Precision monitoring of multiple sclerosis with mobile devices: a transformative approach - this mentored F31 award will provide the trainee, a PHD student in epidemiology at stanford university, with…

Precision monitoring of multiple sclerosis with mobile devices: a transformative approach - this mentored F31 award will provide the trainee, a PHD student in epidemiology at stanford university, with the training necessary to launch an independent research career in computational epidemiology with a focus on neurological disorders. Under the guidance of a multidisciplinary team of expert mentors, his training goals are to: (1) develop advanced expertise constructing mobile digital tools to monitor neurological function, (2) obtain proficiency integrating heterogenous data sources (digital APP, electronic health record [EHR] and imaging data), (3) use advanced statistical techniques (data mining, machine learning) to examine the value of high-dimensional data in predicting clinical outcomes, and (4) gain additional proficiency in preparing manuscripts for publication. Multiple sclerosis (MS) impacts nearly 1 million young adults in the united states, characterized by acute demyelinating brain and spinal cord lesions that cause accumulation of serious neurologic disability over time. The disease course is extremely variable and relapsing remitting MS (rrms) is the most common subtype, affecting 85% of new onset patients. Treatment decisions are often made using limited cross-sectional data, and even prospectively acquired MRI imaging and clinical assessments in the context of randomized trials are too infrequent to capture the evolution of both clinical and subclinical disability as the disease course evolves. To address these gaps, the candidate will use data collected from 143 patients from the stanford MS center to (aim 1) develop and assess a mobile APP paradigm that uses high-resolution passively collected data to retrospectively characterize mobility impairment outside of the clinic, and (aim 2) improve prospective monitoring of patient function using a trimodal paradigm that includes background collection of mobility metrics, active apple researchkit performance tasks, mobile APP-administered surveys regarding patient- reported outcomes (pros). For both aims, the goal is to predict three outcomes (clinical relapses, expanded disability status scale [edss], MRI measures) that will be collected from objective clinical and imaging data. The proposal builds on extensive prior work of the trainee and his sponsors developing clinical research grade digital tools. If successful, the paradigm will improve upon existing methods by being less costly and less burdensome to patients, ultimately resulting in improved assessment of patients in both clinical and clinical research settings.