Systems and methods for identifying disease progression biomarkers using non-ambulatory patient data

Abstract

A diagnostic platform for identifying disease progression biomarkers using wearable device–derived non-ambulatory patient data is described herein. For example, the diagnostic platform can obtain a step-count dataset that includes a time series of step-count data. The diagnostic platform can generate a set of ambulatory windows and a set of non-ambulatory windows associated with the step-count data. The diagnostic platform can generate a metric that is indicative of the continuity of bout duration. The diagnostic platform can determine whether a first human is exhibiting an indicator of a disease. The diagnostic platform can perform an action based on the indicator of the disease.

Description

PATENT Attorney Docket No. 124824.8118. WO01SYSTEMS AND METHODS FOR IDENTIFYING DISEASE PROGRESSION BIOMARKERS USING NON-AMBULATORY PATIENT DATACROSS-REFERENCE TO RELATED APPLICATIONS

[0001] This application claims priority to US Provisional Application No. 63 / 608,987, titled “Systems and Methods for Identifying Disease Progression Biomarkers Using NonAmbulatory Patient Data” and filed on December 12, 2023, which is incorporated by reference herein in its entirety.TECHNICAL FIELD

[0002] Various embodiments concern computer programs and associated computer-implemented techniques for determining digital biomarkers derived from information collected in free-living settings.BACKGROUND

[0003] The term “biomarker” - a portmanteau of “biological” and “marker” - is commonly used to refer to a measurable indicator of a physiological disease (also called a “physiological condition” or “physiological ailment”). Traditionally, biomarkers were evaluated through analysis of blood, urine, or soft tissue taken from a living body, generally for the purpose of predicting the onset of a physiological disease, monitoring the progression of a physiological disease, or determining the pharmacologic response to a therapeutic intervention for a physiological disease.

[0004] An emerging field of biomarkers relies on the analysis of measurements generated by sensors that monitor physiological aspects of living bodies. These biomarkers are commonly called “digital biomarkers,” since insights into the health of a given individual can be surfaced through analysis of measurements generated by sensors embedded in digital computing devices (or simply “computing devices”). Because computing devices with-1- 163869603.8PATENT Attorney Docket No. 124824.8118. WO01embedded sensors are now ubiquitous in society, digital biomarkers can serve as a reliable tool for advancing targeted treatment and guidance in health care.-2- 163869603.8PATENT Attorney Docket No. 124824.8118. WO01BRIEF DESCRIPTION OF THE DRAWINGS

[0005] Figure 1 illustrates a network environment that includes a diagnostic platform that is executed by a computing device.

[0006] Figure 2 illustrates an example of a computing device that is able to implement a diagnostic platform designed to establish digital biomarkers through the analysis of nonambulatory data collected during free-living scenarios.

[0007] Figure 3 depicts an example of a communication environment that includes a diagnostic platform that is configured to receive several types of data.

[0008] Figure 4 depicts another example of a communication environment that includes a diagnostic platform that is configured to obtain data from one or more sources.

[0009] Figure 5 includes a schematic diagram of an approach to identifying digital disease progression biomarkers through the analysis of non-ambulatory data collected during free-living scenarios.

[0010] Figure 6 includes a schematic diagram of an approach to deriving free-living progression biomarkers from free-living digital measurements.

[0011] Figure 7 depicts an illustration of measurements of ambulatory and nonambulatory bouts over the course of a day and over the course of a week for a patient.

[0012] Figure 8 depicts an illustration of measurements of ambulatory and nonambulatory bouts over the course of a day based on the availability of first and last stepcount data for a given day.

[0013] Figure 9A includes illustrations of continuity metrics associated with a relatively continuous non-walking pattern.

[0014] Figure 9B includes illustrations of continuity metrics associated with a less continuous non-walking pattern.

[0015] Figure 10A shows a plot of intraclass correlation (ICC) values against aggregation days for a specific digital measure.-3- 163869603.8PATENT Attorney Docket No. 124824.8118. WO01

[0016] Figure 10B shows a plot of correlation between a specific digital measure and a gait score.

[0017] Figure 10C includes plots of a specific digital measure over time for a group associated with progressive disease and a control group.

[0018] Figure 11 includes a flow diagram of a process for analyzing step-count data to determine ambulatory and non-ambulatory windows associated with determining disease progression.

[0019] Figure 12 includes a flow diagram of a process for correlating clinical measures of disease progression with non-ambulatory bout data associated with free-living scenarios.

[0020] Figure 13 includes a block diagram of a processing system in which at least some operations described herein can be implemented.

[0021] Various embodiments are shown in the drawings for the purpose of illustration. However, those skilled in the art will recognize that alternative embodiments may be employed without departing from the principles of the present disclosure. Accordingly, while certain embodiments are shown in the drawings, the technologies described herein are amenable to various modifications.-4- 163869603.8PATENT Attorney Docket No. 124824.8118. WO01DETAILED DESCRIPTION

[0022] The development of digital biomarkers has led to significant developments in detecting and monitoring physiological diseases. As an example, digital biomarkers have impacted the field of neurology where better approaches to reliably and consistently tracking motor function in a noninvasive manner are needed. Digital biomarkers are quantifiable measures that are established through analysis of measurements collected by sensors. As further discussed below, these sensors are generally included in computing devices that are worn on, implanted in, or digested by a living body. Over time, each of these sensors collects measurements, and these data can be used to explain, influence, or predict health-related outcomes.

[0023] Improvements in wearable technologies and devices enable such sensors to quantify and track data that may not be captured in regular visits to healthcare professionals on a periodic or sporadic (e.g., weekly, monthly, or yearly) basis. For example, digital biomarkers can improve diagnoses of progressive diseases by tracking free-living data over time. Assessments by care providers are generally subjective and limited to diagnoses based on any presented symptoms on a given day, leading to inconsistent or inaccurate diagnoses. For example, assessments for disease progression do not take account of information associated with a patient between regular visits. As such, digital biomarkers based on free-living data (e.g., through wearable technology) can improve the data available for the diagnosis or tracking disease progression over time.

[0024] However, relationships between free-living data and disease progression are not readily characterizable or quantifiable. Free-living data that includes step-count data for a given patient are not naturally indicative of the progression of certain diseases, such as Parkinson’s disease. Furthermore, any changes or trends in step counts may have an unpredictable relationship with disease progression, thereby complicating diagnosis, prognosis, or analysis of such diseases. Free-living data may exhibit issues in a high-frequency collection setting, as data may be noisy and inconsistent. For example, humans may wear wearable devices, such as watches, inconsistently (e.g., at different times every-5- 163869603.8PATENT Attorney Docket No. 124824.8118. WO01day). As such, analysis of free-living data for diagnosis and tracking of progressive diseases can be complicated and unpredictable.

[0025] Pre-existing data analysis methods have not identified systematic approaches to identifying disease progression biomarkers using high-frequency longitudinal data. For example, digital measurements provide both data regarding periods in which a human is walking (called “walking bouts” or “ambulatory bouts”) and periods in which a human is not walking (called “non-walking bouts” or “non-ambulatory bouts”). While non-ambulatory longitudinal data associated with a human can offer information that complements medical or scientific assessments, such information is not readily characterizable or quantifiable by healthcare professionals and has a somewhat unpredictable relationship with a given disease state. For example, no identifiable biological relationship exists between non-ambulatory longitudinal data associated with wearable devices and progressive diseases. Thus, though longitudinal data regarding walking and non-walking bouts can provide information relating to the relationship between a human’s behavior and disease progression, any such relationships are not readily apparent.

[0026] To illustrate, the onset and progression of Parkinson’s disease may be characterized by tremors in hands, arms, legs, the jaw, or the head, as well as muscle stiffness, slowness of movement, and impaired balance or coordination. While doctor visits can aid in the diagnosis and characterization of the progression of such diseases, evaluations may rely on subjective comparisons of a patient’s reported or observed symptoms with literature regarding the likely symptoms of the disease. However, such evaluations may suffer from issues relating to quantification or progression over time. For example, a medical practitioner may struggle to determine whether a tremor is worse or better over time and to what extent any detected changes in these tremors are indicative of disease progression.

[0027] Introduced here are systems and methods for the analysis of non-ambulatory longitudinal data associated with humans for determination of disease progression indicators. For example, a diagnostic platform can receive information relating to a human’s movements over a period of time, such as an indication of the human’s steps over this period-6- 163869603.8PATENT Attorney Docket No. 124824.8118. WO01of time. In some implementations, the diagnostic platform generates indications of ambulatory windows (e.g., periods of time where the human is walking or otherwise active) and non-ambulatory windows (e.g., periods of time where the human is sedentary or otherwise inactive) associated with the human. Based on these windows, the diagnostic platform can generate a continuity metric indicating the continuity of such walking or nonwalking bouts. Based on the continuity metric, the diagnostic platform can determine whether the human is exhibiting an indicator of a disease, thereby enabling the diagnostic platform to react to the diagnosis (e.g., by alerting a healthcare professional, or by stratifying the human for examination or treatment). To illustrate, the diagnostic platform can determine whether the human likely exhibits Parkinson’s disease, and / or determine a degree or marker of progression of the disease over time (e.g., as compared to a previous time).

[0028] In some implementations, the diagnostic platform enables determination of representatives of markers that can be used as indicators for the presence or progression of a disease. For example, the diagnostic platform can obtain sets of aggregated metric time series data for multiple humans. The time-series data can include quantitative metrics or data that characterize the continuity of ambulatory and / or non-ambulatory bouts for each human. The diagnostic platform can retrieve clinical measures associated with these humans and establish whether the aggregated metric time-series data is sufficiently correlated with these clinical measures. For example, the diagnostic platform determines whether the aggregated metric time series satisfies a test-retest reliability criterion, a correlation criterion, and / or a progression rate criterion with respect to the clinical measures. As such, the diagnostic platform enables the determination of characteristics of longitudinal datasets (e.g., from wearable devices) associated with ambulatory or non-ambulatory bouts that are likely indicative of disease presence or progression.

[0029] Embodiments may be described in the context of executable instructions for the purpose of illustration. However, those skilled in the art will recognize that aspects of the technology could be implemented via hardware, firmware, or software. As an example, a computer program that is representative of a diagnostic platform may be executed by the processor of a computing device. The computer program may interface, directly or indirectly,-7- 163869603.8PATENT Attorney Docket No. 124824.8118. WO01with hardware, firmware, or other software implemented on the computing devices. For example, the computer program may obtain data that is collected, computed, or otherwise obtained by a sensor included in the computing device. Additionally or alternatively, the computer program may obtain data that is collected, computed, or otherwise obtained by a sensor included in another computing device. This data may be processed in accordance with the approaches described herein, so as to identify disease progression associated with ambulatory and / or non-ambulatory longitudinal data for humans.Terminology

[0030] References in the present disclosure to “an embodiment” or “some embodiments” means that the feature, function, structure, or characteristic being described is included in at least one embodiment. Occurrences of such phrases do not necessarily refer to the same embodiment, nor do they necessarily refer to alternative embodiments that are mutually exclusive of one another.

[0031] The term “based on” is to be construed in an inclusive sense rather than an exclusive sense. That is, in the sense of “including but not limited to.” Thus, the term “based on” is intended to mean “based at least in part on” unless otherwise noted.

[0032] The terms “connected,” “coupled,” and variants thereof are intended to include any connection or coupling between two or more elements, either direct or indirect. The connection or coupling can be physical, logical, or a combination thereof. For example, elements may be electrically or communicatively connected to one another despite not sharing a physical connection.

[0033] The term “module” may refer broadly to software, firmware, hardware, or combinations thereof. Modules are typically functional components that generate one or more outputs based on one or more inputs. A computer program may include or utilize one or more modules. For example, a computer program may utilize multiple modules that are responsible for completing different tasks, or a computer program may utilize a single module that is responsible for completing multiple tasks.-8- 163869603.8PATENT Attorney Docket No. 124824.8118. WO01

[0034] When used in reference to a list of items, the word “or” is intended to cover all of the following interpretations: any of the items in the list, all of the items in the list, and any combination of items in the list.

[0035] The terms "longitudinal signal" and “longitudinal data” may be used to refer to a series of values that are periodically generated over an interval of time rather than on an ad hoc basis. For example, a parameter may be measured by a sensor multiple times per second (e.g., at a sampling rate of 25, 50, or 100 hertz) in an ongoing manner, or a parameter may be measured once per minute for several hours. The length of time over which the series of values extend may depend on the nature of the parameter being measured. The term “parameter” is generally used to describe the unit of measure of the underlying signal. An example of a parameter is x-axis acceleration.

[0036] The terms “digital measurement” and “measurement” may be used to refer to data that are either collected directly from a computing device or derived from a raw signal output by a sensor included in a computing device. For example, a “measurement” may be a patient-reported outcome provided via a symptom survey, or a “measurement” may be step count as derived from values output by an accelerometer that is indicative of acceleration along one or more axes.

