiments, the reduced rate of progression is as compared to a placebo treatment. In further embodiments, the rate of progression of MSA symptoms is measured using a UMSARS score, or a modification thereof; total UMSARS score, or a modification thereof; CGI-S; and / or MSA-QoL. In yet other embodiments, the rate of progression of MSA symptoms is measured using a UMSARS score, or a modification thereof. In still further embodiments, the rate of progression of MSA symptoms is measured using a total UMSARS score, or a modification thereof. In other embodiments, the rate of progression of MSA symptoms is measured using CGI-S. In further embodiments, the rate of progression of MSA symptoms is measured using MSA-QoL.

[0084] As used herein, the term “reduced rate of progression” refers to reducing the deterioration experienced by a MSA patient or population of MSA patients, e.g., as quantified by UMSARS score, as compared to the deterioration experienced by a MSA patient or population of MSA patients not receiving emrusolmin, over a period of time. For example, the progression may be assessed by an average less worsening of the UMSARS at week 48 compared to baseline, e.g., comparing the UMSARS scores of treated and non-treated groups.

[0085] In certain aspects, emrusolmin reduces the clinical progression by at least about 20% relative to a treatment for MSA lacking emrusolmin. In some embodiments, emrusolmin reduces the clinical progression by at least about 30, about 40, about 50, about 60, about 70, about 80, or about 90% relative to a treatment for MSA lacking emrusolmin. In other embodiments, emrusolmin reduces the clinical progression by about 20 to about 90, about 20 to about 80, about 20 to about 70, about 20 to about 60, about 20 to about 50, about 20 to about 40, about 20 to about 30, about 30 to about 90, about 30 to about 80, about 30 to about 70,about 30 to about 60, about 30 to about 50, about 30 to about 40, about 40 to about 90, about 40 to about 80, about 40 to about 70, about 40 to about 60, about 40 to about 50, about 50 to about 90, about 50 to about 80, about 50 to about 70, about 50 to about 60, about 60 to about 90, about 60 to about 80, about 60 to about 70, about 70 to about 90, about 70 to about 80, or about 80 to about 90% relative to a treatment for MSA lacking emrusolmin. In yet other embodiments, emrusolmin reduces the clinical progression by about 40 to about 60% relative to a treatment for MSA lacking emrusolmin. In still further embodiments, emrusolmin reduces the clinical progression by about 30 to about 40% relative to a treatment for MSA lacking emrusolmin.

[0086] In some aspects of the disclosure, treatment with emrusolmin reduces clinical progression of MSA, as measured by less worsening of the UMSARS scores, or a modification thereof, for example, modified UMSARS part I score, by 2 points (or more), from baseline, as compared to non-treatment with emrusolmin. In some aspects, there is a -2 point from baseline difference in UMSARS scores between those treated with emrusolmin and those not treated with emrusolmin. In some aspects of the disclosure, treatment with emrusolmin reduces the clinical progression of MSA, as measured by less worsening of the UMSARS scores, or a modification thereof, for example, modified UMSARS part I score, by 2 points (or more), from baseline to week 48 of the emrusolmin administration, as compared to non-treatment with emrusolmin. A more detailed description of UMSARS and related treatments is provided herein, for example, with reference to the example study described in reference to FIG. 5.

[0087] In one embodiment, the machine learning model is utilized to determine progression of MSA in a patient with an MSA diagnosis. In another embodiment, the machine learning model can be trained using sensor data from patients with MSA diagnosis. An MSA diagnosis can be categorized as clinically definite MSA, clinically probably MSA, and clinically possible MSA. A diagnosis of clinically definite MSA requires neuropathologic demonstration of CNS a-synuclein-positive glial cytoplasmic inclusions with neurodegenerative changes in striatonigral or olivopontocerebellar structures. A diagnosis of clinically probable MSA requires a sporadic, progressive adult-onset disorder including rigorously defined autonomic failure and poorly levodopa-responsive parkinsonism or cerebellar ataxia. A diagnosis of clinically possible MSA requires a sporadic, progressive adult-onset disease including parkinsonism or cerebellar ataxia and at least one feature suggesting autonomic dysfunction plus one other feature that may be a clinical or a neuroimaging abnormality. Further information on the different categories of MSA diagnosis can be found in Gilman et al. (2008).Second consensus statement on the diagnosis of multiple system atrophy. Neurology, 71(9), 670-676, the entirety of which is incorporated by reference herein.

[0088] As noted above, the system 100 may include one or more remote client devices 130a- 130c. The client devices 130a-130c may comprise, inter alia, a display device. In some examples, the computing device 110 may be configured to generate the notification via the display device of a client device 130a- 130c. The client device 130a may comprise a webportal that, for example, may be presented via a server or other computer. The client device 130b may be a personal computer that, for example, may be associated with a physician or other health care provider, a patient or person related to a patient, pharmaceutical company, etc. The client device 130c may be a mobile device, such as a smartphone, laptop, or tablet, that may be associated with a physician or other health care provider, a patient or person related to a patient, pharmaceutical company, etc.

[0089] FIG. 2 is a simplified block diagram of an example device 200. The device 200 may be an example of a computing device (e.g., one or more servers), such as the computing device 110 of the system 100 of FIG. 1. Alternatively or additionally, the device 200 may be an example of a client device, such as the client devices 130a-130c of the system 100 of FIG. 1. In such instances, the device 200 may include a personal computer, such as a laptop or desktop computer, a tablet device, a cellular phone or smartphone, a server, or another type of client device. The device 200 may be configured to develop, train, or use one or more machine learning models described herein. As shown by FIG. 2, the device 200 may comprise a processor 202, a memory 204, a communication device 206, a display 208, one or more input devices 210, and / or one or more output devices 212. It should be appreciated that the device 200 may include fewer or more components than those shown in FIG. 2.

[0090] The processor 202 may include one or more general purpose processors, special purpose processors, conventional processors, digital signal processors (DSPs), microprocessors, integrated circuits, a programmable logic device (PLD), application specific integrated circuits (ASICs), or the like. The processor 202 may perform signal coding, data processing, image processing, power control, input / output processing, and / or any other functionality that enables the device 200 to perform as described herein.

[0091] The processor 202 may store information in and / or retrieve information from the memory 204. The memory 204 may include a non-removable memory and / or a removable memory. The non-removable memory may include random-access memory (RAM), read-only memory (ROM), a hard disk, or any other type of non-removable memory storage. Theremovable memory may include a subscriber identity module (SIM) card, a memory stick, a memory card, or any other type of removable memory. The memory may be local memory or remote memory external to the device 200. The memory 204 may store instructions which are executable by the processor 202. Different information may be stored in different locations in the memory 204.

[0092] The memory 204 may comprise a computer-readable storage media or machine- readable storage media that maintains computer-executable instructions for performing one or more as described herein. For example, the memory 204 may comprise computer-executable instructions or machine -readable instructions that include one or more portions of the procedures described herein. The processor 202 of the device 200 may access the instructions from memory for being executed to cause the processor 202 of the device 200 to operate as described herein. The memory 204 may comprise computer-executable instructions for executing configuration software. For example, the computer-executable instructions may be executed to perform, in part and / or in their entirety, one or more procedures as described herein. Further, the memory 204 may have stored thereon one or more settings and / or control parameters associated with the device 200.

[0093] The processor 202 may communicate with other devices via the communication device 206. The communication device 206 may transmit and / or receive information over a network (e.g., the network 140), which may include one or more other devices. The communication device 206 may perform wireless and / or wired communications. The communication device 206 may include a receiver, transmitter, transceiver, or other device capable of performing wireless communications via an antenna. The communication device 206 may be capable of communicating via one or more protocols, such as a cellular communication protocol, a Wi-Fi communication protocol, Bluetooth®, a near field communication (NFC) protocol, an internet protocol, another proprietary protocol, or any other radio frequency (RF) or communications protocol. The device 200 may include one or more communication devices 206.

[0094] The processor 202 may be in communication with a display 208 for providing information to a patient. The information may be provided via a patient interface on the display 208. The information may be provided as an image generated on the display 208. The display 208 and the processor 202 may be in two-way communication, as the display 208 may include a touch-screen device capable of receiving information from a patient and providing such information to the processor 202. The processor 202 may be configured to generate, onthe display 208, an indication of one or more notifications described herein, such as an indication of the progression of MSA of the patient, etc.

[0095] The processor 202 may be in communication with input devices 210 and / or output devices 212. The input devices 210 may include a camera, a microphone, a keyboard or other buttons or keys, a mouse, and / or other types of input devices for sending information to the processor 202. The display 208 may be a type of input device, as the display 208 may include touch-screen sensor capable of sending information to the processor 202. The output devices 212 may include speakers, indicator lights, or other output devices capable of receiving signals from the processor 202 and providing output from the device 200. The display 208 may be a type of output device, as the display 208 may provide images or other visual display of information received from the processor 202.

[0096] Although not illustrated, the device 200 may include a power supply. In some examples, the power supply may include one or more batteries. In some examples, the power supply may include an AC to DC power converter. The power supply may be configured to power the processor 202 and the other low voltage circuitry of the device 200.

[0097] Although not illustrated, the device 200 may include a GPS circuit (e.g., in instances where the device 200 is a client device). The processor 202 may be in communication with the GPS circuit for receiving geospatial information. The processor 202 may be capable of determining the GPS coordinates of the device 200 based on the geospatial information received from the GPS circuit. The geospatial information may be communicated to one or more other communication devices to identify the location of the device 200.

[0098] FIG. 3 is a flowchart of a training procedure 300 performed by a device, such as the computing device 110 of the system 100 of FIG. 1. A processor of the computing device may perform the training procedure 300 to train a machine learning model to determine progression of MSA of a patient over a period of time. The processor may be configured to perform the procedure 300 periodically, for example, in response to alarm or schedule that may be configurable. Additionally or alternatively, the processor may be configured to perform the procedure 300 on command. The procedure 300 may start at 310.

[0099] At 312, the processor may receive patient data for a plurality of patients. In one embodiment, the patients for which data is utilized to train the machine learning model are those with clinically confirmed MSA diagnoses. In some embodiments, the data utilized to train the machine learning model are those with clinically probable and / or clinically possible MSA diagnoses. The patients may or may not, individually, or collectively, be prescribed achemical or biologic treatment for MSA symptom mitigation. The patients may be diagnosed with MSA-C or MSA-P. The processor may receive the patient data from one or more sensor devices (e.g., such as the sensor devices 120a- 120n). The plurality of patients may include at least 100 patients. In one example, the plurality of patients may include 160 patients. The patient data may include in-clinic sensor data and at-home sensor data associated with each of the plurality of patients collected over a period of time. The in-clinic sensor data may be acquired via an in-clinic sensor device during structured tasks in a clinical environment as directed by a clinician, whereas the at-home sensor data may be acquired by an at-home sensor device that is the same or different than the in-clinic sensor device during the routine course of the patient’s life.

[0100] The in-clinic sensor data may include one or more of gait speed, stride length, circumduction, dorsiflexion, foot clearance, path deviation, number of steps, step time, mean cadence, a confidential interval of cadence, stride time variability, sit-to-stand duration, sit-to- stand lean angle, stand-to-sit duration, stand-to-sit lean angle, first step duration, first step range of motion, lateral anticipatory postural adjustment (APA) peak, turning speed, turning duration, turning angle, number of steps in turn arm swing variability, arm swing range of motion, arm swing velocity, arm swing asymmetry, arm swing variability, trunk and lumbar coronal, trunk and lumbar sagittal, trunk and lumbar transverse, trunk variability, sway area, path length, jerk, and / or any of the other in-clinic sensor data described in Table 1. The inclinic sensor data may be collected during structured tasks performed by a patient. For example, the structured tasks may include timed stationary standing, timed walk, and / or timed up and go (TUG). The in-clinic sensor data may be collected using a sensor device system. For example, the sensor device system may include one or more wrist-worn sensors, one or more ankle-worn sensors, a sensor platform, a waist-worn sensor, and / or a chest-worn sensor.

[0101] The at-home sensor data may be collected in response to unprompted movements and activities performed by the respective patients. For example, the unprompted movements and activities may include walking, standing, sitting, running, swimming, sleeping, and the like. The at-home sensor data may include one or more of number of steps, step time, mean cadence, a confidential interval of cadence, stride time variability, sit-to-stand duration, active time, inactive time, walking duration, stand duration, sitting / lying duration, total sleep duration, sleep efficiency, wake after sleep onset count, wake after sleep onset duration, and / or wake after sleep onset median. The at-home sensor data may be collected using different sensor devices than the sensor devices used in collecting in-clinic sensor data. For example,the at-home sensor data may be collected using a wrist-worn sensor device.

