A Novel Wearable Digital Biomarker for Detecting Changes in Multiple Sclerosis (MS) Condition

Multiple sclerosis (MS) is lifelong autoimmune disease that is typically first diagnosed in young adults; MS affects the central nervous system and can result in various impairments, including walking, cognition, dexterity, sleep, vision and bladder control. Notably, impairments to gait are the most common and are identified as the most impactful to a person with MS's (PwMS's) quality of life. Furthermore, ambulation is a key metric used to assess the severity of MS and is the basis for the Expanded Disability Status Scale (EDSS) that represents the global standard for assessing a patient's MS condition. For these reasons, clinicians employ a variety of gait tests to assess the severity and progression of the disease, which require frequent clinical visits and lack objective measurements as compared to what can be measured in a laboratory setting. Current scales do not detect subtle progression that could be indicative of early transformation into Secondary Progressive MS (SPMS) from Relapsing Remitting MS (RRMS) or significant progression in progressive forms of MS.

With advancements in wearable technologies and Artificial Intelligence (AI)-based algorithm development, clinicians can be provided with meaningful laboratory grade gait metrics collected in the patient's home environment to assist their practice. Objective walking information can be provided to clinicians to track the personalized progression of the disease to enable a more targeted treatment plan. A subset of this data is also shared with the patients via their smart phone app to keep them informed and motivated.

Several times per week, smart insoles in the patient's shoes will collect data from the embedded sensors (pressure sensors, accelerometer, gyroscope). The wearable smart insoles are fitted into a pair of the patient's "everyday use" shoes, and are very similar to the type of "comfort" insoles available from a local pharmacy. The smart insole data will be used to create AI-based personalized models that compute each individual's walking signature; this includes tracking of subtle changes over time (improvement, deterioration) as well as identifying specific gait phenotypes.