[0037] The term “digital biomarker” may be used to refer to a subset of measurements that are deemed fit for a diagnostic purpose and validated as diagnostically relevant. For example, step count may become a digital biomarker of disease progression (also called a “progression biomarker”) if it can be proven that step count is sensitive to worsening of mobility of the corresponding patient. However, before this relationship is established, step count may simply be a “measurement.”

[0038] The term "free-living scenario" may be used to refer to a scenario in which an individual under observation is permitted to live her life as if not under observation. As an example, fitness tracking devices (also called “fitness trackers”) may monitor movement of individuals by measuring acceleration of the wrist in free-living scenarios. Observation in free-living scenarios is generally less prone to the intentional alteration of habits that commonly occurs in more controlled environments.-9- 163869603.8PATENT Attorney Docket No. 124824.8118. WO01Overview of Diagnostic Platform

[0039] Figure 1 illustrates a network environment 100 that includes a diagnostic platform 102 that is executed by a computing device 104. An individual (also referred to as a “user”) can interact with the diagnostic platform 102 via interfaces 106. For example, a patient may be able to access an interface through which information regarding a physiological disease, such as digital biomarkers or analyses of digital markers, can be reviewed. As another example, a healthcare professional may be able to access an interface through which information regarding patients, such as digital biomarkers or analyses or digital markers, can be reviewed. Depending on the nature of the individual accessing the interfaces 106, the interfaces 106 may allow for the review of physiological data, examination of outputs produced by the diagnostic platform 102, and management of preferences. Some interfaces may be configured to facilitate interactions between patients and healthcare professionals, while other interfaces may be configured to serve as informative dashboards for patients or healthcare professionals.

[0040] The physiological data obtained by the diagnostic platform 102 could be associated with the individual accessing the interfaces 106 or some other person. For example, the interfaces 106 may enable a person diagnosed with a physiological disease to view her own physiological data. Alternatively, the interfaces may enable an individual to view physiological data associated with another person. In such embodiments, the individual may be a healthcare professional who is responsible for monitoring, managing, or treating the other person. Examples of healthcare professionals include physicians, nurses, dietitians, and the like.

[0041] As shown in Figure 1, the diagnostic platform 102 can reside in a network environment 100. Thus, the computing device 104 on which the diagnostic platform 102 resides can be connected to one or more networks 108A-B. Depending on its nature, the computing device 104 could be connected to a personal area network (“PAN”), local area network (“LAN”), wide area network (“WAN”), metropolitan area network (“MAN”), or cellular network. For example, if the computing device 104 is a computer server, then the computing device 104 may be accessible to users via respective mobile phones that are connected to-10- 163869603.8PATENT Attorney Docket No. 124824.8118. WO01the Internet via LANs. The physiological data to be examined by the diagnostic platform 102 may be generated by the respective mobile phones (e.g., sensors included in the respective mobile phones) or acquired by the respective mobile phones.

[0042] Additionally or alternatively, the computing device 104 may be connected to one or more other computing devices over a short-range wireless connectivity technology, such as Bluetooth®, Near Field Communication (“NFC”), Wi-Fi® Direct (also referred to as “WiFi P2P”), and the like. As an example, the diagnostic platform 102 could be embodied as a mobile application that is executed by a mobile phone. In such embodiments, the mobile phone may be communicative connected - via a wireless communication channel - to a source from which to acquire physiological data. The source could be a watch, fitness tracker, or another wearable computing device, for example. The physiological could alternatively be obtained from another computer program executing on the mobile phone. For example, the physiological data could instead be acquired from another mobile application executing on the mobile phone or the operating system of the mobile phone.

[0043] The interfaces 106 may be accessible via a web browser, desktop application, mobile application, or another form of computer program. For example, a patient may be able to access interfaces through which information regarding her own health, such as digital biomarkers or analyses of digital biomarkers, is provided by a mobile application executing on a mobile phone. As another example, a healthcare professional may be able to access interfaces through which information regarding one or more patients can be reviewed via a web browser. Accordingly, the interfaces 106 generated by the diagnostic platform 102 may be accessible on various computing devices, including mobile phones, tablet computers, desktop computers, and the like.

[0044] Generally, the diagnostic platform 102 is executed - at least partially - by a cloud computing service operated by, for example, Amazon Web Services®, Google Cloud Platform™, or Microsoft Azure®. Thus, the computing device 104 may be representative of a computer server that is part of a server system 110. Often, the server system 110 is comprised of multiple computer servers. These computer servers can include different types of data (e.g., physiological data and information regarding patients, such as name,-11- 163869603.8PATENT Attorney Docket No. 124824.8118. WO01demographic information, disease classification, etc.), algorithms for processing incoming data, and other assets. Those skilled in the art will recognize that these data could also be distributed among the server system 110 and one or more computing devices. As an example, data that is input by, or related to, patients may be stored on, and processed by, their own computing devices for security or privacy purposes.

[0045] Components of the diagnostic platform 102 could also be hosted locally. That is, part of the diagnostic platform 102 may reside on the computing device used to access one of the interfaces 106. For example, the diagnostic platform 102 may be embodied as a mobile application executing on a mobile phone as mentioned above. Note, however, that the mobile application may be communicatively connected to the server system 110 on which other components of the diagnostic platform 102 are hosted.

[0046] Figure 2 illustrates an example of a computing device that is able to implement a diagnostic platform designed to establish digital biomarkers through the analysis of nonambulatory data collected during free-living scenarios. As shown in Figure 2, the computing device 200 can include a processor 202, memory 204, display mechanism 206, communication module 208 and sensor suite 210. Each of these components is discussed in greater detail below.

[0047] Those skilled in the art will recognize that different combinations of these components may be present depending on the nature of the computing device 200. For example, if the computing device 200 is a computer server that is part of a server system (e.g., server system 110 of Figure 1), then the computing device 200 may not include the display mechanism 206 or sensor suite 210. Conversely, if the computing device 200 is a mobile phone, then the computing device 200 can include the display mechanism 206 and sensor suite 210.

[0048] As further discussed below, the computing device 200 on which the diagnostic platform 212 resides may not include sensors in some embodiments. In such embodiments, the data examined by the diagnostic platform 212 could instead be generated by one or more sensors 222A-N that are external to the computing device 200. For example, the-12- 163869603.8PATENT Attorney Docket No. 124824.8118. WO01computing device 200 may be a mobile phone, and the sensors 222A-N may be included in a watch or fitness tracker that is communicatively connected to the computing device 200.

[0049] The processor 202 can have generic characteristics similar to general-purpose processors, or the processor 202 may be an application-specific integrated circuit (“ASIC”) that provides control functions to the computing device 200. As shown in Figure 2, the processor 202 can be coupled to all components of the computing device 200, either directly or indirectly, for communication purposes.

[0050] The memory 204 can be comprised of any suitable type of storage medium, such as static random-access memory (“SRAM”), dynamic random-access memory (“DRAM”), electrically erasable programmable read-only memory (“EEPROM”), flash memory, or registers. In addition to storing instructions that can be executed by the processor 202, the memory 204 can also store data generated by the processor 202 (e.g., when executing the modules of the diagnostic platform 212). Note that the memory 204 is merely an abstract representation of a storage environment. The memory 204 could be comprised of actual integrated circuits (also called “chips”).

[0051] The display mechanism 206 can be any mechanism that is operable to visually convey information to a user. For example, the display mechanism 206 can be a panel that includes light-emitting diodes (“LEDs”), organic LEDs, liquid crystal elements, or electrophoretic elements. As further discussed below, outputs produced by the diagnostic platform 212 (e.g., through execution of its modules) can be posted to the display mechanism 206 for review by a user of the computing device 200.

[0052] The communication module 208 may be responsible for managing communications external to the computing device 200. The communication module 208 can be wireless communication circuitry that is able to establish wireless communication channels with other computing devices. Examples of wireless communication circuitry include 2.4 gigahertz (“GHz”) and 5 GHz chipsets compatible with Institute of Electrical and Electronics Engineers (“IEEE”) 802.11 - also referred to as “Wi-Fi chipsets.” Alternatively, the communication module 208 may be representative of a chipset configured for Bluetooth, NFC, and the like. Some computing devices - like mobile phones, tablet computers, and the-13- 163869603.8PATENT Attorney Docket No. 124824.8118. WO01like - are able to wirelessly communicate via separate channels, while other computing devices - like watches and fitness trackers - tend to wirelessly communicate via a single channel. Accordingly, the communication module 208 may be one of multiple communication modules implemented in the computing device 200, or the communication module 208 may be the only communication module implemented in the computing device 200.

[0053] The nature, number, and type of communication channels established by the computing device 200 - and more specifically, the communication module 208 - can depend on (i) the sources from which data is received by the diagnostic platform 212 and (ii) the destinations to which data is transmitted by the diagnostic platform 212. Assume, for example, that the diagnostic platform 212 resides on a mobile phone in the form of a mobile application. In such embodiments, the communication module 208 can communicate with sensors 222A-N external to the computing device 200 from which to obtain data. Moreover, the communication module 208 may communicate with a server system (e.g., server system 110 of Figure 1 ) to which analyses of the data - or the data itself - are transmitted.

[0054] Often, various sensors are implemented in the computing device 200. Collectively, these sensors may be referred to as the “sensor suite” 210 of the computing device 200. For example, the computing device 200 may include a motion sensor whose output is indicative of motion of the computing device 200 as a whole. Examples of motion sensors include accelerometers and gyroscopes. In some embodiments, the motion sensor is implemented in an inertial measurement unit (“IMU”) that measures the force, angular rate, or orientation of the computing device 200. The IMU may accomplish this through the use of one or more accelerometers, one or more gyroscopes, one or more magnetometers, or any combination thereof. As specific examples, the IMU could be a 6-axis IMU that draws low current and therefore is suitable for “always-on” applications in battery-driven computing devices, or the IMU could be a 3-axis IMU that includes logic-level shifting circuitry that can readily interface with a microcontroller. As another example, the computing device 200 may include an ambient light sensor whose output is indicative of the amount of light in the ambient environment.-14- 163869603.8PATENT Attorney Docket No. 124824.8118. WO01

[0055] As mentioned above, data could also be acquired from sensors 222A-N that are external to the computing device 200. These sensors 222A-N could be included in another computing device, such as a watch or fitness tracker, that is directly connected to the computing device 200. Alternatively, these sensors 222A-N could be discrete sensing units that are directly or indirectly connected to the computing device 200. For example, sensor 222A may be a pulse oximeter that monitors the oxygen saturation of the user of the computing device 200 or another person, for the purpose of creating a photoplethysmogram (“PPG”). To monitor the oxygen saturation, sensor 222A may be placed on a thin part of a living body, usually a fingertip or earlobe, and then pass two wavelengths of light through that body part toward a photodetector. The photodetector can measure the changing absorbance at each wavelength, allowing sensor 222A to determine the absorbances due to the pulsing of arterial blood through that body part.

[0056] For convenience, the diagnostic platform 212 is referred to as a computer program that resides within the memory 204. However, the diagnostic platform 212 could be comprised of software, firmware, or hardware that is implemented in, or accessible to, the computing device 200. In accordance with embodiments described herein, the diagnostic platform 212 can include a processing module 214, computation module 216, analytics module 218, and graphical user interface (“GUI”) module 220. These modules could be integral parts of the diagnostic platform 212, or these modules could be logically separate from the diagnostic platform 212 but operate “alongside” it. Together, these modules enable the diagnostic platform 212 to establish and detect signals associated with ambulatory and / or non-ambulatory free-living data to be used to gain insights into the health of a given individual (e.g., for evaluation of the presence or progression of progressive disease). As mentioned above, the given individual could be the user of the computing device 200 or another person.

[0057] The processing module 214 can process data obtained by the diagnostic platform 212 into a format that is suitable for the other modules. For example, the processing module 214 can apply operations to sensor data obtained from the sensor suite 210 or sensors 222A-N in preparation for analysis by the other modules of the diagnostic platform-15- 163869603.8PATENT Attorney Docket No. 124824.8118. WO01212. For example, the processing module 214 can filter or alter the sensor data, such that the sensor data can be more readily analyzed. As another example, the processing module 214 may parse the sensor data in order to temporally align the dataset obtained from each source. Accordingly, the processing module 214 may be responsible for ensuring that the appropriate sensor data is accessible to the other modules of the diagnostic platform 212.

[0058] The computation module 216 (also called a “computation system”) may be responsible for deriving different measurements based on the processed data received from the processing module 214. As further discussed below, these different measurements may depend on the nature of the insights to be surfaced by the diagnostic platform 212. For example, these different measurements may be representative of walking-related measurements, running-related measurements, sleeping-related measurements, eating-related measurements, stress-related measurements, or any combination thereof. The computation module 216 can compute, infer, or otherwise establish measurements for different segments of processed data. These segments may be referred to as “windows” of processed data.