[0102] In one example, the patient data may also include survey data, which is collected in response to a qualitative survey and / or a quantitative survey. The survey may be used to identify symptoms meaningful to the patient and / or connections between MSA-related activity features and the in-clinic sensor data and / or at-home sensor data. The MSA-related activity features may include one or more of gait features, sit-to-stand features, sleep features, walking initiation features, turning features, arm movement features, trunk and low back range of motion features, and / or postural sway features. The survey may provide a series of questions related to a meaningfulness of each of the MSA-related activity features. For example, the processor may map the MSA-related activity features to one or more meaningful symptoms. The processor may determine a subset of the MSA-related activity features based on the survey data; and may train the machine learning model using the subset of the MSA-related activity features.

[0103] The patient data may span (e.g., have been collected over) a certain period of time, for example, a period of 48 weeks. During the period of time, the in-clinic sensor data may be collected at a given frequency for each of the plurality of patients. For example, the given frequency may be every 6 weeks. However, the at-home sensor data may be collected continuously for each of the plurality of patients over the period of time.

[0104] The patient data utilized to train the machine learning model according to method 300 may include clinician and / or patient reported assessments of MSA progression for each patient. For example, the assessments of MSA progression may performed according to the UMSARS and / or MSA-QoL. The data associated with MSA progression of each patient may represent multiple benchmarks across the period of time — for example, the beginning of the period of time, the end of the period of time, and / or at predetermined intervals during the period of time. The training data representing MSA progression may comprise a UMSARS Part I score, modified UMSARS Part I score, UMSARS Part II score, UMSARS Part III values, and / or and UMSARS Part IV score. MSA progression data can be utilized in training the machine learning model as a “label” to provide context to the sensor data processed during the training step of process 300, as described further below.

[0105] At 314, the processor may determine respective values indicative of a plurality of different MSA-related activity features for the patients based on the in-clinic sensor data and the at-home sensor data. As described above, the activity features may be determined by the computing device 110 from the at-home sensor data and / or the in-clinic sensor data. In anotherembodiment, aspects of the at-home sensor data and the in-clinic sensor data outputted from the at-home sensor and / or the in-clinic sensor may be formatted as activity features ( / '.<?., the processing of the raw sensor data into the activity features was performed at the sensor device level).

[0106] At 316, the processor may train a machine learning model to determine progression of MSA over the period of time for the patients using the respective values for the plurality of different MSA-related activity features for each of the plurality of the patients. The machine learning model trained according to step 316 may employ any combination of an instancebased algorithm (e.g., k-Nearest Neighbor (kNN), Learning Vector Quantization (LVQ), SelfOrganizing Map (SOM), Locally Weighted Learning (LWL), Support Vector Machines (SVM), XGBoost, Random Forest (RF), etc.), a regression algorithm (e.g., a linear regression algorithm, a logistic regression algorithm, etc.), a decision tree algorithm, a Bayesian algorithm (e.g., a Naive Bayes classifier), an ensemble algorithm (e.g., a weighted average algorithm), etc.

[0107] In addition to the sensor data, the training data may include covariates associated with each of the plurality of patients. The covariates associated with each of the plurality of patients may include, for example, clinical subtype of the patient (e.g., MSA-P or MSA-C), age of the patient, gender of the patient, and / or neurofilament light chain levels of the patient.

[0108] The processor may train the machine learning model by segmenting the plurality of patients into a plurality of groups. In the plurality of groups, each group may include the same number of patients or different numbers of patients. The processor may train the machine learning model using the respective values for the plurality of different MSA-related activity features for the patients of one or more (e.g., all) of the plurality of groups. The processor may test the machine learning model using the respective values for the plurality of different MSA- related activity features for the patients of a remainder of the plurality of groups.

[0109] Once the training of the machine learning model is final in step 316, the computing device may apply the model to the test data in step 318 to examine model performance (reproducibility, error, etc.), including sums of squares error, and / or "random seed" reproducibility (e.g, the ability of model to produce the same answer). The computing device may examine convergent validity by comparison to QoL measures. The QoL measures may include one or more of the MSA QoL total and sub-scores, UMSARS total and sub-scores, and / or clinical global impression scale-severity and clinical global impression scale-improvement CGI-I.

[0110] After training the machine learning model, the computing device may exit the procedure 300 at 320.

[0111] FIG. 4 is a flowchart of an example procedure 400 performed by a device, such as the computing device 110 of the system 100 of FIG. 1. A processor of the computing device 110 may perform the procedure 400 to use a machine learning model to determine progression of MSA of one or more patients. The machine learning model may be trained according to procedure 300, as described above. The processor may be configured to perform the procedure 400 periodically, for example, in response to alarm or schedule that may be configurable. Additionally or alternatively, the processor may be configured to perform the procedure 400 on command. The procedure 400 may start at 410.

[0112] At 411, a patient may be initiated in a MSA treatment regimen comprising a medication (e.g., a chemical or biologic) for the treatment of MSA. The patient may have a diagnosis of clinically definite MSA, clinically probably MSA, or clinically possible MSA. In addition to standardizing the assessment of MSA progression in patients, the machine learning model of the present description can be utilized to assess efficacy of new or existing treatments for MSA, such as during clinical trials. By employing an objective and standardized tool for the assessment of MSA progression, efficacy of prospective medications for the treatment of MSA or symptoms thereof can be more readily assessed. Examples of such medications include emrusolmin, droxidopa, levodopa, dopamine agonists such as ropinirole and pramipexole, clonazepam, vitamin E, propranolol, baclofen, amantadine, buspirone, ampreloxetine, ENT-01, Lu AF82422, AAV2-GDNF, ION464 and ATH434. However, it is also contemplated that the machine learning model can be utilized to assess MSA progression in patients not taking a medication for the treatment of MSA, such as placebo patients in a clinical trial.

[0113] At 412, the processor may receive patient data for one or more patients. For example, the processor may receive the patient data from one or more sensor devices (e.g., such as the in-clinic and at-home sensor devices 120a-120n). The patient data may include in-clinic sensor data and at-home sensor data. In one example, the patient data may also include survey data, which is collected in response to a qualitative survey and a quantitative survey.

[0114] The survey may be used to identify symptoms meaningful to the patient and / or connections between MSA-related activity features and the in-clinic sensor data or at-home sensor data. The MSA-related activity features may include one or more of gait features, sit- to-stand features, sleep features, walking initiation features, turning features, arm movementfeatures, trunk and low back range of motion features, and / or postural sway features. The survey may provide a series of questions related to a meaningfulness of each of the MSA- related activity features. For example, the processor may map the MSA-related activity features to one or more meaningful symptoms. The processor may determine a subset of the MSA-related activity features based on the survey data; and train the machine learning model using the subset of the MSA-related activity features.

[0115] The patient data may span (e.g., have been collected over) a certain period of time, for example, a period of 48 weeks. During the period of time, the in-clinic sensor data may be collected at a given frequency for each of the plurality of patients. For example, the given frequency may be every 6 weeks. The at-home sensor data may be collected continuously for each of the plurality of patients over the period of time.

[0116] The in-clinic sensor data may include one or more of gait speed, stride length, circumduction, dorsiflexion, foot clearance, path deviation, number of steps, step time, mean cadence, a confidential interval of cadence, stride time variability, sit-to-stand duration, sit-to- stand lean angle, stand-to-sit duration, stand-to-sit lean angle, first step duration, first step range of motion, lateral anticipatory postural adjustment (APA) peak, turning speed, turning duration, turning angle, number of steps in turn arm swing variability, arm swing range of motion, arm swing velocity, arm swing asymmetry, arm swing variability, trunk and lumbar coronal, trunk and lumbar sagittal, trunk and lumbar transverse, trunk variability, sway area, path length, jerk, and / or any of the in-clinic sensor data described in Table 1. The in-clinic sensor data may be collected during structured tasks performed by a patient. For example, the structured tasks may include timed stationary standing, timed walk, and / or timed up and go (TUG). The in-clinic sensor data may be collected using a sensor device system. For example, the sensor device system may include one or more wrist-worn sensor devices, one or more ankle-worn sensor devices, a sensor platform device, a waist-worn sensor device, and / or a chest- worn sensor device.

[0117] The at-home sensor data may be collected in response to unprompted movements and activities performed by the respective patients. For example, the unprompted movements and activities may include walking, standing, sitting, running, swimming, and the like. The at- home sensor data may include one or more of number of steps, step time, mean cadence, a confidential interval of cadence, stride time variability, sit-to-stand duration, active time, inactive time, walking duration, stand duration, sitting / lying duration, total sleep duration, sleep efficiency, wake after sleep onset count, wake after sleep onset duration, and / or wakeafter sleep onset median. The at-home sensor data may be collected using different sensor devices than the sensor devices used in collecting in-clinic sensor data. For example, the at- home sensor data may be collected using a wrist-worn sensor device (e.g., Fitbit or Apple Watch).

[0118] An example of in-clinic and at-home sensor data that can be to train the machine learning model (e.g., a model that is configured to determine progression of MSA of one or more patients) may be described in more detail below with respect to an example of a clinical trial. For instance, the in-clinic and at-home sensor data used to train the machine learning model may be obtained using the client trial described below.

[0119] At 414, the processor may determine respective values indicative of a plurality of different MSA-related activity features for the one or more patients based on the in-clinic sensor data and the at-home sensor data. The determined respective values may include numbers, and / or pass or fail.

[0120] At 416, the processor may use a trained machine learning model to determine progression of MSA over the period of time for the one or more patients using the respective values for the sensor data and / or plurality of different MSA-related activity features for each of the plurality of the patients. The processor may also utilize covariates associated with the one or more patients to determine progression of MSA, where such covariates may include, for example, clinical subtype of the patient e.g., MSA-P or MSA-C), age of the patient, gender of the patient, and / or neurofilament light chain levels of the patient. The trained machine learning model determination representing progression of MSA over the period of time may comprise a change in a UMSARS score over the period of time. For example, progression of MSA over the period of time as determined by the trained machine learning model may comprise a change in UMSARS Part I score, modified UMSARS Part I score, UMSARS Part II scorePart III, or UMSARS Part IV score. Alternatively, progression of MSA over the period of time as determined by the trained machine learning model may comprise a change in combined UMSARS Part I and Part II scores or combined modified UMSARS Part I and Part II scores.

[0121] The computing device may exit the procedure 400 at 418.

[0122] Below is a description of an example client trial, or study, during which in-clinic and / or at-home sensor data can be obtained, where this in-client and / or at-home sensor data can be used to train a machine learning model that is configured to determine progression of MSA of one or more patients. A more detailed description of in-clinic and / or at-home sensors and sensor data is described above.

[0123] Human Populations. According to the disclosure, the stabilizing, limiting, or slowing is relative to a patient population with MSA that is not receiving treatment with emrusolmin.

[0124] The methods described herein are not only applicable to an individual human, but also are applicable to a population of humans. The term “population of humans” refers to a number of humans that can provide statistically relevant data that provides insight regarding the effectiveness of emrusolmin as applied to the general population. For example, a population of humans may include about 3 people, about 5 people, about 10 people, about 15 people, about 20 people, or more.

[0125] The disclosure, thus, provides methods of stabilizing the clinical progression, limiting the clinical progression, and / or slowing the clinical progression of MSA in a population of humans in need thereof. The methods include orally administering, on a daily basis, of 300 mg emrusolmin to each human in the population. The clinical progression is determined as a statistical change in the human populations’ UMSARS scores, or a modification thereof, for example, a statistical change from baseline in the human populations’ UMSARS scores, or a modification thereof. The change from baseline in the clinical progression is statistically significant. In some embodiments, the clinical progression is statistically significant at a 1- sided alpha of about 0.05, as compared to a control population of humans. In some embodiments, the clinical progression is statistically significant at a 1 -sided alpha of about 0.1, as compared to a control population of humans.

[0126] The disclosure also provides methods of stabilizing the clinical progression, limiting the clinical progression, and / or slowing the clinical progression of MSA in a population of humans in need thereof. The methods include orally administering, on a daily basis, of 300 mg emrusolmin to each human in the population. In some embodiments, clinical progression is determined as a statistical change in the human population’s CGI-S score, for example, a statistical change from baseline in the human population’s CGI-s score. In other embodiments, the change in the clinical progression is statistically significant. In further embodiments, the change from baseline in the clinical progression is statistically significant at a 1 -sided alpha of about 0.05, as compared to a control population of humans. In further embodiments, the change from baseline in the clinical progression is statistically significant at a 1 -sided alpha of about 0.1, as compared to a control population of humans.