[0059] The analytics module 218 (also called an “aggregation system”) may implement an analytics engine that, in operation, implements two programs in sequence. As further discussed below, the first program may provide a method for programmatically aggregating the processed data output by the processing module 214 and the computation module 216, while the second program may implement one of multiple methods for determining disease progression indicators associated with the aggregated data.

[0060] The GUI module 220 may be responsible for generating interfaces that are viewable on the display mechanism 206. Various types of information can be presented on these interfaces. For example, digital biomarkers that are calculated, derived, or otherwise obtained by the computation module 216 may be presented on an interface for display to the user. Similarly, analyses of the digital biomarkers may be presented on the interface for display to the user.

[0061] Figure 3 depicts an example of a communication environment 300 that includes a diagnostic platform 302 that is configured to receive several types of data. Here, for-16- 163869603.8PATENT Attorney Docket No. 124824.8118. WO01example, the diagnostic platform 302 receives response data 304, first sensor data 306 that is generated by a first sensor (e.g., sensor 222A of Figure 2), and second sensor data 308 that is generated by a second sensor (e.g., sensor 222B of Figure 2). Those skilled in the art will recognize that these data have been selected for the purpose of illustration. Other types of data, such as treatment data (e.g., including indicators of physiological disease, treatment regimens, etc.), could also be obtained by the diagnostic platform 302.

[0062] These data may be obtained from multiple sources.

[0063] Consider the response data 304, for example. The response data 304 could be obtained directly on the computing device on which the diagnostic platform 302 is executing. For example, if the diagnostic platform 302 is implemented on a mobile phone in the form of a mobile application, the response data 304 may be representative of input provided by an individual through interfaces generated by the mobile application. The input may be representative of responses to queries included in a questionnaire that is designed to elicit responses indicative of health. Alternatively, the response data 304 could be obtained from another computing device. Referring again to the aforementioned example in which the diagnostic platform 302 is implemented as a mobile application executing on a mobile phone, the mobile phone could obtain the response data 304 from a server system (e.g., server system 110 of Figure 1). The server system may manage a datastore in which responses provided by various individuals to questionnaires are documents.

[0064] As mentioned above, sensor data could be obtained from sensors included in the computing device that is responsible for executing the diagnostic platform 302, or sensor data could be obtained from sensors that are external to the computing device that is responsible for executing the diagnostic platform 302. Thus, the first and second sensor data 306, 308 may be generated by sensors included in the computing device on which the diagnostic platform 302 is executing, or the first and second sensor data 306, 308 may be generated by sensors that are external to the computing device on which the diagnostic platform 302 is executing. As an example, the first sensor data 306 may be generated by an IMU implemented in the computing device on which the diagnostic platform 302 is executing, while the second sensor data 308 may be generated by a pulse oximeter that is-17- 163869603.8PATENT Attorney Docket No. 124824.8118. WO01communicatively connected - either directly or indirectly - to the computing device on which the diagnostic platform 302 is executing.

[0065] Wearable devices can include sensors and generate associated sensor datasets (e.g., sets of sensor data). For example, wearable devices can include any wearable technology (e.g., technology designed to be used while worn). Wearable technology can include smartwatches, smart glasses, activity trackers (e.g., pedometers), smart rings, hearing aids, or implants (medical or otherwise). Wearable devices and associated sensors can directly or indirectly collect data associated with a user’s health, such as heart rate, calories burned, steps walked, blood pressure, release of biochemicals, time spent exercising, seizures, physical strain, body composition, and / or water levels. For example, wearable devices can utilize sensor data to determine periods of physical activity, including ambulatory periods, as well as periods of physical inactivity, including periods of sleep. As such, because wearable devices can be worn, such devices can collect data passively, enabling associated users (e.g., humans) to live freely during data collection. As such, wearable devices enable the collection and subsequent analysis of free-living data.

[0066] In some implementations, the sensors associated with the diagnostic platform 302 can include other devices capable of collecting data. For example, sensors 222A-222N can include sensors, such as inertial measurement units or accelerometers, associated with mobile devices, such as smartphones, tablets, or laptop computers. For example, sensors 222A-222N can be housed on a smartphone carried within a human’s pocket, thereby enabling dynamic collection of free-living data associated with the human’s movement. In some implementations, sensor data can include location-based measurements, such as global positioning system (GPS) or network-based location data (e.g., from internet protocol (IP) addresses or connections to associated base stations within a telecommunication network). As such, the data collected at the diagnostic platform 212 can include a variety of sensor measurements associated with movement or health, for example.

[0067] Figure 4 depicts another example of a communication environment 400 that includes a diagnostic platform 402 that is configured to obtain data from one or more sources. Here, the diagnostic platform 402 may obtain data from a mobile phone 404, watch-18- 163869603.8PATENT Attorney Docket No. 124824.8118. WO01406, laptop computer 408, or server system 410 (collectively referred to as the “networked devices”). For example, the diagnostic platform 402 may obtain response data (e.g., response data 304 of Figure 3) from the laptop computer 408 or server system 410. As another example, the diagnostic platform 402 may obtain sensor data (e.g., sensor data 306, 308 of Figure 3) from the mobile phone 404 or watch 406.

[0068] The networked devices can be connected to the diagnostic platform 402 via one or more networks. These networks can include PANs, LANs, WANs, MANs, cellular networks, the Internet, etc. Additionally or alternatively, the networked devices may communicate with one another over a short-range wireless connectivity technology. For example, if the diagnostic platform 402 resides on the mobile phone 404 in the form of a mobile application, data may be obtained from the watch 406 over a Bluetooth communication channel while data may be obtained from the server system 410 over the Internet via a Wi-Fi communication channel.

[0069] Embodiments of the communication environment 400 may include a subset of the networked devices. For example, the communication environment 400 may include a diagnostic platform 402 that obtains, in real time, data from the mobile phone 404 and watch 406 as that data is generated over the course of a free-living scenario. Additional data could be obtained from the server system 410 on a periodic basis (e.g., daily or weekly).Approaches to Identifying Disease Progression Biomarkers

[0070] Figure 5 includes a schematic diagram of an approach to identifying digital disease progression biomarkers through the analysis of non-ambulatory data collected during free-living scenarios.A. Signal Collection

[0071] Initially, a diagnostic platform can collect signals from one or more sources (operation 501 ). For example, the diagnostic platform can collect, from a watch, signals that are representative of measurements generated by sensors included in the watch. As another example, the diagnostic platform can collect, from a server system (e.g., the server system 110 of Figure 1), a signal that is representative of data collected from a variety of devices -19- 163869603.8PATENT Attorney Docket No. 124824.8118. WO01associated with the user (e.g., mobile device accelerometer data), such as data associated with a user account. In some implementations, the diagnostic platform can collect, from a server system, a signal that is representative of responses to a questionnaire that is accessible via interfaces generated by the diagnostic platform.

[0072] In embodiments where the signal collected by the diagnostic platform is representative of a series of measurements generated by a sensor in temporal order, the signal is preferably a longitudinal signal with values generated over an interval of time. For example, the signal could include values that are generated at a predetermined cadence by an accelerometer or an inertial measurement unit (IMU). As another example, the signal could include values that are generated at a predetermined cadence by a pulse oximeter. The duration of the signal, and the cadence of its values, can depend on the nature of the sensor. For example, Figure 6 includes a schematic diagram 600 of an approach to deriving free-living progression biomarkers from free-living digital measurements. At operation 602, the processing module 214 can obtain or receive longitudinal raw sensor passive data, such as from a wearable device. For example, the diagnostic platform 212 can receive data 622 from a sensor associated with a smartwatch, as shown in Figure 6.

[0073] For example, a signal collected by the diagnostic platform at operation 501 can include movement data (e.g., data associated with the movement of a human). Movement data can include inertial measurement unit, accelerometer data, location data, or other data that is implicitly or explicitly indicative of a human’s movement. Movement data can include data associated with short-range movements, such as arm swings, rocking in a chair, or similar movements associated with small distances (e.g., a few meters or less). Additionally or alternatively, movement data can include data associated with longer-range movements, such as walking, running, hiking, cycling, or other movements of a longer distance (e.g., a few meters or move). As such, movement data can provide information relating to humans’ activity levels, including time periods associated with ambulatory (e.g., walking) or nonambulatory (non-walking) bouts. Movement data can inform conclusions associated with health, as changes in human movement and other physical activity can be indicative of symptoms of progressive disease. However, the relationship between movement data-20- 163869603.8PATENT Attorney Docket No. 124824.8118. WO01derived from wearable devices and progressive disease is not immediately apparent or predictable.B. Data Examination

[0074] As mentioned above, the diagnostic platform can include a data processing component - namely, a processing module (e.g., processing module 214 of Figure 2) - that can be responsible for applying a set of preprocessing operations to the signal. Said another way, the processing module can be responsible for examining the data included in the signal (step 502).

[0075] The preprocessing operations that are applied to the signal can depend on the type of data contained therein. Assume, for example, that the signal includes measurements generated by a sensor. In such a scenario, the processing module can apply standard preprocessing operations (e.g., a pre-processing algorithm) such as resampling, bias removal, and noise removal using a low-pass filter or high-pass filter. In the event that the signal includes data that is representative of patient-reported outcomes, the processing module may code the data into a numeric format and then filter the data (e.g., to remove outlier answers, non-responsive answers, incomprehensible answers, etc.). Moreover, the processing module may perform a binning operation, in which the numerically coded data is sorted into categories or “bins.” As an example, patient-reported outcomes may be sorted into different age ranges (e.g., 0-19, 18-30, 30-45, etc.), geographical locations (e.g., by country, region, state, or county), disease classification (e.g., critical, severe, moderate, and mild), races (e.g., White, Asian, Hispanic, Black, American Indian, Native Hawaiian and Other Pacific Islander), gender (e.g., male, female, and non-binary), and the like.

[0076] For example, at operation 604 shown in Figure 6, the processing module 214 can execute a preprocessing algorithm that includes a sampling algorithm to resample given data (e.g., sensor data or a sensor dataset). A sampling algorithm can include an algorithm designed to re-sample or process data to improve data quality or reduce outliers or noise. For example, a sampling algorithm includes pre-processing data to select a subset of data points (e.g., individual measurements) from a set of measurements. For example, sampling-21- 163869603.8PATENT Attorney Docket No. 124824.8118. WO01algorithms can reduce the size of a dataset (e.g., a sensor dataset or associated movement data) while preserving important features or information. The sampling algorithm can include random sampling, stratified sampling, and / or systematic sampling. Random sampling algorithms can include a sampling algorithm by which each sample (e.g., each measurement) exhibits an equal probability of being chosen for the subset. Stratified sampling can include a sampling algorithm by which measurements are divided into subgroups (e.g., strata) based on similar characteristics prior to subsequent random sampling. Systematic sampling can include a sampling algorithm by which each sample within the subset (e.g., each measurement) is chosen at a regular interval (for example, with respect to time, in the case of an ordered list of temporal data). For example, systematic sampling can include sampling the movement dataset to include data associated with a predetermined temporal frequency (e.g., sampled at a particular period of time). Based on a sampling algorithm, the processing module 214 can generate a resampled dataset (e.g., a resampled movement dataset). By sampling data associated with a sensor (e.g., related to movement of humans), the diagnostic platform can reduce the size and, therefore, the computational burden of evaluated data, thereby improving the efficiency of the system.

[0077] The processing module 214 can execute a preprocessing algorithm that includes bias modification. For example, the preprocessing algorithm can determine a bias associated with the movement data, where the bias can include a skew or a shift in data values. For example, the processing module 214 can detect or determine a bias associated with movement data, such as a constant value added to each measurement within the dataset. In some situations, the bias can be heterogeneous across the dataset. In some implementations, the processing module 214 can modify the resampled movement dataset (or another dataset) to shift values within the dataset to remove the determined bias, thereby generating a shifted dataset.

[0078] The processing module 214 can execute a preprocessing algorithm that includes one or more filters. For example, the processing module 214 can provide data, such as the shifted dataset, to a band-pass filter to filter the dataset (e.g., according to a temporal frequency). For example, the processing module 214 can determine frequencies associated-22- 163869603.8PATENT Attorney Docket No. 124824.8118. WO01with the movement data (e.g., through a discrete Fourier transform or a similar algorithm) and remove patterns within the data associated with particular frequencies or ranges of frequencies. To illustrate, the processing module 214 can subtract signals associated with high frequencies (e.g., noise) from movement data to reduce the variability and noise within the data.

[0079] Based on such pre-processing methods, the processing module 214 can generate preprocessed data 624 at operation 604, as shown in Figure 6. The processing module 214 can, in some implementations, perform pre-processing operations that include a single preprocessing algorithm or a combination of preprocessing algorithms described herein.C. _ Measurement Derivation

[0080] The preprocessed data can then be provided to a data computing component -namely, a computation module (e.g., computation module 216 of Figure 2) - that may be responsible for deriving different measurements based on the preprocessed data (step 503). Note that for some patient-reported outcomes, such as vital signs and symptoms experienced, no further derivation may be necessary.

[0081] In some implementations, the computation module 216 can generate or transform sensor data. For example, the computation module 216 can determine, generate, or identify a step-count dataset based on sensor data (e.g., movement data). The computation module 216 can analyze inertial measurement unit, heartbeat, and / or accelerometer data associated with a sensor (e.g., with a wearable device and / or a mobile device) to estimate the number of steps taken within a given time period (and / or the time at which these steps were taken) and store this information within a step-count dataset. For example, the step-count dataset can include a structured, temporally ordered dataset (e.g., a time-series), wherein each element of the structured dataset includes a number of steps taken by the user within a corresponding period of time. Additionally or alternatively, the computation module 216 receives step-count data directly from the signal collection module (e.g., at operation 501).-23- 163869603.8PATENT Attorney Docket No. 124824.8118. WO01

[0082] The computation module can apply heuristics and / or machine-learning models (or simply “models”) to derive the measurements using the preprocessed data (e.g., the processed sensor dataset). For example, to derive a set of walking-related measurements, the computation module may apply one or more activity classifiers can be applied to windows (e.g., test windows) of preprocessed data that includes values generated by an IMU, such that each window is classified as either ambulatory or non-ambulatory. A similar approach could be used to classify each window as sleeping or non-sleeping, stressed or non-stressed, etc. Each window may be representative of a segment of preprocessed data of predetermined length (e.g., a test time period of 5, 10, or 20 seconds). Meanwhile, each activity classifier may be representative of a model that when applied to a window of preprocessed data, classifies that window based on an analysis of the preprocessed data contained therein. The computation module can then identify walking bouts by joining consecutive ambulatory windows and label ambulatory windows of sufficient length (e.g., 10, 20, or 40 seconds) as walking bouts. For each walking bout, the computation module can derive one or more measurements. Examples of such measurements include step count, cadence, bout duration, arm swing magnitude, arm swing velocity, arm swing acceleration, and arm swing range of motion. To illustrate, the computation module 216 can generate an indication of ambulatory bouts, such as ambulatory bout 626, and / or an indication of non-ambulatory bouts, such as non-ambulatory bout 628, as shown in Figure 6, based on the preprocessed data 624.

[0083] For example, at operation 606 shown in Figure 6, the computation module 216 can determine ambulatory bouts based on step counts that are calculated for every 10-second window (or a window associated with another length of time, such as 5, 15, or 30 seconds) of the pre-processed signal data. The computation module 216 can classify the window as ambulatory if there are greater than (and / or an equal number of) a threshold number of steps within the window, where the threshold number of steps may be adjusted based on human-related or device-related factors. The computation module 216 can join consecutive windows that are classified as ambulatory to determine an ambulatory (e.g., walking) bout. In some implementations, the computation module 216 determines to join these windows in response to a determination that the total length of the joined windows is -24- 163869603.8PATENT Attorney Docket No. 124824.8118. WO01greater than or equal to a threshold time period (e.g., 30 seconds). The computation module 216 can derive different walking measurements, such as a daily number of ambulatory bouts, or a mean ambulatory bout duration.

[0084] The computation module 216 can determine non-ambulatory bouts based on step counts that are calculated for every 10-second window (or a window associated with another length of time, such as 5, 10, 15, or 30 seconds) of the pre-processed signal data. The computation module 216 can classify the window as non-ambulatory if there are fewer than (and / or an equal number of) a threshold number of steps within the window, where the threshold number of steps may be adjusted based on human-related or device-related factors. The computation module 216 can include periods associated with sleep within the classification of non-ambulatory bouts. The computation module 216 can join consecutive windows that are classified as non-ambulatory to determine the length of non-ambulatory bouts. In some embodiments, the computation module 216 determines to join these windows in response to a determination that the total length of the joined windows is greater than or equal to a threshold time period (e.g., 30 seconds). The computation module 216 can derive different non-walking measurements, such as a daily number of non-ambulatory bouts, or a mean non-ambulatory bout duration.

[0085] The nature and number of measurements derived from the preprocessed data will depend on various factors, including the nature of the preprocessed data and intended application of measurements derived therefrom. Consider, for example, walking-related measurements versus sleeping-related measurements. To compute these measurements, the computation module may rely on the same preprocessed data, namely, values generated by an IMU. However, the measurements may differ. Examples of walking-related measurements include step count and cadence as discussed above, while examples of sleeping-related measurements include total sleep duration and wake count.

[0086] Figure 7 depicts an illustration of measurements of ambulatory and non-ambulatory bouts over the course of a day (e.g., schematic 700) and over the course of a week for a patient (e.g., schematics 706 and 708). For example, the computation module 216, at operation 503, enables the determination of time periods that correspond to non--25- 163869603.8PATENT Attorney Docket No. 124824.8118. WO01walking bouts (e.g., time period 702) and / or walking bouts (e.g., time period 704). To illustrate, the computation module 216 can determine, for each window within a step-count dataset, a number of steps performed within the given window. The computation module 216 can compare the number of steps with a threshold number of steps (e.g., a step-count parameter). For example, in situations where a given window includes a number of steps greater than the step-count parameter, the computation module 216 can determine that the window corresponds to a walking bout. Additionally or alternatively, where a given window includes a number of steps equal to or less than the step-count parameter, the computation module 216 can determine that the window corresponds to a non-walking bout. Adjacent windows with the same bout type (e.g., adjacent non-walking bout-type windows, or adjacent walking bout-type windows) can be combined to determine the time period 702 or the time period 704, as shown in Figure 7.

[0087] In some embodiments, the computation module 216 can determine walking or non-walking bouts (and associated time periods) over a variety of time periods. For example, the schematic 706 illustrates periods of time throughout a week corresponding to ambulatory bouts, while the schematic 708 illustrates periods of time throughout the week corresponding to non-ambulatory bouts.

[0088] In some embodiments, the computation module 216 can determine ambulatory or non-ambulatory bouts based on sleep data associated with the human. For example, the computation module 216 can determine whether a human is on a bed or off a bed based on an on-bed classification algorithm. The computation module 216 can combine information relating to ambulatory bouts and sleeping bouts to determine non-ambulatory bouts (e.g., time periods in which the human is not walking).

[0089] Figure 8 depicts an illustration 800 of measurements of ambulatory and non-ambulatory bouts over the course of a day based on the availability of first and last stepcount data for a given day. For example, the computation module 216 can determine, based on the sensor data, a first time at which step-count data becomes available (e.g., when the associated human begins wearing a smartwatch in the morning), as well as a second time at which step-count data becomes unavailable (e.g., when the associated human removes-26- 163869603.8PATENT Attorney Docket No. 124824.8118. WO01the smartwatch prior to sleeping). As such, the computation module 216 can determine walking and / or non-walking bouts between the first time and the second time, while discounting data from outside of these times.

[0090] Based on the determination of ambulatory and non-ambulatory bouts associated with the preprocessed signal data, the computation module 216 can further derive free-living digital measurements (e.g., at operation 608 shown in Figure 6). For example, the computation module 216 can determine continuity metrics or indicators that are indicative of bout duration 630, 632 and / or consistency over time. For example, generated continuity metrics can provide behavioral information (e.g., continuity) regarding a walking or non-walking pattern associated with a human. Continuity measurements can Gini measure, non-ambulatory / ambulatory percentage metrics, and / or skewness metrics associated with bout duration.

[0091] Figure 9A includes illustrations 900 of continuity metrics associated with a relatively continuous non-walking pattern 902. Figure 9B includes illustrations 920 of continuity metrics associated with a less continuous non-walking pattern 922.

[0092] For example, the computation module 216, at operation 503, can determine Gini measures associated with a given human. A Gini measure can include a measure of the statistical dispersion of bouts with respect to bout length. For example, the computation module 216 can calculate a Gini value that indicates a relationship between a cumulative distribution of a set of ambulatory bouts and corresponding percentiles of the ambulatory bouts. To illustrate, the Gini value can represent, mathematically, a value associated with a Lorenz-type plot, as shown in plot 904 of Figure 9A (e.g., for a more continuous non-walking pattern) and plot 924 of Figure 9B (e.g., for a less continuous non-walking pattern). The Lorenz-type plot can represent a plot of the proportion of the total length of ambulatory bouts (on a virtual vertical axis) that is cumulatively associated with a given percent of the considered ambulatory bouts (associated with the associated virtual horizontal axis). Furthermore, on the same plot, a line indicative of equality (e.g., where all ambulatory bouts are the same length) can be plotted. The Gini measure can include a ratio of the area that lies between the line indicative of equality and the Lorenz-type plot over the total area under-27- 163869603.8PATENT Attorney Docket No. 124824.8118. WO01the line indicative of equality. As such, the measure indicates how dispersed or disparate bout lengths associated with a given human are, where a higher measure indicates that the total bout time is mainly due to a very small proportion of long walking bouts (e.g., that the bout lengths are generally variable). Alternatively, a lower measure (e.g., closer to 0) indicates that the majority of walking bouts are short and all lengths of bouts contribute more equally to the total time, indicating a more fragmented pattern of bout accumulation.

[0093] In some embodiments, the computation module 216, at operation 503, can determine a non-ambulatory / ambulatory percentage metric (e.g., a minute-to- % time) associated with the bouts (e.g., non-walking bouts). For example, the non-ambulatory percentage metric can include a metric indicative of a bout duration associated with a predetermined (e.g., adjustable) fraction of non-ambulatory bouts with a duration less than the bout duration. For example, the measure can be defined as a bout duration associated with a K% of the day a human spends not walking that comes from non-walking bouts that are shorter than the bout duration, where K can be adjusted or tuned to improve predictability of disease progression indicators. As such, a smaller non-ambulatory percentage metric can include an indication that non-walking bouts are less continuous, indicating that the nonwalking bouts are mostly of short duration. To illustrate, the more continuous non-walking pattern 902 of Figure 9A, with many longer non-walking bouts, can indicate a larger minute-to-80% time, while the less continuous non-walking pattern 922 of Figure 9B, with few longer non-walking bouts, can indicate a lower minute-to-80% time metric.

[0094] In some embodiments, the computation module 216, at operation 503, can determine a skewness metric associated with the bouts (e.g., non-walking bouts). For example, the computation module 216 can determine a statistical measure of skewness for a bout duration distribution (e.g., non-ambulatory and / or ambulatory). For example, skewness can include an indication of a difference between a mean and a median bout duration value, where the difference is divided by the distribution’s standard deviation. For example, a skewness value of -0.5 to 0.5 can indicate a fairly symmetrical distribution of bout duration. For example, plot 906 of Figure 9A demonstrates that the more continuous non-walking pattern 902 indicates a lower skewness value. A skewness value of -1 to -0.5-28- 163869603.8PATENT Attorney Docket No. 124824.8118. WO01or 0.5 to 1 can indicate a moderately skewed bout duration distribution. A skewness value of greater than 1 or less than -1 can indicate a highly skewed bout duration distribution. For example, plot 926 of Figure 9B demonstrates that the more continuous non-walking pattern 922 indicates a greater skewness value.

[0095] Such continuity metrics associated with non-ambulatory bouts can be indicative of the health of a given human. For example, progressive diseases, such as Parkinson’s disease, can affect the mobility of humans. However, the relationship between continuity metrics and non-walking periods of time are not immediately apparent, as bout continuity is not expected to be directly proportional to or indicative of mobility changes associated with progressive diseases.D. Analytics

[0096] Thereafter, another data computing component - namely, an analytics module (e.g., analytics module 218 of Figure 2) - can perform aggregation (e.g., at operation 504 shown in Figure 5 and / or operation 610 shown in Figure 6) and progression analysis (operation 505). As shown in Figure 5, the analytics module can implement an analytics engine 500 in order to perform aggregation and progression analysis. At a high level, the analytics engine 500 (or simply “engine”) is representative of the core logic of the analytics module that, in operation, takes an input (e.g., the preprocessed data), performs at least one operation, and then produces an output (e.g., a data structure with data that is temporally aggregated). More specifically, the analytics engine 500 may be responsible for aggregating continuity metrics and other derived measurements on a daily basis using mathematical operators, as well as for identifying features and trends in continuity metrics associated with disease progression.