[0127] The disclosure also provides methods of stabilizing the clinical progression, limiting the clinical progression, and / or slowing the clinical progression of MSA in a population of humans in need thereof. The methods include orally administering, on a daily basis, of 300 mgemrusolmin to each human in the population. In some embodiments, the clinical progression is determined as a statistical change in the human population’s MSA-QoL score, for example, a statistical change from baseline in the human population’s MSA-QoL score. In other embodiments, the change in the clinical progression is statistically significant, preferably statistically significant at a 1 -sided alpha of about 0.05, as compared to a control population of humans. In further embodiments, the change from baseline in the clinical progression is statistically significant at a 1 -sided alpha of about 0.1, as compared to a control population of humans.

[0128] Desirably, administration of emrusolmin in any other methods of the disclosure does not result in a clinically serious adverse event in the human or the population of humans. In some embodiments, presentation of serious adverse events is not statistically significant as compared to a control human or a control population of humans.

[0129] The human may have received at least one therapy for the treatment of MSA prior to administration of emrusolmin. In some embodiments, the at least one therapy for MSA is levodopa. In other embodiments, the human is resistant to treatment with levodopa. In further embodiments, the human has acquired resistance to treatment with levodopa.

[0130] In some aspects, certain clinical assessments are utilized in performing the methods disclosure herein. Examples of such assessments include, without limitation, rating scales such as the Unified Multiple System Atrophy Rating Scale (UMSARS) score, a modified UMSARS score, a modified UMSARS part I (excluding question 11, item scoring rescaled 0-3), a total UMSARS score (part I and part II combined), a modified total UMSARS score (part I and part II combined), a CGI-S score, or a MSA-QoL score.

[0131] In some embodiments, the clinical assessment is the UMSARS. See, e.g., Wenning, “Development and validation of the Unified Multiple System Atrophy Rating Scale (UMSARS),” Mov. Disord. 2004; 19(12): 1391-402, the entirety of which is incorporated by reference herein. The UMSARS is a multidimensional, validated scale for semi-quantitative clinical assessments of MSA patients. It is comprised of 4 parts: part I, historical review of disease-related impairments, 12 items; part II, motor examination, 14 items; part III, autonomic examination; and part IV, global disability scale. As UMSARS is a unified scale, each item in parts I and II achieves a single score using a range of 0 (no impairment) to 4 (severe impairment). To obtain the UMSARS part I score and UMSARS part II score, respectively, the values for the individual items are summed up with higher scores corresponding to a more pronounced clinical disease severity. A total UMSARS score includes the sums for parts I andII.

[0132] In other embodiments, the clinical assessment is a modified UMSARS, i.e., Unified Multiple System Atrophy Rating Scale - Modified Score (UMSARS-MS). The UMSARS-MS uses all data collected within part I and part II of the UMSARS. To obtain the classical total UMSARS score, values of 0, 1, 2, 3, or 4 (depending on the level of impairment) are assigned to each item of the UMSARS part I and part II and summed up, i.e., a modified total UMSARS score. For the UMSARS-MS, those values are transformed to reweighted and rescaled values before they are summed up. As the UMSARS-MS is a modified evaluation using different values assigned to the different levels of impairment within the items assessed, the source data for calculation of the score is identical to the data used for the calculation of the classical total UMSARS score. The range of scores (minimum to maximum) is preserved in UMSARS-MS compared to the classical UMSARS. In certain aspects, in a modified UMSARS part, I score, item 11 (sexual function) is omitted and the item scoring rescaled 0 to 0, 1 to 0, 2 to 1, 3 to 2, and 4 to 3.

[0133] In further embodiments, the clinical assessment is the Multiple System Atrophy - Quality of Life (MSA-QoL). The MSA-QoL is a patient-reported validated measure correlated with quality of life in a MSA population. The MSA-QoL shows internal consistency and construct validity (Schrag, “Measuring health-related quality of life in MSA: the MSA-QoL,” Mov. Disord. 2007; 22(16):2332-8). It was shown that MSA-QoL correlates negatively with disease progression in MSA (Schrag, “Health related quality of life in multiple system atrophy,” Mov. Disord. 2006; 21(6):809-15; Schrag 2007). The 40-item MSA-QoL questionnaire is self-administered, and each item is rated on a Likert type scale (0: No problem) to (4: Extreme problem). It is comprised of 3 subscales relevant to MSA: motor (14 items), non-motor (12 items), and emotional / social (14 items). The MSA-QoL total score is the sum of all the items and lower scores indicate better QoL.

[0134] In yet other embodiments, the clinical assessment is the Clinical Global Impression Scale - Severity (CGLS). The CGLS scale permits a global evaluation of the patient’s current severity of illness on a Likert type scale ranging from 1 to 7, where l=normal / not at all ill, 2=borderline mentally ill, 3=mildly ill, 4=moderately ill, 5=markedly ill, 6=severely ill, and 7=among the most extremely ill patients (Guy 1976).

[0135] The methods may further include administering an additional therapeutic agent. As used herein, an additional therapeutic agent refers to an agent other than emrusolmin. In some embodiments, the additional therapeutic agent treats an aspect of the patient’s MSA symptom,such as, for example, hypokinetic -rigid symptoms, ataxia, orthostatic dizziness and lightheadedness. Exemplary additional therapeutic agents include, e.g., levodopa, droxidopa, dopamine agonists, N methyl-D-aspartate receptor antagonists, botulinum toxin, antimuscarinic agents or sympathomimetic agents. The additional therapeutic agent may be administering concomitant with, prior to, or following administration of emrusolmin.

[0136] Example 1

[0137] A. Study Objectives, Endpoints, and Estimands

[0138] (i) Primary and Secondary Study Objectives and Endpoints

[0139] (ii) Primary Estimand

[0140] For the primary efficacy endpoint, the following estimand attributes will be employed:• Treatment: TEV-56286 300 mg QD or matching placebo for approximately 48 weeks.• Population: All randomized adult patients with clinically possible or probable MSA.• Endpoint: Change from baseline to week 48 in modified UMSARS part I score.• Population-level summary: Difference between treatment groups in mean change from baseline to 48 weeks in modified UMSARS part I score.

[0141] The treatment policy strategy will be applied for the intercurrent events of early study discontinuation and concomitant medication / treatment noncompliance, meaning that the patient’s data will be used in the analysis regardless of these intercurrent events.

[0142] (iii) Exploratory Study Objectives and Endpoints

[0143] B. Study Design

[0144] (i) General Study Design and Study Schematic Diagram

[0145] This is a Phase 2, 56-week, multi-centered, randomized, double blind, 2 arm, placebo controlled study to evaluate the safety and efficacy of TEV-56286, also known as ANLE138b, in adult patients with MSA.

[0146] The study will consist of a screening period (up to 4 weeks), a 48-week double-blind treatment period, and a follow-up visit (approximately 4 weeks after the end of the double-blind treatment period). Patients who qualify for the study based on the inclusion / exclusion criteria will be randomized in a 1:1 ratio (stratified by region and by typeof MSA [parkinsonism-predominant versus cerebellar- ataxia-predominant]) to receive either 300 mg test Investigational Medical Product (IMP) or placebo.

[0147] Throughout the study, patients will interact regularly with investigational center personnel, during in clinic visits and by telephone for the evaluation of safety and tolerability.

[0148] The study schematic diagram is shown in FIG. 5.

[0149] The total duration of patient participation in the study is planned to be approximately 56 weeks, of which approximately 48 weeks are double -blind treatment with IMP.

[0150] At the baseline visit, patients will be randomly assigned to 1 of 2 treatment groups with TEV-56286 or placebo in a 1:1 ratio. There is no reference IMP in this study. For each group, 10 x 30 mg capsules or placebo capsules will be administered QD.

[0151] Patients who complete all scheduled visits will have procedures and assessments performed at week 52 end of study (EOS). Patients who withdraw from the study before completing the 52-week period will have visit 9 [EOT] early termination procedures and assessments performed at their final visit. Patients who complete this study may be eligible to enroll in an open-label extension study to assess the continued effectiveness and long-term safety of TEV-56286, if and when such an option becomes available. This potential open-label extension study is beyond the scope of this protocol. A separate protocol will be issued for the open-label extension study as applicable.

[0152] The end of study is defined as the last follow-up visit of the final patient.

[0153] The study duration will be approximately 27 months.

[0154] The study used a combination of in-clinic sensors and at-home sensors. The in-clinic sensors include a combination of six sensors, and the patients were asked to perform three different tasks. After the first in-clinic visit, the patient was provided with an at-home sensor. The at-home sensor is configured to monitor the patient during all activities while they are at- home. For example, the at-home sensor may include an accelerometer. The patient is instructed to not remove the at-home sensor. The at-home sensor may be in the form of a wrist-watch. At subsequent in-client visits, the at-home sensor is obtained from the user so that the data can be extracted, and a new at-home sensor is provided to the patient for the next at- home interval (e.g., 30 days). Although described in reference to this specific study, the method and machine learning algorithms described herein may be trained and used using any combination of the in-clinic sensors and / or at-home sensors described herein.

[0155] (ii) Planned Number of Patients and Countries

[0156] Approximately 190 patients will be screened to achieve approximately 160 enrolled patients.

[0157] Per the primary estimand, all enrolled patients will be evaluable for safety and efficacy analysis in this study following randomization.

[0158] C. Selection and Withdrawal of Patients

[0159] (ii) Patient Inclusion Criteria

[0160] Patients may be randomized / enrolled in this study only if they meet all of the following criteria: a. is a male or female >30 years old at screening; b. is considered to be “clinically possible” or “clinically probable” MSA as determined by the Gilman criteria (See, Gilman, Second consensus statement on the diagnosis of multiple system atrophy. Neurology 2008; 71(91:670-6); c. is able to ambulate at least 10 meters without the assistance of another person at screening; the use of assistive devices (e.g., walker or cane) is allowed; d. If patient is receiving treatment for symptomatic MSA prior to randomization, therapy regimen must have been stable for a period of >28 days with no foreseeable changes predicted during the study including but not limited to: drugs acting against parkinsonism (e.g., levodopa, dopamine-agonists, amantadine, and monoamine oxidase-B inhibitors); drugs acting against autonomic dysfunction (e.g., ephedrine, midodrine, fludrocortisone, octreotide, desmopressin, and oxybutynine); and / or antidepressant drugs (e.g., selective serotonin reuptake inhibitors, serotonin and norepinephrine reuptake inhibitors). e. is able to swallow the IMP, as determined by a dysphagia evaluation during the screening period and in follow-up evaluations during the course of the study; and is willing to swallow the IMP capsules whole, and as scheduled, throughout the duration of the study; f. is medically and psychiatrically stable, as indicated by medical and psychiatric history, as well as physical and neurological examination; g. In the investigator’s opinion, the patient and / or patient’s caregiver(s) has (have) the ability to understand the nature of the study and its procedures, and is (are) able andwilling to comply with the requirements of the study; h. If patient is personally unable to provide written informed consent, caregiver(s) is (are) able and willing to provide written informed consent after being assigned the role by the patient; i. Females of child bearing potential (CBP) may be included only if they have a negative pregnancy test at the screening and baseline visits; j. Females of CBP whose male partners are potentially fertile ( / '.<?., no vasectomy) must use highly effective birth control methods beginning 28 days before the first administration of IMP, the duration of the study, and 28 days after the last dose of IMP; and k. Males who are potentially fertile / reproductively competent (not surgically [e.g., vasectomy] or congenitally sterile) and their female partners who are of CBP must use, together with their female partners, highly effective birth control methods for the duration of the study and for 28 days after the last dose of IMP.

[0161] (ii) Patient Exclusion Criteria

[0162] Patients will not be randomized / enrolled in this study if they meet any of the following criteria: a. has any clinically significant uncontrolled medical or psychiatric condition (treated or untreated), other than MSA that could jeopardize or would compromise the patient’s ability to participate in any study assessments; b. is severely affected with a UMSARS part IV score of 5; c. is suspected of having a neurodegenerative disease other than MSA; d. has 2 or more relatives with history of MSA, suggestive of an alternative diagnosis other than MSA; e. has participated in another clinical study involving administration of an IMP within 3 months or 5 half-lives (whichever is longer) of this IMP prior to screening, or has previously participated in other studies of neuroprotective therapies (e.g., deferiprone, exenatide, or alpha- synuclein antibodies or vaccinations, anti-sense oligonucleotides); f. has a history of an intracranial or intraspinal space occupying lesion, increased intracranial pressure, or any other known condition in the medical history that may preclude the patient from the procedure of cerebrospinal fluid (CSF) sampling; g. has a history of, or acknowledges, alcohol or other substance abuse in the 12 monthsbefore screening; h. is a female patient who is pregnant or breastfeeding, or plans to become pregnant during the study; i. is diagnosed with concurrent severe depression defined by a score >30 on the Beck Depression Inventory-II and / or has a significant risk of committing suicide based on the patient’s medical history / or the C-SSRS (lifetime); patients with a C-SSRS (current) positive response to suicidal ideation items 3, 4, or 5 are not eligible; j. has a known hypersensitivity to any components of the IMP; k. has moderate to severe renal impairment as assessed by estimated glomerular filtration rate <60 mL / min; l. has moderate to severe hepatic impairment as assessed by Child-Pugh Score of 7 or above; m. has any disorder that may interfere with absorption, distribution, metabolism, or excretion of IMPs (including relevant gastrointestinal surgery); n. is of a vulnerable population (e.g., people kept in detention); and / or o. patient is using or consuming any prohibited concomitant medications within the specified exclusionary windows of this study .