[0097] The analytics engine 500 can include two programs that, in operation, are performed in sequence. The first program may provide a method for programmatically aggregating data included in the signal collected as input. The second program may provide multiple methods for determining whether a given measurement is indicative of a digital-29- 163869603.8PATENT Attorney Docket No. 124824.8118. WO01progression biomarker (e.g., a statistical indicator for progressive diseases based on the continuity of non-ambulatory bouts).1. Aggregation

[0098] The analytics engine 500 can generate daily aggregated data associated with measurements or distributions of bouts (e.g., non-ambulatory or ambulatory) at operation 504. For example, the analytics engine 500 can generate aggregated values, using mathematical operators, for bout durations on a daily level or with respect to another time (e.g., daily, weekly, monthly, yearly, etc.). As an illustrative example, the analytics engine 500 generates, using a maximum operator, a maximum bout duration (e.g., a bout duration for the longest bout within a given day) for each day within a set of consecutive days. In some embodiments, the analytics engine 500 generates values associated with minimum values, or bout durations associated with particular percentiles (e.g., a 95thpercentile). By aggregating data, the analytics module 218 enables standardization of any inconsistencies in temporal data associated with the signal data.II. Progression Analysis

[0099] The analytics engine 500 can execute (e.g., at operation 505 shown in Figure 5 or at operation 612 shown in Figure 6) associated progression analyses with the derived measurements to determine whether such measurements are indicative of disease progression (e.g., are associated with digital progression biomarkers). For example, the analytics engine 500 can determine whether a given measurement fulfills one or more of a set of criteria. The criteria can include a test-retest reliability criterion, a correlation criterion, and / or a progression criterion. In some implementations, the analytics engine 500 determines a given measurement’s satisfaction of these three criteria in order to determine that the measurement can be indicative of disease progression or presence. Alternatively or additionally, the analytics engine 500 determines the given measurement’s satisfaction of a subset of criteria in order to determine that the measurement is indicative of disease progression or presence.-30- 163869603.8PATENT Attorney Docket No. 124824.8118. WO01

[0100] For example, the analytics engine 500 can determine whether measurements (e.g., an aggregated time series of metrics) satisfy a test-retest reliability criterion associated with the aggregate time period. To illustrate, the analytics engine 500 can determine a test-retest reliability of 30-day (or another value, such as 2-week or 4-week) aggregated metric data and compare this test-retest reliability to a threshold value (e.g., 0.75). For example, the analytics engine 500 can determine that the test-retest reliability is greater than the threshold value and, therefore, that the aggregated data satisfies a first criterion.

[0101] In some embodiments, the analytics engine 500 can quantify the test-retest reliability based on determining an intra-class correlation (ICC) metric between adjacent aggregation periods. For example, the analytics engine 500 can determine whether the test-retest reliability is sufficient (e.g., exceeds a threshold value) within a given daily aggregation period of a certain length (e.g., 30 days) based on an analysis of plots 634. For example, Figure 10A shows a plot 1000 of intraclass correlation (ICC) values against aggregation days for a specific digital measure. For example, the plot 1000 demonstrates ICC values against different aggregation days on the horizontal axis for a specific digital measure (e.g., the 95thpercentile daily walking bout duration). Each line represents the range of ICC values (associated with the 95thpercentile confidence interval), where the 95thpercentile daily walking bout duration is aggregated in the n-day manner (where n indicates days of the horizontal axis). The horizontal line indicates the 0.75 ICC threshold value. The point estimate of test-retest reliability achieved a value greater than the threshold value at 30 days, thereby fulfilling the test-retest reliability criterion in this example.

[0102] For example, Fisher’s original formulation of ICC may be used to estimate test-retest reliability as set forth below in Eq. 1-3. To compute a given ICC, the data may need to be organized into a pair of vectors ( xn l, xn 2) having columnar form, as shown in Figures 9A-C. These vectors can be constructed as set forth above.r = ^ln=i ^ (xnil- x)(xni2- x), Eq. 1 whereX = ^ln=i > (xn,i + xn,2~), and Eq. 2-31- 163869603.8PATENT Attorney Docket No. 124824.8118. WO01S2= {2n=i > (%n,l - X)2+ Sn=l On, 2 “ X)2}. Eq. 3

[0103] For each reliability value, both the point estimate and confidence interval (Cl) of the ICC may be provided. The point estimate can be obtained by computing the ICC from the observed data, while the Cl can be computed using a bootstrapping method. Specifically, the aggregation module can resample the observed data with replacement being performed m times, where m is an integer value (e.g., between 100 and 2,500, and preferably ~1,000). For each set of resampled data, the value of the ICC for that given sample can be obtained, and in total, m number of ICCs may be obtained. A (100 - a) percent Cl can be obtained by taking the a / 2 and (100 - a / 2) percentiles of the resultant m ICCs.

[0104] In some embodiments, the analytics engine 500 can determine whether measurements (e.g., an aggregated time series of metrics) satisfy a correlation criterion (e.g., a criterion corresponding to a correlation between a metric and the set of clinical measures). For example, the analytics engine 500 can determine a Spearman’s rank correlation coefficient between the measurement and a clinical measure (e.g., associated with disease progression) at different time points (e.g., corresponding to the availability of clinical data). The analytics engine 500 can determine whether a confidence interval (e.g., at a value, such as the 95thpercentile) of the correlation coefficient excludes zero, thereby indicating correlation between the given free-living measurement-derived measure and the clinical measure. For example, Figure 10B shows a plot 1020 of correlation between a specific digital measure and a gait score. For example, Figure 10B shows an example of the correlation between the monthly 95thpercentile daily bout duration and the Unified Parkinson’s Disease Rating Scale (UPDRS) part 3.10 Gait Score, demonstrating a Spearman’s rank correlation of 0.28 (e.g., indicating a weak correlation).

[0105] In some embodiments, the analytics engine 500 can determine whether measurements (e.g., an aggregated time series of metrics) satisfy a progression rate criterion. For example, the progression rate can be quantified by the digital measure for participants with a progressive disease (e.g., Parkinson’s disease), as well as for participants without a progressive disease (e.g., healthy participants). Based on a comparison of progression rates, the analytics engine 500 can determine whether a given trend in digital -32- 163869603.8PATENT Attorney Docket No. 124824.8118. WO01measurements indicates a change in the expected direction given known symptoms and / or research associated with the diseased or healthy participants. The analytics engine 500 can determine satisfaction of the progression rate criterion based on a comparison between a first subset of measurements associated with humans with a progressive disease and a second subset of measurements associated with humans that are not associated with a progressive disease.

[0106] For example, the analytics engine 500 can leverage a linear mixed effects model to estimate the difference of an average monthly progression rate (or, for example, with respect to another temporal period) in a digital measure (e.g., aggregated metric time series data) between participants with progressive disease and the healthy comparator (e.g., the healthy participants). For example, the model can include a change from baseline in the digital measure at each post-baseline clinical visit as an outcome. The model can incorporate the fixed effects of time, time and cohort interaction, and other factors, such as age, sex, and time-varying effects of seasonality and / or public safety lockdowns (e.g., in response to a pandemic or epidemic).

[0107] For example, for a given participant i = 1,..., N and post-baseline aggregation visit time j = l,...,n the analytics engine 500 can determine a change from baseline (CFBij) indicating a difference in the associated measure from a baseline month Fi7- Yu. For example, the analytics engine can utilize the following formula for determination of the change from baseline:CFBij = ft * timetj +?2* cohort^ * timeij + (3 * covariateSij + bi0-I- timetj * btl+ ei7where CFBtj= Ytj- Ytlis the change from baseline (CFB) at month j, cohort^ is a binary indicator for cohort, i.e., 0 for HC and 1 for PD, (3 = ( / ?i,..., f3p) is a p x 1 column vector of the fixed-effect coefficients, time is treated as a continuous variable, covariates include a list of confounding factors such as age at baseline, sex, and time-varying factors such as seasonality, COVID lockdown type (i.e., Iockdown=strict, moderate, mild, or no lockdown, which can be an ordinal variable), and bt - bt0,bt^) is the random intercept and random slope of time respectively, and assumed to follow a multivariate normal distribution-33- 163869603.8PATENT Attorney Docket No. 124824.8118. WO01with mean zero and covariance matrix G. Specifically, bi0and btlare the participant-specific intercept and slope for each participant, i. For example, et= (eiy) is a ntx 1 column vector of the residuals and assumed to follow a multivariate normal distribution with mean zero and covariance matrix ln.o2.

[0108] In some embodiments, the parametercorresponds to the monthly rate of change in the real-life measure in the healthy comparator cohort, while+ / ?2corresponds to the monthly rate of change in the progressive disease cohort. For estimating the model parameters, restricted maximum likelihood (REML) can be used. The covariance matrix for the random intercept and random slope can be assumed to be unstructured.

[0109] For example, Figure 10C includes an illustration 1040 of plots 1042 and 1044 of a specific digital measure over time for a group associated with progressive disease and a control group, respectively. Utilizing hypothesis testing, the analytics engine 500 can determine whether a given trend in digital measures is statistically significant by comparing the measures associated with humans known to be associated with progressive disease with measures associated with humans known to be healthy. For example, plot 1042 indicates the 95thpercentile daily bout duration over 2 years of time as associated with the progressive disease cohort (e.g., demonstrating a statistically significant trend in the associated digital measure over time compared to the healthy comparator cohort). In comparison, plot 1044 indicates the 95thpercentile daily bout duration over 2 years of time as associated with the healthy comparator cohort (e.g., demonstrating no trend in the associated digital measure).

[0110] As such, criteria, such as the test-retest reliability criterion, the Spearman’s rank correlation criterion, and / or the progression rate criterion enable determination of the nature and significance of relationships between the existence and / or progression of progressive disease within humans on the basis of free-living data (e.g., as associated with wearable devices), thereby elucidating relationships between, for example, non-ambulatory bout duration / continuity and associated progressive disease, which are not immediately predictable or apparent.-34- 163869603.8PATENT Attorney Docket No. 124824.8118. WO01E. Determination of Disease Progression Indicators

[0111] Based on the satisfaction of one or more criteria described above, the diagnostic platform 212 can generate indications or determinations of disease progression indicators 636 (e.g., at operation 506). These disease progression indicators 636 may also be called “progression biomarkers.” For example, the diagnostic platform 212 can determine that a given continuity metric (e.g., of a particular type) is indicative of a disease for a human. In some embodiments, the diagnostic platform 212 can determine that metrics can be representative of markers that can be used as an indicator of presence or progression of a disease (e.g., generally, in one or more humans). For example, in some embodiments, the diagnostic platform 212 can generate a diagnosis message indicating a probability that a given patient is associated with progressive disease (e.g., through a hypothesis testing method).

[0112] Based on determination of disease progression associated with a human, the diagnostic platform 212 can generate, determine, obtain, or retrieve a treatment plan associated with the disease. The treatment plan can include an indication of therapies, treatments, medications, surgeries, or lifestyle suggestions for the human based on an indication that the human likely is associated with a progressive disease (e.g., Parkinson’s disease).

[0113] In some embodiments, based on the determination of the disease progression indicator, the diagnostic platform can perform other actions associated with the human. For example, the diagnostic platform can store an indication of the associated metric or measure in a digital profile associated with the human, so as to maintain a record of whether the indicator is present over time. By doing so, the diagnostic platform enables progression to be monitored in a more quantifiable manner, as well as enabling degradation (e.g., disease progression) to be identified more easily, leading to prompt treatment.

[0114] In some embodiments, the diagnostic platform can generate a message for transmission to a device associated with a healthcare professional or the human (e.g., an alert for display on a user interface). For example, the message can include an alert suggesting further examination by a healthcare professional.-35- 163869603.8PATENT Attorney Docket No. 124824.8118. WO01

[0115] In some embodiments, the diagnostic platform can stratify the human for examination or treatment. For example, the diagnostic platform can generate a metric that indicates an urgency level for examination of the human (e.g., independent of or relative to other humans).

[0116] In some embodiments, the diagnostic platform can assign a human to a given treatment regimen, including physical activity, medication, or to a treatment plan associated with a corresponding patient cohort (e.g., corresponding to an associated disease severity level).

[0117] As such, by evaluating a relationship between digital, free-living measurementbased derived metrics and disease progression indicators, the diagnostic platform 212 enables the generation of a framework for identifying and tracking disease progression over time in a manner that supplements clinical evaluations. For example, clinical evaluations rely on self-reported information from patients associated with symptoms that occur between clinical assessments. The systems and methods disclosed herein enable dynamic tracking and supplementation of clinical information with free-living information associated with data between these assessments, thereby improving the accuracy and efficiency of disease progression detection (e.g., for Parkinson’s disease patients).Methodologies for Identifying Disease Progression Biomarkers

[0118] Several approaches to identifying disease progression biomarkers based on free-living measurements are set forth below. These approaches are best understood when read in conjunction with the disclosure corresponding to Figures 5-10.

[0119] Figure 11 includes a flow diagram of a process 1100 for analyzing step-count data to determine ambulatory and non-ambulatory windows associated with determining disease progression.

[0120] At operation 1101, the diagnostic platform can obtain a step-count dataset associated with a first human. For example, the diagnostic platform can obtain a step-count dataset (or other suitable data) that includes a time series of step-count data associated with a first human of a set of humans. As an illustrative example, the diagnostic platform can-36- 163869603.8PATENT Attorney Docket No. 124824.8118. WO01receive sensor data associated with a wearable device, such as a watch, and determine a measure of steps taken by the first user over time (e.g., a number of steps within each window of a set of time windows). By receiving such information, the diagnostic platform enables tracking of activity data associated with the first human, thereby providing information relating to the human’s health for further identification and / or diagnosis of characteristics of the human.

[0121] In some embodiments, the diagnostic platform obtains a longitudinal dataset that includes values arranged in temporal order. For example, each value of the longitudinal dataset can be generated by a sensor in a computing device associated with a human at a corresponding point in time (e.g., a wearable device, such as a smartwatch). In some embodiments, the diagnostic platform can segment the longitudinal dataset into a series of datasets, each of which is associated with a corresponding one of a series of windows. As such, the diagnostic platform can analyze human activity measured by a sensor associated with the first human over time, thereby enabling analytics of the human’s behavior over time.

[0122] In some embodiments, the diagnostic platform can pre-process the longitudinal dataset to improve its quality. For example, the diagnostic platform can perform preprocessing algorithms including resampling, bias removal, and / or filtering. The diagnostic platform can generate, based on the longitudinal dataset, a resampled dataset for the human, where the resampled dataset includes data associated with a predetermined temporal frequency. The diagnostic platform can determine a bias associated with the longitudinal dataset. Based on the bias, the diagnostic platform can modify the resampled dataset to generate a shifted dataset. The diagnostic platform can provide the shifted dataset to a filter to generate a filtered dataset. The diagnostic platform can update the longitudinal dataset to include the filtered dataset. By performing such measures, the diagnostic platform improves the quality of step-count data and other measurements derived from sensor data by reducing the noise and other undesirable artifacts within the longitudinal data.

[0123] At operation 1102, the diagnostic platform can generate a set of ambulatory windows and a set of non-ambulatory windows. For example, the diagnostic platform can generate, based on the step-count dataset and a step-count parameter, a set of ambulatory-37- 163869603.8PATENT Attorney Docket No. 124824.8118. WO01windows and a set of non-ambulatory windows. As an illustrative example, the diagnostic platform determines periods of time for the first human likely associated with non-walking activities (e.g., sleeping or sitting on a chair), as well as periods of time for the first human likely associated with walking-related activities or other ambulatory activities. For example, the diagnostic platform can determine whether the step count within a given window is greater than or less than (or equal to) a threshold step count. Based on this determination, the diagnostic platform can classify the window as ambulatory or non-ambulatory. Doing so enables the diagnostic platform to evaluate the first human’s behavior (and, for example, possible disease progression) based on data associated with the duration and / or continuity of walking or non-walking bouts, thereby providing information unavailable to medical practitioners during clinical assessments.

[0124] In some embodiments, the diagnostic platform can determine the ambulatory and non-ambulatory windows by comparing numbers of steps with threshold step counts. For example, for each window in the series of windows, the diagnostic platform can determine a number of steps taken by the human during a corresponding one of the series of datasets, and establish that the window is either (i) representative of a non-ambulatory period in response to a determination that the number of steps is less than or equal to a threshold, or (ii) representative of an ambulatory period in response to a determination that the number of steps is more than the threshold.

[0125] In some embodiments, the diagnostic platform can identify non-ambulatory and ambulatory bouts based on the classification of the windows. For example, the diagnostic platform can identify (i) non-ambulatory bouts that correspond to windows, if any, established as representative of non-ambulatory periods and (ii) ambulatory bouts that correspond to windows, if any, established as representative of ambulatory periods. As an illustrative example, the diagnostic platform can determine to join adjacent ambulatory windows and adjacent non-ambulatory windows if such joined windows represent a time period greater than a threshold time period. By doing so, the diagnostic platform can generate an indication of lengths of time that characterize the human’s ambulatory and non--38- 163869603.8PATENT Attorney Docket No. 124824.8118. WO01ambulatory bouts, thereby providing information relating to the continuity of such bouts for further analysis and evaluation.

[0126] In some embodiments, the diagnostic platform can utilize sleeping data to determine non-ambulatory bouts and non-ambulatory bouts. For example, the diagnostic platform can obtain sleeping data associated with the human. The diagnostic platform can determine sleeping time periods based on the sleeping data. The diagnostic platform can identify the ambulatory bouts to be associated with time periods not associated with the sleeping time periods and windows representative of non-ambulatory periods. As an illustrative example, the diagnostic platform can determine, using an on-bed classification algorithm, whether the human is on a bed or off a bed in a given window of time based on the associated sensor data. Based on this determination, the diagnostic platform can classify these windows as non-ambulatory bouts, for example, thereby classifying other windows as ambulatory. By doing so, the diagnostic platform improves the accuracy of determinations of ambulatory and non-ambulatory bouts associated with sensor data.

[0127] At operation 1103, the diagnostic platform can generate a metric that is indicative of the continuity of bout duration. For example, the diagnostic platform can generate, based on the set of ambulatory windows or the set of non-ambulatory windows, a metric that is indicative of continuity of bout duration. As an illustrative example, the diagnostic platform can generate metrics associated with the maximum bout duration of ambulatory or non-ambulatory bouts associated with the human. In some embodiments, the diagnostic platform can generate continuity metrics associated with the distribution of bout durations, such as a Gini metric, a skewness metric, and / or a time-percentage characterization of the bout durations. As such, by considering the distribution of non-ambulatory and ambulatory bouts, the diagnostic platform enables the generation of information generally not available to healthcare professionals (e.g., information relating to the distribution of walking and non-walking time-periods associated with a given human).

[0128] In some embodiments, the metric can include walking measurements associated with the number or duration of ambulatory bouts. For example, the diagnostic platform can generate a set of walking measurements based on the ambulatory bouts. The-39- 163869603.8PATENT Attorney Docket No. 124824.8118. WO01set of walking measurements can include at least one of (1) a daily number of ambulatory bouts or (2) a mean ambulatory bout duration. The diagnostic platform can determine, based on the set of walking measurements, that the metric is indicative of the disease. As an illustrative example, diagnostic platform can characterize periods of time where the human is determined to be walking based on information relating to ambulatory bouts, thereby improving the information associated with the human’s behavior for further evaluation with respect to disease progression indicators.

[0129] In some embodiments, the ambulatory bout measurement can include other characterizations of walking bouts. For example, the diagnostic platform can generate an ambulatory bout measurement based on the longitudinal dataset. For example, the ambulatory bout measurement includes at least one of (1) a step count, (2) a number of bouts, (3) a cadence, (4) a bout duration, (5) an arm swing magnitude, or (6) an arm swing range of motion. The diagnostic platform can determine, based on the ambulatory bout measurement, that the metric is indicative of the disease. As an illustrative example, the diagnostic platform can characterize time periods during which the human may be walking using a variety of different metrics, thereby improving the quality and breadth of information used in evaluating the health of the individuals.

[0130] In some embodiments, the diagnostic platform can generate continuity measurements based on a statistical representation of the distribution of bout durations associated with non-ambulatory (and / or ambulatory) bouts. For example, the diagnostic platform can generate a metric that is an indication of (i) a Gini measure, (ii) a nonambulatory percentage metric, or (iii) a skewness for a bout duration distribution associated with the set of ambulatory windows and the set of non-ambulatory windows. To illustrate, the diagnostic platform can generate continuity metrics that characterize the distribution, consistency, and / or variability in the duration of walking or non-walking bouts associated with the human. By considering such information, the diagnostic platform captures information absent from a clinical evaluation of humans, by including information that is not easily describable or measurable by non-free-living measurements or surveys. As such, the-40- 163869603.8PATENT Attorney Docket No. 124824.8118. WO01diagnostic platform can leverage such information to improve the diagnosis, identification, or characterization of progressive diseases, such as Parkinson’s disease.

[0131] In some embodiments, the diagnostic platform can aggregate the measurements according to a determined time period (e.g., on a daily basis). For example, the diagnostic platform can generate, based on the set of ambulatory windows or the set of non-ambulatory windows, a set of continuity metrics corresponding to a set of time periods. The diagnostic platform can apply one or more mathematical operators to the set of continuity metrics to generate an aggregated metric time series. Each aggregated metric of the aggregated metric time series can be associated with a daily time period. The diagnostic platform can perform a progression analysis of the aggregated metric time series to determine that the first human is exhibiting an indicator of the disease. As an illustrative example, the diagnostic platform can generate a characterization or summary of bout duration measurements or other movement-related measurements at a pre-determined periodicity of time. For example, the diagnostic platform can generate metrics associated with non-walking and / or walking bouts for the human that characterize the human’s behavior on a daily, weekly, or monthly basis. By doing so, the diagnostic platform can smooth out smaller-scale variations in the human’s behavior, thereby improving the accuracy of further analyses associated with this data.

[0132] In some embodiments, the mathematical operator can include a minimum, maximum, or a percentile operator. For example, the diagnostic platform can apply a minimum operator, maximum operator, or pre-determined percentile operator to the set of continuity metrics to generate the aggregated metric time series. As an illustrative example, the diagnostic platform can generate an aggregated metric for each day in a given time period, where the aggregated metric includes a bout length for the longest bout associated with the given day. Additionally or alternatively, the aggregated metric includes a bout length for a bout at a given percentile with respect to bout lengths (e.g., the 95thpercentile, the 5thpercentile, or another suitable percentile). By doing so, the diagnostic platform can characterize bouts on a periodic basis to generate uniform data, thereby improving the quality of evaluations of the human’s health based on the aggregated metric time series.-41- 163869603.8PATENT Attorney Docket No. 124824.8118. WO01

[0133] In some embodiments, the diagnostic platform generates a metric (e.g., a continuity metric) based on a Gini value. For example, the diagnostic platform generates a Gini value indicative of a continuity of ambulatory bouts. The Gini value indicates a relationship between a cumulative distribution of a set of ambulatory bouts and corresponding percentiles of the ambulatory bouts in the set of ambulatory bouts. The diagnostic platform generates the metric based on the Gini value. As an illustrative example, the diagnostic platform can generate a metric that indicates whether the total bout time within a given time period (e.g., within a day) is due to a small proportion of longer bouts (e.g., walking or non-walking), or whether the total bout time is more fragmented (e.g., due to a more equal distribution of bouts throughout the day). As such, the Gini value can characterize the continuity and consistency of bouts and bout durations, thereby providing information that can supplement and improve the diagnosis and characterization of progressive diseases.

[0134] In some embodiments, the diagnostic platform generates the metric based on a a time metric associated with a percentile or percentage associated with the distribution of bout duration (e.g., a percentage metric that corresponds to a time, such as a minute-to-K%-time metric). For example, the diagnostic platform can generate a non-ambulatory time metric indicative of a bout duration associated with a pre-determined fraction of nonambulatory bouts with a duration less than the bout duration. The diagnostic platform can generate the metric based on the non-ambulatory time metric. For example, the diagnostic platform can characterize the continuity of non-walking bouts. As an illustrative example, a smaller non-ambulatory time metric can indicate a less continuous pattern of non-walking bouts. As such, the time metric can be indicative of the health of a human based on a characterization of non-ambulatory periods of time associated with the human.

[0135] In some embodiments, the diagnostic platform generates the metric based on a skewness metric. For example, the diagnostic platform generates a skewness for a bout duration distribution associated with the set of ambulatory windows and the set of non-ambulatory windows. The diagnostic platform can generate the metric based on the skewness. As an illustrative example, the diagnostic platform can generate a metric that-42- 163869603.8PATENT Attorney Docket No. 124824.8118. WO01indicates a difference between the mean and median of a distribution of bout lengths (e.g., normalized by the standard deviation). By generating such a value, the diagnostic platform can determine information associated with the symmetry of the distribution of bout lengths, thereby characterizing the continuity and distribution of bouts associated with the human. As such, the skewness can provide information associated with the human’s health that is not easily captured in periodic clinical evaluations.

[0136] At operation 1104, the diagnostic platform can determine whether the human is exhibiting an indicator of a disease. For example, the diagnostic platform can determine, based on the metric, whether the first human is exhibiting an indicator of a disease. The indicator of the disease can include a trend in bout duration for the first human. As an illustrative example, the diagnostic platform can calculate a trend in the continuity metric and determine whether this trend is likely associated with a progressive disease (e.g., Parkinson’s disease). As such, the diagnostic platform can leverage free-living data from wearable devices, such as smartwatches, to evaluate humans for progressive diseases, thereby incorporating data not readily accessible to healthcare professionals.