[0163] D. Treatments

[0164] (i) Investigational Medicinal Products Used in the Study

[0165] The IMP in this study is TEV-56286. There is a matching placebo composed of excipients only. Neither the IMP nor placebo are authorized and marketed products.

[0166] (a) Dosing

[0167] TEV-56286 or matching placebo will be administered orally at a dose of 300 mg QD (10 x 30 mg) capsules with a sufficient amount of water. Patients will be instructed to selfadminister at approximately the same time each day, irrespective of food intake, when in-clinic administration is not already scheduled. Administration will be performed during the treatment period of 48 weeks.

[0168] (b) Placebo Investigational Medicinal Product

[0169] The placebo IMP consists of excipient ingredients only which shall be administered identically at a dose of 300 mg QD (10 x 30 mg) capsules. Administration will be performed over a 48-week period.

[0170] (c) Other Components

[0171] A descriptive table of TEV-56286 and the placebo is included below:Table 2

[0172] (ii) Prior Prohibited and Concomitant Medication and Therapies

[0173] Any prior or concomitant medication a patient has had before IMP administration and up to the end of study, including follow-up, will be recorded.

[0174] Use of any over-the-counter (OTC) medications, herbal medications, diet aids, or nutritional supplements is prohibited during the study and follow-up period, with the exception of:1. approved forms of contraception2. vitamins within expected ranges of dosing3. approved doses of acetaminophen or ibuprofen4. medications required to treat pre-existing medical conditions

[0175] Medications prescribed for treatment of acute or chronic medical conditions, otherthan excluded conditions, may be permitted. The prescribed medications may be inclusive of medical marijuana, cannabidiol, OTC, vitamins and nutritional supplements as long as the medicine is maintained under the supervision of a prescribing health care provider.Medications required for the treatment of adverse events shall be permitted under supervision.

[0176] Concomitant medication and treatment will be queried and recorded until the completion of study activities or the time of patient withdrawal.

[0177] (a) Prohibited Medications and Therapies

[0178] Strong or moderate inhibitors of CYP1A2: Moderate and strong inhibitors of CYP1A2 are disallowed from the study 14 days or 5 half-lives (whichever is longer) prior to first dose of TEV-56286 and until 4 days after last dose of TEV-56286.

[0179] (iii) Assessments of Efficacy

[0180] (a) Unified Multiple System Atrophy Rating Scale (UMSARS)

[0181] The UMSARS is a multidimensional, validated scale for semi-quantitative clinical assessments of MSA patients (Wenning, Development and validation of the Unified Multiple System Atrophy Rating Scale (UMSARS). Mov. Disord. 2004; 19(12): 1391— 402). It is comprised of 4 parts: part I, historical review of disease-related impairments, 12 items; part II, motor examination, 14 items; part III, autonomic examination; and part IV, global disability scale. As UMSARS is a unified scale, each item in parts I and II achieves a single score using a range of 0 (no impairment) to 4 (severe impairment). To obtain the UMSARS part I score and UMSARS part II score, respectively, the values for the individual items are summed up with higher scores corresponding to a more pronounced clinical disease severity.

[0182] In the derivation of the modified UMSARS part I score, item 11 (sexual function) will be omitted and the item scoring rescaled 0 to 0, 1 to 0, 2 to 1, 3 to 2, and 4 to 3.

[0183] (b) Unified Multiple System Atrophy Rating Scale - Modified Score (UMSARS-MS)

[0184] The UMSARS-MS uses all data collected within part I and part II of the UMSARS. To obtain the classical total UMSARS score, values of 0, 1, 2, 3, or 4 (depending on the level of impairment) are assigned to each item of the UMSARS part I and part II and summed up. For the UMSARS-MS, those values are transformed to reweighted and rescaled values before they are summed up. As the UMSARS-MS is a modified evaluation using different values assigned to the different levels of impairment within the items assessed, the source data for calculation of the score is identical to the data used for the calculation of the classical total UMSARS score. The range of scores (minimum to maximum) is preserved in UMSARS-MScompared to the classical UMSARS.

[0185] In trial simulations based on data from previous MSA cohorts (Dodel, “Minocycline 1-year therapy in multiple-system-atrophy: effect on clinical symptoms and [ C] (R)-PK11195 PET (MEMSA-trial),” Mov, Disord. 2010; 25( :97-107; Levin, “Safety and efficacy of epigallocatechin gallate in multiple system atrophy (PROMESA): a randomised, doubleblind, placebo-controlled trial. Lancet Neurol,” 2019; 18(81:724-35), the UMSARS-MS had slightly higher statistical power to detect a slowing in disease progression.

[0186] (c) Multiple System Atrophy - Quality of Life (MSA-QoL)

[0187] The MSA-QoL is a patient-reported validated measure correlated with quality of life in a MSA population. The MSA-QoL shows internal consistency and construct validity (Schrag 2007 above). It was shown that MSA-QoL correlates negatively with disease progression in MSA (Schrag 2006 above; Schrag 2007 above). The 40-item MSA-QoL questionnaire is self-administered and each item is rated on a Likert type scale (0: No problem) to (4: Extreme problem). It is comprised of 3 subscales relevant to MSA: motor (14 items), non-motor (12 items), and emotional / social (14 items). The MSA-QoL total score is the sum of all the items and lower scores indicate better QoL.

[0188] (d) Clinical Global Impression Scale - Severity (CGLS)

[0189] The CGLS scale permits a global evaluation of the patient’s current severity of illness on a Likert type scale ranging from 1 to 7, where l=normal / not at all ill, 2=borderline mentally ill, 3=mildly ill, 4=moderately ill, 5=markedly ill, 6=severely ill, and 7=among the most extremely ill patients (Guy, Clinical Global Impressions Scale (CGI). APA PsycTests 1976).

[0190] (e) Clinical Global Impression Scale - Improvement (CGLI)

[0191] The CGLI scale permits a global evaluation of the patient’s overall improvement in symptoms. The CGLI scale rates the patient’s improvement relative to their symptoms at baseline on a 7-point Likert type scale ranging from 1 (very much improved) to 7 (very much worse) (Guy 1976 above).

[0192] (f) Digital Biomarker Assessments

[0193] Patient digital biomarker assessment will be assessed during in-clinic visits by wearing a digital gait monitor for the duration of the gait assessment. This procedure may be combined with the timed 10-meter walk assessment. Patient's activity and sleep will be passively monitored out of clinic by actigraphy watch worn for 2 weeks following each in-clinic visit. Analysis of both in-clinic gait and actigraphy will be reported separately.

[0194] (g) Neurofilament Light Chain (NfL)

[0195] Level of NfL has been identified as a reliable biomarker for neuroaxonal damage in CSF and blood samples of neurodegenerative diseases (Hu, Cerebrospinal fluid levels of neurofilament light chain in multiple system atrophy relative to Parkinson’s disease: a metaanalysis. Neurol Sci 2017;38:407-14; Singer, Alpha- sy nuclein oligomers and neurofilament light chain in spinal fluid differentiate multiple system atrophy from Lewy body synucleinopathies. Ann Neurol 2020;88:503-12). NfL assessment is significantly elevated in MSA in many studies by more than the factor of 2 compared to healthy controls (Hans son, Blood-based NfL; A biomarker for differential diagnosis of parkinsonian disorder. Neurology. 2017 Mar 7;88(10):930-937; Zhang, Neurofilament light chain predicts disease severity and progression in multiple system atrophy. Mov. Disord. 2022; 37(29:421 -6) and has been shown to be a suitable parameter for detecting neuroprotective treatment effects in other neurodegenerative diseases such as spinal muscular atrophy, a condition in which response to treatment is predicted by a decrease in NfL (Wurster, Neurofilament light chain in serum of adolescent and adult SMA patients under treatment with nusinersen. J. Neurol. 2020; 267(l):36-44).

[0196] (h) MRI Regional Volumetric Changes in the Brain

[0197] In order to further investigate the correlation between brain volume change and the progression of MSA in TEV-56286 and placebo-treated patients, magnetic resonance imaging (MRI) regional volumetric changes will be assessed.

[0198] E. Assessment of Pharmacokinetics, Pharmacodynamics, Biomarkers and Pharmacogenetic s

[0199] (i) Pharmacokinetics Assessment

[0200] Blood of sparse pharmacokinetics samples will be collected throughout the study for measurement of sample concentration of TEV-56286 (and potential TEV-56286 metabolites).

[0201] When the pharmacokinetics blood sample collection coincides with safety assessments or related unscheduled study visits, pharmacokinetics blood samples should be collected as close to the scheduled timepoint as possible. Assessment of vital signs and ECG should precede pharmacokinetics blood sampling.

[0202] The dates and times of IMP administration and the date and timepoint (24-hour clock time) of each sample collection will be recorded both on the source documentation and the patient diary.

[0203] Samples will be analyzed for concentration of TEV-56286 using an appropriate validated method. Incurred sample reanalysis may be performed. Samples collected from patients who receive placebo will not be analyzed.

[0204] (ii) Pharmacokinetics / Pharmacodynamics Assessment

[0205] Pharmacodynamic parameters evaluated in this study include selected safety and efficacy endpoints. If data permits, the relationship between the plasma IMP concentration and pharmacodynamics endpoints will be investigated. The results of the pharmacokinetic s / pharmacodynamics analysis may be presented in a separate report.

[0206] (iii) Biomarker Assessment

[0207] Biomarkers are defined as biological substances that monitor physiological effects, assessing activity of the medicinal product, predicting clinical outcome, safety, and treatment response.

[0208] Collection of blood samples for biomarker analysis will be a part of this study. This may include misfolded alpha synuclein, NfL in plasma and CSF, glial fibrillary acidic protein concentrations, and disease / pathway implicative proteins. Additional biomarkers may be selected at a later stage before the analysis so as to allow updating with new scientific information.

[0209] Samples may be stored for a maximum of 15 years after the last patient’s last visit.

[0210] Biomarker blood samples will be collected via venipuncture.

[0211] (iv) Pharmacogenomics Assessment

[0212] Blood samples will be stored and analyzed for biomarker / genetic variants thought to play a role in MSA.

[0213] Collection of blood samples for pharmacogenomics analysis will be a part of this study. This shall include genetic polymorphisms (e.g., CYP1A2 genetic variation) to evaluate their association with observed clinical responses to TEV-56286 (including TEV-56286 pharmacokinetics). Additional pharmacogenomics outcomes may be selected at a later stage before the analysis so as to allow updating with new scientific information.

[0214] Samples may be stored for a maximum of 15 years after the last patient’s last visit.

[0215] The pharmacogenomics blood samples will be collected via venipuncture.

[0216] (v) Efficacy Analysis

[0217] (a) Primary Endpoint

[0218] The primary efficacy endpoint will be change from baseline to week 48 (EOT) as measured by modified UMSARS part I score (excluding item 11, item scoring rescaled 0-3).

[0219] (b) Supplementary Endpoints

[0220] As a supplementary analysis of the primary endpoint of modified UMSARS part I score change from baseline, linear slope of progression in modified UMSARS part I score from baseline to week 48, as well as from week 48 to week 52, will be estimated and compared between treatment groups.

[0221] (c) Planned Method of Analysis

[0222] The ITT / mITT / PP Analysis Sets will be used for all efficacy analyses. Summaries will be presented by treatment group.

[0223] (d) Primary Efficacy Analysis

[0224] The mITT Analysis Set will be used for the primary efficacy endpoint analysis.

[0225] The primary analysis will be a mixed model for repeated measures (MMRM) with modified UMSARS part I score change from baseline as the dependent variable; fixed effects for treatment group, week as a categorical variable, and treatment-by-week interaction; and baseline modified UMSARS part I score, and the same baseline stratification factors used for randomization as covariates. The unstructured covariance model will be used for repeated observations within patients. If the model does not converge, then simpler covariance structures with fewer parameters will be used, according to the following order: heterogeneous autoregressive (1), heterogeneous compound symmetry, autoregressive (1), and compound symmetry.

[0226] The primary comparison will be treatment contrast at week 48. The difference in least squares (LS) mean of the change in modified UMSARS part I score from baseline to week 48 (active versus placebo) will be compared using a 1 -sided test for superiority at a nominal significance level of a=0.05.

[0227] The LS mean and standard error for the treatment groups, the LS mean difference, the 2-sided 95% confidence interval, and the p-value for the comparison (active versus placebo) at week 48 will be presented.

[0228] (e) Secondary Efficacy Analysis

[0229] Treatment difference in total UMSARS score (part I and part II combined) change from baseline to week 48 will be analyzed using MMRM with total UMSARS score change from baseline as the dependent variable; fixed effects for treatment group, week as acategorical variable, and treatment-by-week interaction; and baseline total UMSARS score, and the same baseline stratification factors used for randomization as covariates. The unstructured covariance model will be used for repeated observations within patients. If the model does not converge, then simpler covariance structures with fewer parameters will be used as specified for the primary efficacy endpoint analysis.

[0230] Treatment difference in MSA-QoL scale score change from baseline to week 48 will be analyzed using MMRM with MSA-QoL scale score change from baseline as the dependent variable; fixed effects for treatment group, week as a categorical variable, and treatment-by- Ffigurweek interaction; and baseline MSA-QoL scale score, and the same baseline stratification factors used for randomization as covariates. The unstructured covariance model will be used for repeated observations within patients. If the model does not converge, then simpler covariance structures with fewer parameters will be used as specified for the primary efficacy endpoint analysis.

[0231] Treatment difference in the CGLS scale score change from baseline to week 48 will be analyzed using MMRM with CGLS scale score change from baseline as the dependent variable; fixed effects for treatment group, week as a categorical variable, and treatment-by- week interaction; and baseline CGLS scale score and the same baseline stratification factors used for randomization as covariates. The unstructured covariance model will be used for repeated observations within patients. If the model does not converge, then simpler covariance structures with fewer parameters will be used as specified for the primary efficacy endpoint analysis.

[0232] In addition, since the CGLS may not be normally distributed, the Wilcoxon rank sum test will be performed on the change from baseline to week 48 of the CGLS scale score.

[0233] (f) Supplementary Analysis

[0234] For the supplementary analysis of the primary efficacy endpoint of modified UMSARS part I score change from baseline, linear slope as a function of continuous time will be estimated and compared between treatment groups. The statistical model will be a repeated measures mixed linear model with modified UMSARS part I score change from baseline as dependent variable; fixed effects for week as a continuous variable and treatment-by-week interaction; and interaction between week and each of: baseline modified UMSARS part I score and the same baseline stratification factors used for randomization as covariates. The individual patient linear slope over time will also be included in the model as a random effect. The model will fix the y-intercept (change from baseline at week 0) to be 0.

[0235] (vi) Tolerability Analysis

[0236] Tolerability analysis will be based on the number (%) of patients who failed to complete the study and the number (%) of patients who failed to complete the study due to adverse events. Time to study withdrawal will be presented by Kaplan-Meier curves for each treatment group.

[0237] (vii) Pharmacokinetics Analysis

[0238] Observed plasma and CSF concentrations will be summarized using descriptive statistics (n, mean, SD, median, minimum, geometric mean, and coefficient of variation). Correlation of observed CSF to plasma concentrations of TEV-56286 (and metabolites, if applicable) may be conducted, as data permits.

[0239] The following observed pharmacokinetics parameters for TEV-56286 (and potential metabolites) will be determined:• Concentrations of IMP reported by dose, visit, and time.• Minimum concentration (Cmin) prior to dose administration for TEV-56286 reported by visit.

[0240] Pharmacokinetics parameters will be summarized using descriptive statistics, including the geometric mean and coefficient of variation, as appropriate. If available, a separate summary may also include pharmacokinetics parameters from the concentration data of patients who discontinue or are withdrawn from the study. Additional pharmacokinetics parameters may be determined if data permits.

[0241] Another potential summary may include stratification by smoking status. If data permits, pharmacokinetics of TEV-56286 will also be evaluated using a population pharmacokinetics approach. TEV-56286 concentration data from this study will be combined with data from previous studies to characterize the following population pharmacokinetics parameters and derived estimates:

[0242] Steady state:• Apparent clearance (CL / F) and volume (V / F),• Area under the plasma concentration-time curve (AUC) over the dosing interval (AUCtau),• Maximum concentration at steady state (Cmax.ss),Minimum concentration at steady state (Cmin,ss) prior to dosing,Average plasma concentration (Cavg) over a dosing interval, Time to maximum concentration (tmax).

[0243] Single-dose:• AUC from time 0 to infinity (AUCo-ao),• Cmax,• tmax,• Terminal elimination half-life (b / 2),• Additional pharmacokinetics parameters may be calculated if data permits.

[0244] The results of the population pharmacokinetics analysis may be presented in a separate pharmacometrics report.

[0245] (viii) Pharmacokinetic s / Pharmacodynamics Analysis

[0246] Population pharmacokinetic s / pharmacodynamics analysis of relevant pharmacodynamics variables and / or exposure-response analysis of relevant efficacy or safety endpoints may be performed.

[0247] If relevant data permits, the relationship between the plasma and / or CSF concentration of TEV-56286 and pharmacodynamics endpoints may be investigated.

[0248] The pharmacodynamics endpoints for analysis may include modified UMSARS part I, total UMSARS score (part I and part II combined), CGI-S, patient survival, adverse events of interest, NfL in plasma, and MRI brain regional volumetric changes. Additional relevant pharmacodynamics endpoints may be included.

[0249] The results of the pharmacokinetic s / pharmacodynamics analysis may be presented in a separate report.

[0250] (ix) Biomarker Analysis

[0251] Change from baseline analysis of selected biomarkers (which may include misfolded alpha-synuclein, NfL in plasma and CSF, glial fibrillary acidic protein concentrations, and disease / pathway-implicative proteins) may be assessed in the relevant matrix.

[0252] Biomarker analysis plans and results, with the exception of NfL concentrations in the plasma, may be reported separately from the main study results.

[0253] Additional relevant biomarker endpoints may be included. The final list of biomarkers to be evaluated will be determined at the time of analysis to be able to account for the most current research.

[0254] (x) Pharmacogenomics Analysis

[0255] Genetic variation assessed from blood such as in CYP1A2 and corresponding metabolizer status may be assessed using descriptive statistics for an association with study endpoints, including safety endpoints, IMP concentration, and pharmacokinetics parameters.

[0256] Pharmacogenetics analysis plans and results may be reported separately from the main study results.

[0257] Additional relevant pharmacogenomics endpoints may be included. The final list of biomarkers to be evaluated will be determined at the time of analysis to be able to account for the most current research.

[0258] (xi) Digital Biomarker Analysis

[0259] Change from baseline in selected digital biomarkers (which may include parameters of gait, general activity and sleep) may be assessed across various timepoints.

[0260] Digital biomarker analysis plans and results may be reported separately from the main study results.

[0261] Raw study data may be used to develop new algorithms to detect disease related and / or general health related features yet to be identified in the study population. Development of novel algorithms may require splitting the data into testing and validation data sets. Details would be outlined in the digital biomarker analysis plan. The final list of digital biomarkers to be evaluated will be determined at the time of analysis to be able to account for the most current research.

[0262] G. Prohibited Medications and Interactions

[0263] The following describes the restricted medications for the study due to potential interactions with TEV-56286.

[0264] The following drugs and compounds are excluded per the study inclusion criteria:• deferiprone• byetta• alpha-synuclein antibodies or vaccinations• anti-sense oligonucleotides• GLP-1 agonists not used for the sole purpose of treating type 2 diabetes mellitus

[0265] Moderate and strong Cytochrome P450 1A2 inhibitors are prohibited from 2 weeks or 5 half-lives (whichever is of longer duration) prior to first dose of TEV-56286 and until 4 daysafter last dose of TEV-56286. Below is a partial list of moderate and strong CYP450 1A2 inhibitors prohibited:• fluvoxamine• ciprofloxacin• enoxacin• viloxazine• clinafloxacin• idrocilamide• etintidine• genistein• methoxsalen (8-methoxypsoralen)• mexiletine• vemurafenib• osilodrostat• phenylpropanolamine• pipemidic acid• estrogen-containing medications

[0266] In addition to what has been described herein, the methods and systems may also be implemented in a computer program(s), software, or firmware incorporated in one or more computer-readable media for execution by a computer(s) or processor(s), for example.Examples of computer-readable media include electronic signals (transmitted over wired or wireless connections) and tangible / non-transitory computer-readable storage media. Examples of tangible / non-transitory computer-readable storage media include, but are not limited to, a read only memory (ROM), a random-access memory (RAM), removable disks, and optical media such as CD-ROM disks, and digital versatile disks (DVDs).

[0267] While this disclosure has been described in terms of certain embodiments and generally associated methods, alterations and permutations of the embodiments and methods will be apparent to those skilled in the art. Accordingly, the above description of example embodiments does not constrain this disclosure. Other changes, substitutions, and alterations are also possible without departing from the spirit and scope of this disclosure.

Claims

CLAIMSWhat is claimed is:

1. A method for detecting the progression of multiple system atrophy (MSA) over a time period using a trained machine learning model, the method comprising: receiving in-clinic sensor data and at-home sensor data for a patient over the time period; determining respective values indicative of a plurality of different MSA-related activity features for the user based on the in-clinic sensor data and the at-home sensor data; and determining, using the trained machine learning model, the progression of MSA over the time period for the user using the respective values for the plurality of different MSA- related activity features for the patient, wherein the machine learning model was trained to determine the progression of MSA over the time period using respective values for the plurality of different MSA-related activity features for each of a plurality of different patients.

2. The method of claim 1, wherein the progression of MSA over the time period comprises at least one of a change in Unified Multiple System Atrophy Rating Scale (UMSARS) Part I score, a change in modified UMSARS Part I score, a change in UMSARS Part II score, a change in UMSARS Part IV score, a change in combined Unified Multiple System Atrophy Rating Scale (UMSARS) Part I and Part II scores, or a change in combined modified UMSARS Part I and Part II scores.