[0137] At operation 1105, the diagnostic platform can perform an action based on the indicator of the disease. For example, the diagnostic platform can perform an action based on whether the first human is determined to exhibit the indicator of the disease. As an illustrative example, the diagnostic platform can generate a warning message or other data for transmission to another entity based on determining that a human’s disease is progressing. For example, the diagnostic platform can recommend a visit to a healthcare professional to the human through a message transmitted to the human’s wearable device or another associated device (e.g., a smartphone). By doing so, the diagnostic platform enables the human to receive earlier preventative care, even prior to any clinical evaluations for disease progression.

[0138] In some embodiments, the diagnostic platform can generate or obtain a treatment plan for the human. For example, the diagnostic platform can obtain, for the human, a treatment plan associated with the disease. The diagnostic platform can transmit, for display on the computing device, a representation of the treatment plan. For example,-43- 163869603.8PATENT Attorney Docket No. 124824.8118. WO01the diagnostic platform can determine a therapy regimen, a diet, a medication plan, or another set of instructions or suggestions associated with treating the detected progressive disease. For example, the diagnostic platform can classify the continuity metrics and the disease indicator to determine a disease type, retrieve data (e.g., patient data or health records) associated with the human, and / or obtain a treatment plan for transmission to the human. As such, the diagnostic platform enables preventative actioin and intervention to alleviate progressive disease symptoms on the basis of free-living digital measurements.

[0139] Figure 12 includes a flow diagram of a process 1200 for correlating clinical measures of disease progression with non-ambulatory bout data associated with free-living scenarios.

[0140] At operation 1201, the diagnostic platform can obtain a set of aggregated metric time series for a set of humans. For example, each aggregated metric time series of the set of aggregated metric time series is associated with a corresponding one of the set of humans. Each aggregated metric time series of the set of aggregated metric time series can include at least one metric that characterizes, for the corresponding human, a set of associated ambulatory windows and / or a set of associated non-ambulatory windows. As an illustrative example, the diagnostic platform can generate metrics (e.g., from preprocessed data from wearable devices) that indicate the continuity or otherwise characterize the nature of non-ambulatory and / or ambulatory bouts associated with a set of humans. The set of humans can include humans associated with progressive disease (e.g., patients known to be afflicted with Parkinson’s disease), as well as humans known not to be associated with progressive disease (e.g., healthy comparators). As such, the aggregated data can provide free-living digital measurements that are not readily available during a conventional clinical examination of patients.

[0141] At operation 1202, the diagnostic platform can retrieve a set of clinical measures associated with the set of humans. For example, the diagnostic platform can retrieve information associated with humans diagnosed with a progressive disease and / or humans that are not diagnosed with a progressive disease (e.g., a health comparator cohort and a progressive disease cohort). This information enables the evaluation of the relationship-44- 163869603.8PATENT Attorney Docket No. 124824.8118. WO01between the characterization of non-ambulatory or ambulatory bouts and the existence or presence of progressive disease.

[0142] At operation 1203, the diagnostic platform can determine whether the set of aggregated metric time series satisfies a set of criteria. For example, the diagnostic platform can determine whether the set of aggregated metric time series satisfies a set of criteria by establishing whether the set of aggregated metric time series fulfills (i) a test-retest reliability criterion associated with an aggregate time period, (ii) a correlation criterion corresponding to correlation between the at least one metric and the set of clinical measures, and (iii) a progression rate criterion. The progression rate criterion may be representative of an aggregated metric time series between humans with and without progressive disease.

[0143] In some embodiments, the diagnostic platform can determine whether the set of aggregated metric time series satisfies a subset of these criteria. The progression rate criterion can correspond to a comparison between (a) a first subset of aggregated metric time series associated with humans of the set of humans associated with a progressive disease, and (b) a second subset of aggregated metric time series associated with humans of the set of humans not associated with the progressive disease. To illustrate, the diagnostic platform can determine whether the aggregated metric time series (or, for example, other digital data derived from free-living measurements) corresponds to indicators of progressive disease or trends thereof, thereby enabling the diagnostic platform to evaluate the metrics more accurately.

[0144] In some embodiments, the diagnostic platform can determine whether the aggregated metric time series satisfies a test-retest reliability criterion. For example, the diagnostic platform generates, according to a pre-determined daily aggregation period, the set of aggregated metric time series. The diagnostic platform can determine an intra-class correlation metric between adjacent periods within the set of aggregated metric time series. The diagnostic platform can compare the intra-class correlation metric with a threshold correlation metric to determine that the intra-class correlation metric exceeds the threshold correlation metric. Based on determining that the intra-class correlation metric exceeds the threshold correlation metric, the diagnostic platform can establish that the set of aggregated-45- 163869603.8PATENT Attorney Docket No. 124824.8118. WO01metric time series fulfills the test-retest reliability criterion. As an illustrative example, the diagnostic platform can determine the reliability of the diagnostic metrics to ensure that any determined indicators of disease are consistent over time. As such, the diagnostic platform improves the reliability of evaluations for disease progression.

[0145] In some embodiments, the diagnostic platform can determine a correlation between clinical measures and the continuity metrics, such as through a Spearman’s rank correlation coefficient. For example, the diagnostic platform can generate, between the set of aggregated metric time series and the set of clinical measures, a Spearman’s rank correlation coefficient and a corresponding confidence interval. Based on determining that the corresponding confidence interval excludes a threshold correlation coefficient, the diagnostic platform can establish that the set of aggregated metric time series fulfills the correlation criterion. As an illustrative example, the diagnostic platform can utilize clinical measures to confirm the accuracy of the determined measures as an indicator of clinical relevance (e.g., with respect to a LIPDRS part 3.10 Gait score). By doing so, the diagnostic platform can ensure that any identified disease indicators are clinically and statistically significant.

[0146] In some embodiments, the diagnostic platform can determine whether a rate of disease progression is statistically significant through a comparison between a healthy cohort and a cohort associated with progressive disease. For example, the diagnostic platform can determine, based on the comparison, a rate difference between the first subset and the second subset. The diagnostic platform can execute a hypothesis test for the rate difference to determine a linear mixed effects parameter associated with the set of humans associated with the progressive disease. The diagnostic platform can determine that the linear mixed effects parameter exceeds a threshold parameter value to established that the set of aggregated metric time series fulfills the progression rate criterion. As an illustrative example, the diagnostic platform executes hypothesis testing to determine a rate difference in a given metric (e.g., a continuity metric) with respect to a cohort associated with disease progression and a healthy cohort. For example, the diagnostic platform can determine whether a fb. in the linear mixed effects model is zero. For example, if a two-sided p value-46- 163869603.8PATENT Attorney Docket No. 124824.8118. WO01corresponding to in the linear mixed effects model is less than 0.05 (or, for example, another suitable value, such as 0.10), the proposed system can include that the difference in the monthly rate of change of the measure between the two cohorts is significant. If the sign of the 2 parameter indicates worsening of the disease in the progressive disease cohort, the diagnostic platform can conclude that the progression rate criterion is fulfilled. By doing so, the diagnostic platform can determine whether the continuity metric corresponds to an indicator if disease progression.

[0147] At operation 1204, the diagnostic platform can determine that the set of aggregated metric time series satisfies the criteria in order to determine that a metric is representative of a disease indicator. For example, based on a determination that the set of aggregated metric time series satisfies the set of criteria, the diagnostic platform can determine that the at least one metric is representative of a marker that can be used as an indicator of presence or progression of a disease. As an illustrative example, the diagnostic platform can determine that each of the set of criteria (e.g., the test-retest reliability criterion, the correlation criterion, and the progression rate criterion) are all satisfied for a given aggregated metric time series. Additionally or alternatively, the diagnostic platform can determine that some of these criteria are satisfied in order to determine that the metric is representative of the marker. As such, the diagnostic platform can confirm and / or characterize a relationship between a free-living digital measurement derived from wearable device data and a disease progression indicator, thereby providing information that supplements clinically derived data.

[0148] In some embodiments, the diagnostic platform can diagnose or identify a disease progression biomarker for a given human based on the determination that a given measure is a marker of disease progression or presence. For example, the diagnostic platform can receive, from a set of wearable devices, a sensor dataset comprising movement data for a patient. The diagnostic platform can determine, for the patient and based on the sensor dataset, a set of ambulatory windows and a set of non-ambulatory windows for the patient. The diagnostic platform can generate a set of continuity metrics based on the set of ambulatory windows and the set of non-ambulatory windows. The-47- 163869603.8PATENT Attorney Docket No. 124824.8118. WO01diagnostic platform can generate an aggregation of continuity metrics of the set of continuity metrics. The diagnostic platform can determine, based on the aggregation of the continuity metrics, a probability that the patient is associated with the progressive disease. The diagnostic platform, based on determining that the probability is greater than a threshold probability, can transmit a diagnosis message to the patient. For example, the diagnosis message indicates that the patient is associated with the progressive disease. As such, the diagnostic platform enables diagnosis, identification, and evaluation of a patient based on free-living indicators inaccessible to healthcare professionals in a clinical setting, thereby improving the accuracy and sensitivity of disease progression tracking.

[0149] From the foregoing, it will be appreciated that specific embodiments of the invention have been described herein for purposes of illustration, but that various modifications may be made without deviating from the scope of the invention. Accordingly, the invention is not limited except as by the appended claims.Processing System

[0150] Figure 13 includes a block diagram of a processing system 1300 in which at least some operations described herein can be implemented. For example, components of the processing system 1300 may be hosted on a computing device that includes a diagnostic platform.

[0151] The processing system 1300 can include a processor 1302, main memory 1306, non-volatile memory 1310, network adapter 1312, video display 1318, input / output devices 1320, control device 1322 (e.g., a keyboard or pointing device such as a computer mouse or trackpad), drive unit 1324 including a storage medium 1326, and signal generation device 1330 that are communicatively connected to a bus 1316. The bus 1316 is illustrated as an abstraction that represents one or more physical buses or point-to-point connections that are connected by appropriate bridges, adapters, or controllers. The bus 1316, therefore, can include a system bus, a Peripheral Component Interconnect (“PCI”) bus or PCI-Express bus, a HyperTransport (“HT”) bus, an Industry Standard Architecture (“ISA”) bus, a Small Computer System Interface (“SCSI”) bus, a Universal Serial Bus (“USB”) data interface, an-48- 163869603.8PATENT Attorney Docket No. 124824.8118. WO01Inter-Integrated Circuit (“l2C”) bus, or a high-performance serial bus developed in accordance with Institute of Electrical and Electronics Engineers (“IEEE”) 1394.

[0152] While the main memory 1306, non-volatile memory 1310, and storage medium 1326 are shown to be a single medium, the terms “machine-readable medium” and “storage medium” should be taken to include a single medium or multiple media (e.g., a centralized / distributed database and / or associated caches and servers) that store one or more sets of instructions 1328. The terms “machine-readable medium” and “storage medium” shall also be taken to include any medium that is capable of storing, encoding, or carrying a set of instructions for execution by the processing system 1300.

[0153] In general, the routines executed to implement the embodiments of the disclosure can be implemented as part of an operating system or a specific application, component, program, object, module, or sequence of instructions (collectively referred to as “computer programs”). The computer programs typically comprise one or more instructions (e.g., instructions 1304, 1308, 1328) set at various times in various memory and storage devices in a computing device. When read and executed by the processors 1302, the instruction(s) cause the processing system 1300 to perform operations to execute elements involving the various aspects of the present disclosure.

[0154] Further examples of machine- and computer-readable media include recordable-type media, such as volatile memory devices and non-volatile memory devices (e.g., non-volatile memory 1310), removable disks, hard disk drives, and optical disks (e.g., Compact Disk Read-Only Memory (“CD-ROMs”) and Digital Versatile Disks (“DVDs”)), and transmission-type media, such as digital and analog communication links.

[0155] The network adapter 1312 enables the processing system 1300 to mediate data in a network 1314 with an entity that is external to the processing system 1300 through any communication protocol supported by the processing system 1300 and the external entity. The network adapter 1312 can include a network adaptor card, a wireless network interface card, a router, an access point, a wireless router, a switch, a multilayer switch, a protocol converter, a gateway, a bridge, bridge router, a hub, a digital media receiver, a repeater, or any combination thereof.-49- 163869603.8PATENT Attorney Docket No. 124824.8118. WO01Remarks

[0156] The foregoing description of various embodiments of the claimed subject matter has been provided for the purposes of illustration and description. It is not intended to be exhaustive or to limit the claimed subject matter to the precise forms disclosed. Many modifications and variations will be apparent to one skilled in the art. Embodiments were chosen and described in order to best describe the principles of the invention and its practical applications, thereby enabling those skilled in the relevant art to understand the claimed subject matter, the various embodiments, and the various modifications that are suited to the particular uses contemplated.