3. The method of claim 1, wherein the in-clinic sensor data comprises one or more of Anticipatory Postural Adjustment (APA) - Duration, APA - First Step Duration, APA - First Step Range of Motion, APA - Forward APA Peak, APA - Lateral APA Peak, Gait / Lower Limb - Cadence (Left), Gait / Lower Limb - Cadence (Right), Gait / Lower Limb - Circumduction (Left), Gait / Lower Limb - Circumduction (Right), Gait / Lower Limb - Cycle Validity Flags (Left), Gait / Lower Limb - Cycle Validity Flags (Right), Gait / Lower Limb - Double Support (Left), Gait / Lower Limb - Double Support (Right), Gait / Lower Limb - Elevation at Midswing (Left), Gait / Lower Limb - Elevation at Midswing (Right), Gait / Lower Limb - Foot Strike Angle (Left), Gait / Lower Limb - Foot Strike Angle (Right), Gait / Lower Limb - Gait Cycle Duration (Left), Gait / Lower Limb - Gait Cycle Duration (Right), Gait / Lower Limb - GaitCycle Validity (Left), Gait / Lower Limb - Gait Cycle Validity (Right), Gait / Lower Limb - Gait Speed (Left), Gait / Lower Limb - Gait Speed (Right), Gait / Lower Limb - Lateral Step Variability (Left), Gait / Lower Limb - Lateral Step Variability (Right), Gait / Lower Limb - N, Gait / Lower Limb - Single Limb Support (Left), Gait / Lower Limb - Single Limb Support (Right), Gait / Lower Limb - Stance (Left), Gait / Lower Limb - Stance (Right), Gait / Lower Limb- Step Duration (Left), Gait / Lower Limb - Step Duration (Right), Gait / Lower Limb - Stride Length (Left), Gait / Lower Limb - Stride Length (Right), Gait / Lower Limb - Swing (Left), Gait / Lower Limb - Swing (Right), Gait / Lower Limb - Terminal Double Support (Left), Gait / Lower Limb - Terminal Double Support (Right), Gait / Lower Limb - Toe Off Angle (Left), Gait / Lower Limb - Toe Off Angle (Right), Gait / Lower Limb - Toe Out Angle (Left), Gait / Lower Limb - Toe Out Angle (Right), Gait / Lumbar - Coronal Range of Motion, Gait / Lumbar - Sagittal Range of Motion, Gait / Lumbar - Transverse Range of Motion, Gait / Trunk - Coronal Range of Motion, Gait / Trunk - Sagittal Range of Motion, Gait / Trunk - Transverse Range of Motion, Gait / Upper Limb - Arm Range of Motion (Left), Gait / Upper Limb - Arm Range of Motion (Right), Gait / Upper Limb - Arm Swing Velocity (Left), Gait / Upper Limb - Arm Swing Velocity (Right), Postural Sway / Acc - 95% Ellipse Axis 1 Radius, Postural Sway / Acc - 95% Ellipse Axis 2 Radius, Postural Sway / Acc - 95% Ellipse Rotation, Postural Sway / Acc - Centroidal Frequency, Postural Sway / Acc - Centroidal Frequency (Coronal), Postural Sway / Acc - Centroidal Frequency (Sagittal), Postural Sway / Acc- Frequency Dispersion, Postural Sway / Acc - Frequency Dispersion (Coronal), Postural Sway / Acc - Frequency Dispersion (Sagittal), Postural Sway / Acc - Jerk, Postural Sway / Acc - Jerk (Coronal), Postural Sway / Acc - Jerk (Sagittal), Postural Sway / Acc - Mean Velocity, Postural Sway / Acc - Mean Velocity (Coronal), Postural Sway / Acc - Mean Velocity (Sagittal), Postural Sway / Acc - Path Length, Postural Sway / Acc - Path Length (Coronal), Postural Sway / Acc - Path Length (Sagittal), Postural Sway / Acc - RMS Sway, Postural Sway / Acc - RMS Sway (Coronal), Postural Sway / Acc - RMS Sway (Sagittal), Postural Sway / Acc - Range, Postural Sway / Acc - Range (Coronal), Postural Sway / Acc - Range (Sagittal), Postural Sway / Acc - Sway Area, Postural Sway / Angles - 95% Ellipse Axis 1 Radius, PosturalSway / Angles - 95% Ellipse Axis 2 Radius, Postural Sway / Angles - 95% Ellipse Rotation, Postural Sway / Angles - Duration, Postural Sway / Angles - RMS Sway, Postural Sway / Angles - RMS Sway (Coronal), Postural Sway / Angles - RMS Sway (Sagittal), Postural Sway / Angles - Sway Area, Turns - Angle, Turns - Duration, Turns - N, Turns - Steps in Turn, Turns - Turn Velocity, Gait / Lower Limb - Gait Cycle Timing (Left), Gait / Lower Limb - Gait Cycle Timing(Right), Gait / Lower Limb - Toe Off Timing (Left), Gait / Lower Limb - Toe Off Timing (Right), Gait / Lower Limb - Midswing Timing (Left), Gait / Lower Limb - Midswing Timing (Right), Gait / Lower Limb - Initial Contact Next Timing (Left), Gait / Lower Limb - Initial Contact Next Timing (Right), Gait / Lower Limb - All Cycles Initial Contact Timing (Left), Gait / Lower Limb - All Cycles Initial Contact Timing (Right), Turns - Turn Timing, Duration, Sit to Stand - Duration, Sit to Stand - Lean Angle, Sit to Stand - N, Sit to Stand - Sit to Stand Timing, Stand to Sit - Stand to Sit Timing, Duration Sway test, or Duration Tug test.

4. The method of claim 1, wherein the in-clinic sensor data is collected during structured tasks performed by a patient, the structured tasks comprising timed stationary standing, timed walk, or timed up and go (TUG).

5. The method of claim 1, wherein the at-home sensor data comprises number of steps, step time, mean cadence, a confidential interval of cadence, stride time variability, sit-to-stand duration, active time, inactive time, walking duration, stand duration, sitting / lying duration, total sleep duration, sleep efficiency, wake after sleep onset count, wake after sleep onset duration, and wake after sleep onset median.

6. The method of claim 1, wherein the at-home sensor data is collected in response to unprompted movements and activities performed by the respective patient.

7. The method of claim 1, wherein the in-clinic sensor data is collected using a different sensor device than the at-home sensor data.

8. The method of claim 1, wherein the at-home sensor data is collected using an at-home sensor that comprises a single wearable component, and wherein the in-clinic sensor data is collected using an in-clinic sensor that comprises a plurality of wearable components.

9. The method of claim 8, wherein the at-home sensor data is collected using a wrist-worn sensor device, and the in-clinic sensor data is collected using a sensor device system comprising one or more wrist- worn sensor devices, one or more ankle-worn sensor devices, a sensor platform device, a waist- worn sensor device, and a chest- worn sensor device.

10. The method of claim 9, wherein the in-clinic sensor data comprises more types of sensor data than the at-home sensor data.

11. The method of claim 1, further comprising: receiving survey data for the plurality of patients over the time period; and training the machine learning model using the survey data.

12. The method of claim 11, wherein the survey data is collected in response to a survey prompting a patient to identify symptoms meaningful to the patient or connections between the MSA-related activity features and the in-clinic sensor data or at-home sensor data.

13. The method of claim 12, wherein the survey provides a series of questions related to a meaningfulness of each of the MSA-related activity features.

14. The method of claim 13, further comprising: mapping the MSA-related activity features to one or more meaningful MSA-related symptoms.

15. The method of claim 11, further comprising: determining a subset of the MSA-related activity features based on the survey data; and training the machine learning model using the subset of the MSA-related activity features.

16. The method of claim 1, wherein the plurality of MSA-related activity features comprises gait features, sit-to-stand features, sleep features, walking initiation features, turning features, arm movement features, trunk and low back range of motion features, or postural sway features.

17. The method of claim 1, wherein the time period is 48 weeks.

18. The method of claim 1, wherein the in-clinic sensor data and the survey data are collected at a given frequency for each of the plurality of patients over the time period, and wherein the at-home sensor data is collected continuously for each of the plurality of patients over the time period.

19. The method of claim 18, wherein the given frequency is every 6 weeks.

20. The method of claim 1, wherein the plurality of patients comprises at least 100 patients.

21. The method of claim 1, wherein at least a subset of the plurality of patients are administered a medication associated with MSA, and the remainder of the plurality of patients are not administered the medication.

22. The method of claim 21, wherein the medication is a chemical or biologic medication.

23. The method of claim 21, wherein the medication is a treatment for one or more symptoms of MSA.

24. The method of claim 21, wherein the medication is administered pursuant to a clinical trial.

25. The method of claim 21, wherein the medication comprises emrusolmin, droxidopa, levodopa, ropinirole, pramipexole, clonazepam, vitamin E, propranolol, baclofen, amantadine, buspirone, ampreloxetine, ENT-01, Lu AF82422, AAV2-GDNF, ION464 or ATH434.

26. The method of claim 1, further comprising: administering to the patient a medication associated with MSA.

27. The method of claim 26, wherein the medication is a chemical or biologic medication.

28. The method of claim 26, wherein the medication is a treatment for one or more symptoms of MSA.

29. The method of claim 26, wherein the medication is administered pursuant to a clinical trial.

30. The method of claim 26, wherein the medication comprises emrusolmin, droxidopa, levodopa, ropinirole, pramipexole, clonazepam, vitamin E, propranolol, baclofen, amantadine, buspirone, ampreloxetine, ENT-01, Lu AF82422, AAV2-GDNF, ION464 or ATH434.

31. The method of claim 1 , further comprising: segmenting the plurality of patients into a plurality of groups; training the machine learning model using the respective values for the plurality of different MSA-related activity features for the patients of one of the plurality of groups; and testing the machine learning model using the respective values for the plurality of different MSA-related activity features for the patients of a remainder of the plurality of groups.

32. The method of claim 31, wherein each group of the plurality of groups comprises the same number of patients.

33. The method of claim 1, further comprising: training the machine learning model using covariates associated with each of the plurality of patients.

34. The method of claim 33, wherein the covariates comprise clinical MSA subtype of the patient, age of the patient, gender of the patient, or neurofilament light chain levels of the patient.

35. The method of claim 1, wherein the machine learning model comprises a classification and regression model.

36. The method of claim 1, further comprising: orally administering a medication, wherein the medication comprises one or more of emrusolmin, droxidopa, levodopa, ropinirole, pramipexole, clonazepam, vitamin E, propranolol, baclofen, amantadine, buspirone, ampreloxetine, ENT-01, Lu AF82422, AAV2- GDNF, ION464 and ATH434.

37. The method of claim 36, wherein the medication is emrusolmin, and about 300 mg / day of emrusolmin is orally administered to the patient.

38. The method of claim 1, wherein the patient has a diagnosis of clinically definite MSA, clinically probably MSA, or clinically possible MSA.

39. The method of claim 1, wherein each of the plurality of patients has a diagnosis of clinically definite MSA, clinically probably MSA, or clinically possible MSA.

40. A computer-readable storage medium comprising instructions, that when executed by one or more processors, cause the one or more processors to: receive in-clinic sensor data and at-home sensor data for a patient over a time period; determine respective values indicative of a plurality of different multiple system atrophy (MSA)-related activity features for the patient based on the in-clinic sensor data and the at-home sensor data; and determine, using a trained machine learning model, the progression of MSA over the time period for the patient using the respective values for the plurality of different MSA-related activity features for the patient, wherein the machine learning model was trained to determine the progression of MSA over the time period using respective values for the plurality of different MSA-related activity features for each of a plurality of different patients.

41. The computer-readable storage medium of claim 40, wherein the progression of MSA over the time period comprises at least one of a change in Unified Multiple System Atrophy Rating Scale (UMSARS) Part I score, a change in modified UMSARS Part I score, a change in UMSARS Part II score, a change in UMSARS Part IV score, a change in combined Unified Multiple System Atrophy Rating Scale (UMSARS) Part I and Part II scores, or a change in combined modified UMSARS Part I and Part II scores.

42. The computer-readable storage medium of claim 40, wherein the in-clinic sensor data comprises one or more of Anticipatory Postural Adjustment (APA) - Duration, APA - First Step Duration, APA - First Step Range of Motion, APA - Forward APA Peak, APA - Lateral APA Peak, Gait / Lower Limb - Cadence (Left), Gait / Lower Limb - Cadence (Right), Gait / Lower Limb - Circumduction (Left), Gait / Lower Limb - Circumduction (Right),Gait / Lower Limb - Cycle Validity Flags (Left), Gait / Lower Limb - Cycle Validity Flags (Right), Gait / Lower Limb - Double Support (Left), Gait / Lower Limb - Double Support (Right), Gait / Lower Limb - Elevation at Midswing (Left), Gait / Lower Limb - Elevation at Midswing (Right), Gait / Lower Limb - Foot Strike Angle (Left), Gait / Lower Limb - Foot Strike Angle (Right), Gait / Lower Limb - Gait Cycle Duration (Left), Gait / Lower Limb - Gait Cycle Duration (Right), Gait / Lower Limb - Gait Cycle Validity (Left), Gait / Lower Limb - Gait Cycle Validity (Right), Gait / Lower Limb - Gait Speed (Left), Gait / Lower Limb - Gait Speed (Right), Gait / Lower Limb - Lateral Step Variability (Left), Gait / Lower Limb - Lateral Step Variability (Right), Gait / Lower Limb - N, Gait / Lower Limb - Single Limb Support (Left), Gait / Lower Limb - Single Limb Support (Right), Gait / Lower Limb - Stance (Left), Gait / Lower Limb - Stance (Right), Gait / Lower Limb - Step Duration (Left), Gait / Lower Limb - Step Duration (Right), Gait / Lower Limb - Stride Length (Left), Gait / Lower Limb - Stride Length (Right), Gait / Lower Limb - Swing (Left), Gait / Lower Limb - Swing (Right), Gait / Lower Limb - Terminal Double Support (Left), Gait / Lower Limb - Terminal Double Support (Right), Gait / Lower Limb - Toe Off Angle (Left), Gait / Lower Limb - Toe Off Angle (Right), Gait / Lower Limb - Toe Out Angle (Left), Gait / Lower Limb - Toe Out Angle (Right), Gait / Lumbar - Coronal Range of Motion, Gait / Lumbar - Sagittal Range of Motion, Gait / Lumbar - Transverse Range of Motion, Gait / Trunk - Coronal Range of Motion, Gait / Trunk - Sagittal Range of Motion, Gait / Trunk - Transverse Range of Motion, Gait / Upper Limb - Arm Range of Motion (Left), Gait / Upper Limb - Arm Range of Motion (Right), Gait / Upper Limb - Arm Swing Velocity (Left), Gait / Upper Limb - Arm Swing Velocity (Right), Postural Sway / Acc - 95% Ellipse Axis 1 Radius, Postural Sway / Acc - 95% Ellipse Axis 2 Radius, Postural Sway / Acc - 95% Ellipse Rotation, Postural Sway / Acc - Centroidal Frequency, Postural Sway / Acc - Centroidal Frequency (Coronal), Postural Sway / Acc - Centroidal Frequency (Sagittal), Postural Sway / Acc - Frequency Dispersion, Postural Sway / Acc - Frequency Dispersion (Coronal), Postural Sway / Acc - Frequency Dispersion (Sagittal), Postural Sway / Acc - Jerk, Postural Sway / Acc - Jerk (Coronal), Postural Sway / Acc - Jerk (Sagittal), Postural Sway / Acc - Mean Velocity, Postural Sway / Acc - Mean Velocity (Coronal), Postural Sway / Acc - Mean Velocity (Sagittal), Postural Sway / Acc - Path Length, Postural Sway / Acc - Path Length (Coronal), Postural Sway / Acc - Path Length (Sagittal), Postural Sway / Acc - RMS Sway, Postural Sway / Acc - RMS Sway (Coronal), Postural Sway / Acc - RMS Sway (Sagittal), Postural Sway / Acc - Range, Postural Sway / Acc - Range (Coronal), Postural Sway / Acc - Range (Sagittal), Postural Sway / Acc - Sway Area, PosturalSway / Angles - 95% Ellipse Axis 1 Radius, Postural Sway / Angles - 95% Ellipse Axis 2 Radius, Postural Sway / Angles - 95% Ellipse Rotation, Postural Sway / Angles - Duration, Postural Sway / Angles - RMS Sway, Postural Sway / Angles - RMS Sway (Coronal), Postural Sway / Angles - RMS Sway (Sagittal), Postural Sway / Angles - Sway Area, Turns - Angle, Turns - Duration, Turns - N, Turns - Steps in Turn, Turns - Turn Velocity, Gait / Lower Limb - Gait Cycle Timing (Left), Gait / Lower Limb - Gait Cycle Timing (Right), Gait / Lower Limb - Toe Off Timing (Left), Gait / Lower Limb - Toe Off Timing (Right), Gait / Lower Limb - Midswing Timing (Left), Gait / Lower Limb - Midswing Timing (Right), Gait / Lower Limb - Initial Contact Next Timing (Left), Gait / Lower Limb - Initial Contact Next Timing (Right), Gait / Lower Limb - All Cycles Initial Contact Timing (Left), Gait / Lower Limb - All Cycles Initial Contact Timing (Right), Turns - Turn Timing, Duration, Sit to Stand - Duration, Sit to Stand - Lean Angle, Sit to Stand - N, Sit to Stand - Sit to Stand Timing, Stand to Sit - Stand to Sit Timing, Duration Sway test, or Duration Tug test.

43. The computer-readable storage medium of claim 40, wherein the in-clinic sensor data is collected during structured tasks performed by a patient, the structured tasks comprising timed stationary standing, timed walk, or timed up and go (TUG).

44. The computer-readable storage medium of claim 40, wherein the at-home sensor data comprises number of steps, step time, mean cadence, a confidential interval of cadence, stride time variability, sit-to-stand duration, active time, inactive time, walking duration, stand duration, sitting / lying duration, total sleep duration, sleep efficiency, wake after sleep onset count, wake after sleep onset duration, and wake after sleep onset median.

45. The computer-readable storage medium of claim 40, wherein the at-home sensor data is collected in response to unprompted movements and activities performed by the respective patient.

46. The computer-readable storage medium of claim 40, wherein the in-clinic sensor data is collected using different sensor device than the at-home sensor data.

47. The computer-readable storage medium of claim 40, wherein the at-home sensor data is collected using an at-home sensor that comprises a single wearable component, and whereinthe in-clinic sensor data is collected using an in-clinic sensor that comprises a plurality of wearable components.

48. The computer-readable storage medium of claim 47, wherein the at-home sensor data is collected using a wrist- worn sensor device.

49. The computer-readable storage medium of claim 47, wherein the in-clinic sensor data is collected using a sensor device system comprising one or more wrist-worn sensor devices, one or more ankle- worn sensor devices, a sensor platform device, a waist-worn sensor device, and a chest-worn sensor device.

50. The computer-readable storage medium of claim 40, wherein the computer-readable storage medium further comprises instructions, that when executed by the one or more processors, cause the one or more processors to: receive survey data for the plurality of patients over the time period; and train the machine learning model using the survey data.

51. The computer-readable storage medium of claim 50, wherein the survey data is collected in response to a survey prompting a patient to identify symptoms meaningful to the patientor connections between the MSA-related activity features and the in-clinic sensor data or at-home sensor data.

52. The computer-readable storage medium of claim 51, wherein the survey provides a series of questions related to a meaningfulness of each of the MSA-related activity features.

53. The computer-readable storage medium of claim 52, wherein the computer-readable storage medium further comprises instructions, that when executed by the one or more processors, cause the one or more processors to: map the MSA-related activity features to one or more meaningful symptoms.

54. The computer-readable storage medium of claim 50, wherein the computer-readable storage medium further comprises instructions, that when executed by the one or more processors, cause the one or more processors to:determine a subset of the MSA-related activity features based on the survey data; and training the machine learning model using the subset of the MSA-related activity features.

55. The computer-readable storage medium of claim 40, wherein the plurality of MSA- related activity features comprises gait features, sit-to-stand features, sleep features, walking initiation features, turning features, arm movement features, trunk and low back range of motion features, or postural sway features.

56. The computer-readable storage medium of claim 40, wherein the time period is 48 weeks.

57. The computer-readable storage medium of claim 40, wherein the in-clinic sensor data and the survey data are collected at a given frequency for each of the plurality of patients over the time period, and wherein the at-home sensor data is collected continuously for each of the plurality of patients over the time period.

58. The computer-readable storage medium of claim 57, wherein the given frequency is every 6 weeks.

59. The computer-readable storage medium of claim 40, wherein the plurality of patients comprises at least 100 patients.

60. The computer-readable storage medium of claim 40, wherein a subset of the plurality of patients are taking a medication, and the remainder of the plurality of patients are not taking the medication.

61. The computer-readable storage medium of claim 40, wherein the medication comprises one or more of emrusolmin, droxidopa, levodopa, ropinirole, pramipexole, clonazepam, vitamin E, propranolol, baclofen, amantadine, buspirone, ampreloxetine, ENT-01, Lu AF82422, AAV2-GDNF, ION464 and ATH434.

62. The computer-readable storage medium of claim 40, wherein the computer-readable storage medium further comprises instructions, that when executed by the one or more processors, cause the one or more processors to: segment the plurality of patients into a plurality of groups; train the machine learning model using the respective values for the plurality of different MSA-related activity features for the patients of one of the plurality of groups; and test the machine learning model using the respective values for the plurality of different MSA-related activity features for the patients of a remainder of the plurality of groups.

63. The computer-readable storage medium of claim 62, wherein each group of the plurality of groups comprises the same number of patients.

64. The computer-readable storage medium of claim 40, wherein the computer-readable storage medium further comprises instructions, that when executed by the one or more processors, cause the one or more processors to: train the machine learning model using covariates associated with each of the plurality of patients.

65. The computer-readable storage medium of claim 64, wherein the covariates comprise clinical MSA subtype of the patient, age of the patient, gender of the patient, or neurofilament light chain levels of the patient.

66. The computer-readable storage medium of claim 40, wherein the machine learning model comprises a classification and regression model.

67. A method for training a machine learning model to detect progression of multiple system atrophy (MSA) over a time period, the method comprising: receiving in-clinic sensor data and at-home sensor data for a plurality of patients over the time period; determining respective values indicative of a plurality of different MSA-related activity features for each of the plurality of patients based on the in-clinic sensor data and at-home sensor data; andtraining the machine learning model to determine the progression of MSA over the time period using the respective values for the plurality of different MSA-related activity features for each of the plurality of patients.

68. The method of claim 67, wherein the progression of MSA over the time period comprises at least one of a change in Unified Multiple System Atrophy Rating Scale (UMSARS) Part I score, a change in modified UMSARS Part I score, a change in UMSARS Part II score, a change in UMSARS Part IV score, a change in combined Unified Multiple System Atrophy Rating Scale (UMSARS) Part I and Part II scores, or a change in combined modified UMSARS Part I and Part II scores.

69. The method of claim 67, wherein the in-clinic sensor data comprises one or more of Anticipatory Postural Adjustment (APA) - Duration, APA - First Step Duration, APA - First Step Range of Motion, APA - Forward APA Peak, APA - Lateral APA Peak, Gait / Lower Limb- Cadence (Left), Gait / Lower Limb - Cadence (Right), Gait / Lower Limb - Circumduction (Left), Gait / Lower Limb - Circumduction (Right), Gait / Lower Limb - Cycle Validity Flags (Left), Gait / Lower Limb - Cycle Validity Flags (Right), Gait / Lower Limb - Double Support (Left), Gait / Lower Limb - Double Support (Right), Gait / Lower Limb - Elevation at Midswing (Left), Gait / Lower Limb - Elevation at Midswing (Right), Gait / Lower Limb - Foot Strike Angle (Left), Gait / Lower Limb - Foot Strike Angle (Right), Gait / Lower Limb - Gait Cycle Duration (Left), Gait / Lower Limb - Gait Cycle Duration (Right), Gait / Lower Limb - Gait Cycle Validity (Left), Gait / Lower Limb - Gait Cycle Validity (Right), Gait / Lower Limb - Gait Speed (Left), Gait / Lower Limb - Gait Speed (Right), Gait / Lower Limb - Lateral Step Variability (Left), Gait / Lower Limb - Lateral Step Variability (Right), Gait / Lower Limb - N, Gait / Lower Limb - Single Limb Support (Left), Gait / Lower Limb - Single Limb Support (Right), Gait / Lower Limb - Stance (Left), Gait / Lower Limb - Stance (Right), Gait / Lower Limb- Step Duration (Left), Gait / Lower Limb - Step Duration (Right), Gait / Lower Limb - Stride Length (Left), Gait / Lower Limb - Stride Length (Right), Gait / Lower Limb - Swing (Left), Gait / Lower Limb - Swing (Right), Gait / Lower Limb - Terminal Double Support (Left), Gait / Lower Limb - Terminal Double Support (Right), Gait / Lower Limb - Toe Off Angle (Left), Gait / Lower Limb - Toe Off Angle (Right), Gait / Lower Limb - Toe Out Angle (Left), Gait / Lower Limb - Toe Out Angle (Right), Gait / Lumbar - Coronal Range of Motion, Gait / Lumbar - Sagittal Range of Motion, Gait / Lumbar - Transverse Range of Motion,Gait / Trunk - Coronal Range of Motion, Gait / Trunk - Sagittal Range of Motion, Gait / Trunk - Transverse Range of Motion, Gait / Upper Limb - Arm Range of Motion (Left), Gait / Upper Limb - Arm Range of Motion (Right), Gait / Upper Limb - Arm Swing Velocity (Left), Gait / Upper Limb - Arm Swing Velocity (Right), Postural Sway / Acc - 95% Ellipse Axis 1 Radius, Postural Sway / Acc - 95% Ellipse Axis 2 Radius, Postural Sway / Acc - 95% Ellipse Rotation, Postural Sway / Acc - Centroidal Frequency, Postural Sway / Acc - Centroidal Frequency (Coronal), Postural Sway / Acc - Centroidal Frequency (Sagittal), Postural Sway / Acc - Frequency Dispersion, Postural Sway / Acc - Frequency Dispersion (Coronal), Postural Sway / Acc - Frequency Dispersion (Sagittal), Postural Sway / Acc - Jerk, Postural Sway / Acc - Jerk (Coronal), Postural Sway / Acc - Jerk (Sagittal), Postural Sway / Acc - Mean Velocity, Postural Sway / Acc - Mean Velocity (Coronal), Postural Sway / Acc - Mean Velocity (Sagittal), Postural Sway / Acc - Path Length, Postural Sway / Acc - Path Length (Coronal), Postural Sway / Acc - Path Length (Sagittal), Postural Sway / Acc - RMS Sway, Postural Sway / Acc - RMS Sway (Coronal), Postural Sway / Acc - RMS Sway (Sagittal), Postural Sway / Acc - Range, Postural Sway / Acc - Range (Coronal), Postural Sway / Acc - Range (Sagittal), Postural Sway / Acc - Sway Area, Postural Sway / Angles - 95% Ellipse Axis 1 Radius, Postural Sway / Angles - 95% Ellipse Axis 2 Radius, Postural Sway / Angles - 95% Ellipse Rotation, Postural Sway / Angles - Duration, Postural Sway / Angles - RMS Sway, Postural Sway / Angles - RMS Sway (Coronal), Postural Sway / Angles - RMS Sway (Sagittal), Postural Sway / Angles - Sway Area, Turns - Angle, Turns - Duration, Turns - N, Turns - Steps in Turn, Turns - Turn Velocity, Gait / Lower Limb - Gait Cycle Timing (Left), Gait / Lower Limb - Gait Cycle Timing (Right), Gait / Lower Limb - Toe Off Timing (Left), Gait / Lower Limb - Toe Off Timing (Right), Gait / Lower Limb - Midswing Timing (Left), Gait / Lower Limb - Midswing Timing (Right), Gait / Lower Limb - Initial Contact Next Timing (Left), Gait / Lower Limb - Initial Contact Next Timing (Right), Gait / Lower Limb - All Cycles Initial Contact Timing (Left), Gait / Lower Limb - All Cycles Initial Contact Timing (Right), Turns - Turn Timing, Duration, Sit to Stand - Duration, Sit to Stand - Lean Angle, Sit to Stand - N, Sit to Stand - Sit to Stand Timing, Stand to Sit - Stand to Sit Timing, Duration Sway test, or Duration Tug test.