[0157] Although the Detailed Description describes certain embodiments and the best mode contemplated, the technology can be practiced in many ways no matter how detailed the Detailed Description appears. Embodiments can vary considerably in their implementation details, while still being encompassed by the specification. Particular terminology used when describing certain features or aspects of various embodiments should not be taken to imply that the terminology is being redefined herein to be restricted to any specific characteristics, features, or aspects of the technology with which that terminology is associated. In general, the terms used in the following claims should not be construed to limit the technology to the specific embodiments disclosed in the specification, unless those terms are explicitly defined herein. Accordingly, the actual scope of the technology encompasses not only the disclosed embodiments, but also all equivalent ways of practicing or implementing the embodiments.

[0158] The language used in the specification has been principally selected for readability and instructional purposes. It may not have been selected to delineate or circumscribe the subject matter. It is therefore intended that the scope of the technology be limited not by this Detailed Description, but rather by any claims that issue on an application based hereon. Accordingly, the disclosure of various embodiments is intended to be illustrative, but not limiting, of the scope of the technology as set forth in the following claims.-50- 163869603.8

Claims

PATENT Attorney Docket No. 124824.8118. WO01CLAIMSWhat is claimed is:

1. A method performed by a computer program executed on a computing device, the method comprising:receiving, from a set of wearable devices, a sensor dataset that includes movement data for multiple humans over a period of time; executing a preprocessing algorithm to which the sensor dataset is provided as input, to generate a processed sensor dataset for a first human of the multiple humans;determining, based on the processed sensor dataset, a step-count dataset that includes a time series of step-count data associated with the first human;generating a set of test windows associated with the time series of step-count data, wherein each test window is of a test time period;for a first test window in the set of test windows:determining a first number of steps associated with the step-count dataset during the first test window, andbased on determining that the first number of steps is less than or equal to a step-count parameter, determining that the first test window includes a non-ambulatory period for the first human; determining, for the first human and based on the step-count dataset, a set of ambulatory windows and a set of non-ambulatory windows, wherein the set of non-ambulatory windows includes the first test window;generating a set of continuity metrics of a first type based on the set of ambulatory windows and the set of non-ambulatory windows; and based on an aggregation of the set of continuity metrics, determining that continuity metrics of the first type are indicative of a disease.-51- 163869603.8PATENT Attorney Docket No. 124824.8118. WO012. The method of claim 1, wherein executing the preprocessing algorithm comprises:generating, based on the movement data, a resampled movement dataset for the multiple humans,wherein the resampled movement dataset includes movement data associated with a predetermined temporal frequency; determining a bias associated with the movement data;based on the bias, modifying the resampled movement dataset to generate a shifted movement dataset;providing the shifted movement dataset to a band-pass filter to generate a filtered movement dataset; andgenerating the processed sensor dataset based on the filtered movement dataset.

3. The method of claim 1, comprising:generating a set of walking measurements based on the set of ambulatory windows, wherein the set of walking measurements includes at least one of (1 ) a daily number of ambulatory bouts or (2) a mean ambulatory bout duration; and determining, based on the set of walking measurements, that the continuity metrics of the first type are indicative of the disease.

4. The method of claim 1, comprising:generating an ambulatory bout measurement based on the step-count dataset, wherein the ambulatory bout measurement includes at least one of (1) a step count, (2) a number of bouts, (3) a cadence, (4) a bout duration, (5) an arm swing magnitude, or (6) an arm swing range of motion; and-52- 163869603.8PATENT Attorney Docket No. 124824.8118. WO01determining, based on the ambulatory bout measurement, that continuity metrics of the first type are indicative of the disease.

5. The method of claim 1, wherein determining the set of non-ambulatory windows comprises:obtaining sleeping data associated with the first human;determining sleeping time periods based on the sleeping data; andgenerating the set of non-ambulatory windows to include time periods not associated with the sleeping time periods and the set of ambulatory windows.

6. The method of claim 1, comprising:obtaining, for the first human, a treatment plan associated with the disease; and transmitting a representation of the treatment plan to a wearable device of the first human.

7. A method comprising:obtaining a longitudinal dataset that includes values arranged in temporal order,wherein each value is generated by a sensor in a computing device associated with a human at a corresponding point in time; segmenting the longitudinal dataset into a series of datasets, each of which is associated with a corresponding one of a series of windows;for each window in the series of windows,determining a number of steps taken by the human during a corresponding one of the series of datasets, and establishing that window is either(i) representative of a non-ambulatory period in response to a determination that the number of steps is less than or equal to a threshold, or-53- 163869603.8PATENT Attorney Docket No. 124824.8118. WO01(ii) representative of an ambulatory period in response to a determination that the number of steps is more than the threshold;identifying (i) non-ambulatory bouts that correspond to windows, if any, established as representative of non-ambulatory periods and (ii) ambulatory bouts that correspond to windows, if any, established as representative of ambulatory periods;computing a metric from which information regarding non-ambulatory pattern of the human is derivable based on an analysis of the non-ambulatory bouts;determining, based on the metric, whether the human is exhibiting an indicator of disease; andperforming an action based on whether the human is determined to exhibit the indicator of disease.

8. The method of claim 7, comprising:generating, based on the longitudinal dataset, a resampled dataset for the human, wherein the resampled dataset includes data associated with a predetermined temporal frequency;determining a bias associated with the longitudinal dataset;based on the bias, modifying the resampled dataset to generate a shifted dataset; providing the shifted dataset to a filter to generate a filtered dataset; and updating the longitudinal dataset to include the filtered dataset.

9. The method of claim 7, comprising:generating a set of walking measurements based on the ambulatory bouts, wherein the set of walking measurements includes at least one of (1) a daily number of ambulatory bouts or (2) a mean ambulatory bout duration; and-54- 163869603.8PATENT Attorney Docket No. 124824.8118. WO01determining, based on the set of walking measurements, that the metric is indicative of the disease.

10. The method of claim 7, comprising:generating an ambulatory bout measurement based on the longitudinal dataset, wherein the ambulatory bout measurement includes at least one of (1) a step count, (2) a number of bouts, (3) a cadence, (4) a bout duration, (5) an arm swing magnitude, or (6) an arm swing range of motion; and determining, based on the ambulatory bout measurement, that the metric is indicative of the disease.

11. The method of claim 7, wherein identifying the ambulatory bouts comprises: obtaining sleeping data associated with the human;determining sleeping time periods based on the sleeping data; andidentifying the ambulatory bouts to be associated with time periods not associated with the sleeping time periods and windows representative of non-ambulatory periods.

12. The method of claim 7, comprising:obtaining, for the human, a treatment plan associated with the disease; and transmitting, for display on the computing device, a representation of the treatment plan.

13. A computing device including:(i) one or more processors; and(ii) one or more non-transitory media storing instructions that, when executed by the one or more processors, cause the computing device to perform operations comprising:-55- 163869603.8PATENT Attorney Docket No. 124824.8118. WO01obtaining a step-count dataset that includes a time series of step-count data associated with a first human of a set of humans;generating, based on the step-count dataset and a step-count parameter, a set of ambulatory windows and a set of non-ambulatory windows; generating, based on the set of ambulatory windows or the set of nonambulatory windows, a metric that is indicative of continuity of bout duration;determining, based on the metric, whether the first human is exhibiting an indicator of a disease,wherein the indicator of the disease comprises a trend in bout duration for the first human; andperforming an action based on whether the first human is determined to exhibit the indicator of the disease.

14. The computing device of claim 13, wherein the instructions for generating the metric that is indicative of the continuity of bout duration cause the computing device to perform operations comprising generating a metric that is an indication of (i) a Gini measure, (ii) a non-ambulatory percentage metric, or (iii) a skewness for a bout duration distribution associated with the set of ambulatory windows and the set of non-ambulatory windows.

15. The computing device of claim 13, wherein the instructions for determining whether the first human is exhibiting an indicator of the disease cause the computing device to perform operations comprising:generating, based on the set of ambulatory windows or the set of non-ambulatory windows, a set of continuity metrics corresponding to a set of time periods; applying one or more mathematical operators to the set of continuity metrics to generate an aggregated metric time series,wherein each aggregated metric of the aggregated metric time series is associated with a daily time period; and-56- 163869603.8PATENT Attorney Docket No. 124824.8118. WO01performing a progression analysis of the aggregated metric time series to determine that the first human is exhibiting an indicator of the disease.

16. The computing device of claim 15, wherein the instructions for generating the aggregated metric time series cause the computing device to perform operations comprising applying a minimum operator, maximum operator, or pre-determined percentile operator to the set of continuity metrics to generate the aggregated metric time series.

17. The computing device of claim 13, wherein the instructions for generating the metric that is indicative of continuity of bout duration cause the computing device to perform operations comprising:generating a Gini value indicative of a continuity of ambulatory bouts,wherein the Gini value indicates a relationship between a cumulative distribution of a set of ambulatory bouts and corresponding percentiles of the ambulatory bouts in the set of ambulatory bouts; and generating the metric based on the Gini value.

18. The computing device of claim 13, wherein the instructions for generating the metric that is indicative of continuity of bout duration cause the computing device to perform operations comprising:generating a non-ambulatory time metric indicative of a bout duration associated with a pre-determined fraction of non-ambulatory bouts with a duration less than the bout duration; andgenerating the metric based on the non-ambulatory time metric.

19. The computing device of claim 13, wherein the instructions for generating the metric that is indicative of continuity of bout duration cause the computing device to perform operations comprising:generating a skewness for a bout duration distribution associated with the set of ambulatory windows and the set of non-ambulatory windows; and-57- 163869603.8PATENT Attorney Docket No. 124824.8118. WO01generating the metric based on the skewness.

20. One or more non-transitory media storing instructions that, when executed by one or more processors, cause a computing device to perform operations comprising: obtaining a set of aggregated metric time series for a set of humans,wherein each aggregated metric time series of the set of aggregated metric time series is associated with a corresponding one of the set of humans, wherein each aggregated metric time series of the set of aggregated metric time series includes at least one metric that characterizes, for the corresponding human, a set of associated ambulatory windows and / or a set of associated non-ambulatory windows;retrieving a set of clinical measures associated with the set of humans; determining whether the set of aggregated metric time series satisfies a set of criteria by establishing whether the set of aggregated metric time series fulfills:(i) a test-retest reliability criterion associated with an aggregate time period, (ii) a correlation criterion corresponding to correlation between the at least one metric and the set of clinical measures, and(iii) a progression rate criterion corresponding to a comparison between:(a) a first subset of aggregated metric time series associated with humans of the set of humans associated with a progressive disease, and(b) a second subset of aggregated metric time series associated with humans of the set of humans not associated with the progressive disease; andbased on a determination that the set of aggregated metric time series satisfies the set of criteria, determining that the at least one metric is representative of a marker that can be used as an indicator of presence or progression of a disease.-58- 163869603.8PATENT Attorney Docket No. 124824.8118. WO0121. The one or more non-transitory media of claim 20, wherein the instructions cause the computing device to perform operations comprising:receiving, from a set of wearable devices, a sensor dataset comprising movement data for a patient;determining, for the patient and based on the sensor dataset, a set of ambulatory windows and a set of non-ambulatory windows for the patient; generating a set of continuity metrics based on the set of ambulatory windows and the set of non-ambulatory windows;generating an aggregation of continuity metrics of the set of continuity metrics; determining, based on the aggregation of the continuity metrics, a probability that the patient is associated with the progressive disease; andbased on determining that the probability is greater than a threshold probability, transmitting a diagnosis message to the patient,wherein the diagnosis message indicates that the patient is associated with the progressive disease.

22. The one or more non-transitory media of claim 20, wherein the instructions cause the computing device to perform operations comprising:generating, according to a pre-determined daily aggregation period, the set of aggregated metric time series;determining an intra-class correlation metric between adjacent periods within the set of aggregated metric time series;comparing the intra-class correlation metric with a threshold correlation metric to determine that the intra-class correlation metric exceeds the threshold correlation metric; andbased on determining that the intra-class correlation metric exceeds the threshold correlation metric, establishing that the set of aggregated metric time series fulfills the test-retest reliability criterion.-59- 163869603.8PATENT Attorney Docket No. 124824.8118. WO0123. The one or more non-transitory media of claim 20, wherein the instructions cause the computing device to perform operations comprising:generating, between the set of aggregated metric time series and the set of clinical measures, a Spearman’s rank correlation coefficient and a corresponding confidence interval; andbased on determining that the corresponding confidence interval excludes a threshold correlation coefficient, establishing that the set of aggregated metric time series fulfills the correlation criterion.

24. The one or more non-transitory media of claim 20, wherein the instructions cause the computing device to perform operations comprising:determining, based on the comparison, a rate difference between the first subset and the second subset;executing a hypothesis test for the rate difference to determine a linear mixed effects parameter associated with the set of humans associated with the progressive disease; anddetermining that the linear mixed effects parameter exceeds a threshold parameter value to established that the set of aggregated metric time series fulfills the progression rate criterion.-60- 163869603.8

Images