70. The method of claim 67, wherein the in-clinic sensor data is collected during structured tasks performed by a patient, the structured tasks comprising timed stationary standing, timed walk, or timed up and go (TUG).

71. The method of claim 67, wherein the at-home sensor data comprises number of steps, step time, mean cadence, a confidential interval of cadence, stride time variability, sit-to-stand duration, active time, inactive time, walking duration, stand duration, sitting / lying duration, total sleep duration, sleep efficiency, wake after sleep onset count, wake after sleep onset duration, and wake after sleep onset median.

72. The method of claim 67, wherein the at-home sensor data is collected in response to unprompted movements and activities performed by the respective patient.

73. The method of claim 67, wherein the in-clinic sensor data is collected using a different sensor device than the at-home sensor data.

74. The method of claim 67, wherein the at-home sensor data is collected using an at-home sensor that comprises a single wearable component, and wherein the in-clinic sensor data is collected using an in-clinic sensor that comprises a plurality of wearable components.

75. The method of claim 74, wherein the at-home sensor data is collected using a wrist- worn sensor device.

76. The method of claim 74, wherein the in-clinic sensor data is collected using a sensor device system comprising one or more wrist-worn sensor devices, one or more ankle-worn sensor devices, a sensor platform device, a waist- worn sensor device, and a chest-worn sensor device.

77. The method of claim 67, further comprising: receiving survey data for the plurality of patients over the time period; and training the machine learning model using the survey data.

78. The method of claim 77, wherein the survey data is collected in response to a survey prompting a patient to identify symptoms meaningful to the patient or connections between the MSA-related activity features and the in-clinic sensor data or at-home sensor data.

79. The method of claim 78, wherein the survey provides a series of questions related to a meaningfulness of each of the MSA-related activity features.

80. The method of claim 79, further comprising: mapping the MSA-related activity features to one or more meaningful symptoms.

81. The method of claim 77, further comprising: determining a subset of the MSA-related activity features based on the survey data; and training the machine learning model using the subset of the MSA-related activity features.

82. The method of claim 67, wherein the plurality of MSA-related activity features comprise gait features, sit-to-stand features, sleep features, walking initiation features, turning features, arm movement features, trunk and low back range of motion features, or postural sway features.

83. The method of claim 67, wherein the time period is 48 weeks.

84. The method of claim 67, wherein the in-clinic sensor data and the survey data is collected at a given frequency for each of the plurality of patients over the time period, and wherein the at-home sensor data is collected continuously for each of the plurality of patients over the time period.

85. The method of claim 84, wherein the given frequency is every 6 weeks.

86. The method of claim 67, wherein the plurality of patients comprises at least 100 patients.

87. The method of claim 67, wherein a subset of the plurality of patients are taking a medication, and the remainder of the plurality of patients are not taking the medication.

88. The method of claim 87, wherein the medication comprises one or more of emrusolmin, droxidopa, levodopa, ropinirole, pramipexole, clonazepam, vitamin E, propranolol, baclofen,amantadine, buspirone, ampreloxetine, ENT-01, Lu AF82422, AAV2-GDNF, ION464 and ATH434.

89. The method of claim 67, further comprising: segmenting the plurality of patients into a plurality of groups; training the machine learning model using the respective values for the plurality of different MSA-related activity features for the patients of one of the plurality of groups; and testing the machine learning model using the respective values for the plurality of different MSA-related activity features for the patients of a remainder of the plurality of groups.

90. The method of claim 89, wherein each group of the plurality of groups comprises the same number of patients.

91. The method of claim 67, further comprising: training the machine learning model using covariates associated with each of the plurality of patients.

92. The method of claim 91, wherein the covariates comprise clinical MSA subtype of the patient, age of the patient, gender of the patient, or neurofilament light chain levels of the patient.

93. The method of claim 67, wherein the machine learning model comprises a classification and regression model.

94. A method for detecting the progression of multiple system atrophy (MSA) over a time period using a trained machine learning model, the method comprising: receiving in-clinic sensor data and at-home sensor data for a patient over the time period; determining respective values indicative of a plurality of different MSA-related activity features for the user based on the in-clinic sensor data and the at-home sensor data; determining, using the trained machine learning model, the progression of MSA over the time period for the user using the respective values for the plurality of different MSA-related activity features for the patient, wherein the machine learning model was trained to determine the progression of MSA over the time period using respective values for the plurality of different MSA-related activity features for each of a plurality of different patients; and orally administering, on a daily basis, emrusolmin based on the progression of MSA over the time period for the user.

95. The method of claim 94, wherein 300 mg of emrusolmin is administered based on the progression of MSA over the time period for the user.

96. The method of claim 94, wherein the progression of MSA over the time period comprises at least one of a change in Unified Multiple System Atrophy Rating Scale (UMSARS) Part I score, a change in modified UMSARS Part I score, a change in UMSARS Part II score, or a change in UMSARS Part IV score.

97. The method of claim 94, wherein the progression of MSA over the period of time comprises at least one of a change in combined Unified Multiple System Atrophy Rating Scale (UMSARS) Part I and Part II scores, or a change in combined modified UMSARS Part I and Part II scores.

98. The method of claim 94, wherein the in-clinic sensor data comprises one or more of Anticipatory Postural Adjustment (APA) - Duration, APA - First Step Duration, APA - First Step Range of Motion, APA - Forward APA Peak, APA - Lateral APA Peak, Gait / Lower Limb - Cadence (Left), Gait / Lower Limb - Cadence (Right), Gait / Lower Limb - Circumduction (Left), Gait / Lower Limb - Circumduction (Right), Gait / Lower Limb - Cycle Validity Flags (Left), Gait / Lower Limb - Cycle Validity Flags (Right), Gait / Lower Limb - Double Support (Left), Gait / Lower Limb - Double Support (Right), Gait / Lower Limb - Elevation at Midswing (Left), Gait / Lower Limb - Elevation at Midswing (Right), Gait / Lower Limb - Foot Strike Angle (Left), Gait / Lower Limb - Foot Strike Angle (Right), Gait / Lower Limb - Gait Cycle Duration (Left), Gait / Lower Limb - Gait Cycle Duration (Right), Gait / Lower Limb - Gait Cycle Validity (Left), Gait / Lower Limb - Gait Cycle Validity (Right), Gait / Lower Limb - Gait Speed (Left), Gait / Lower Limb - Gait Speed (Right), Gait / Lower Limb - Lateral Step Variability (Left), Gait / Lower Limb - Lateral Step Variability (Right), Gait / Lower Limb - N,Gait / Lower Limb - Single Limb Support (Left), Gait / Lower Limb - Single Limb Support (Right), Gait / Lower Limb - Stance (Left), Gait / Lower Limb - Stance (Right), Gait / Lower Limb- Step Duration (Left), Gait / Lower Limb - Step Duration (Right), Gait / Lower Limb - Stride Length (Left), Gait / Lower Limb - Stride Length (Right), Gait / Lower Limb - Swing (Left), Gait / Lower Limb - Swing (Right), Gait / Lower Limb - Terminal Double Support (Left), Gait / Lower Limb - Terminal Double Support (Right), Gait / Lower Limb - Toe Off Angle (Left), Gait / Lower Limb - Toe Off Angle (Right), Gait / Lower Limb - Toe Out Angle (Left), Gait / Lower Limb - Toe Out Angle (Right), Gait / Lumbar - Coronal Range of Motion, Gait / Lumbar - Sagittal Range of Motion, Gait / Lumbar - Transverse Range of Motion, Gait / Trunk - Coronal Range of Motion, Gait / Trunk - Sagittal Range of Motion, Gait / Trunk - Transverse Range of Motion, Gait / Upper Limb - Arm Range of Motion (Left), Gait / Upper Limb - Arm Range of Motion (Right), Gait / Upper Limb - Arm Swing Velocity (Left), Gait / Upper Limb - Arm Swing Velocity (Right), Postural Sway / Acc - 95% Ellipse Axis 1 Radius, Postural Sway / Acc - 95% Ellipse Axis 2 Radius, Postural Sway / Acc - 95% Ellipse Rotation, Postural Sway / Acc - Centroidal Frequency, Postural Sway / Acc - Centroidal Frequency (Coronal), Postural Sway / Acc - Centroidal Frequency (Sagittal), Postural Sway / Acc- Frequency Dispersion, Postural Sway / Acc - Frequency Dispersion (Coronal), Postural Sway / Acc - Frequency Dispersion (Sagittal), Postural Sway / Acc - Jerk, Postural Sway / Acc - Jerk (Coronal), Postural Sway / Acc - Jerk (Sagittal), Postural Sway / Acc - Mean Velocity, Postural Sway / Acc - Mean Velocity (Coronal), Postural Sway / Acc - Mean Velocity (Sagittal), Postural Sway / Acc - Path Length, Postural Sway / Acc - Path Length (Coronal), Postural Sway / Acc - Path Length (Sagittal), Postural Sway / Acc - RMS Sway, Postural Sway / Acc - RMS Sway (Coronal), Postural Sway / Acc - RMS Sway (Sagittal), Postural Sway / Acc - Range, Postural Sway / Acc - Range (Coronal), Postural Sway / Acc - Range (Sagittal), Postural Sway / Acc - Sway Area, Postural Sway / Angles - 95% Ellipse Axis 1 Radius, PosturalSway / Angles - 95% Ellipse Axis 2 Radius, Postural Sway / Angles - 95% Ellipse Rotation, Postural Sway / Angles - Duration, Postural Sway / Angles - RMS Sway, Postural Sway / Angles - RMS Sway (Coronal), Postural Sway / Angles - RMS Sway (Sagittal), Postural Sway / Angles - Sway Area, Turns - Angle, Turns - Duration, Turns - N, Turns - Steps in Turn, Turns - Turn Velocity, Gait / Lower Limb - Gait Cycle Timing (Left), Gait / Lower Limb - Gait Cycle Timing (Right), Gait / Lower Limb - Toe Off Timing (Left), Gait / Lower Limb - Toe Off Timing (Right), Gait / Lower Limb - Midswing Timing (Left), Gait / Lower Limb - Midswing Timing (Right), Gait / Lower Limb - Initial Contact Next Timing (Left), Gait / Lower Limb - InitialContact Next Timing (Right), Gait / Lower Limb - All Cycles Initial Contact Timing (Left), Gait / Lower Limb - All Cycles Initial Contact Timing (Right), Turns - Turn Timing, Duration, Sit to Stand - Duration, Sit to Stand - Lean Angle, Sit to Stand - N, Sit to Stand - Sit to Stand Timing, Stand to Sit - Stand to Sit Timing, Duration Sway test, or Duration Tug test.

99. The method of claim 94, wherein the in-clinic sensor data is collected during structured tasks performed by a patient, the structured tasks comprising timed stationary standing, timed walk, or timed up and go (TUG).

100. The method of claim 94, wherein the at-home sensor data comprises number of steps, step time, mean cadence, a confidential interval of cadence, stride time variability, sit-to-stand duration, active time, inactive time, walking duration, stand duration, sitting / lying duration, total sleep duration, sleep efficiency, wake after sleep onset count, wake after sleep onset duration, and wake after sleep onset median.

101. The method of claim 94, wherein the at-home sensor data is collected in response to unprompted movements and activities performed by the respective patient.

102. The method of claim 94, wherein the in-clinic sensor data is collected using a different sensor device than the at-home sensor data.

103. The method of claim 94, wherein the at-home sensor data is collected using an at-home sensor that comprises a single wearable component, and wherein the in-clinic sensor data is collected using an in-clinic sensor that comprises a plurality of wearable components.

104. The method of claim 104, wherein the at-home sensor data is collected using a wrist- worn sensor device.

105. The method of claim 104, wherein the in-clinic sensor data is collected using a sensor device system comprising one or more wrist-worn sensor devices, one or more ankle-worn sensor devices, a sensor platform device, a waist- worn sensor device, and a chest-worn sensor device.

Images