Key Clinical Summary

• A patient-centered remote therapeutic monitoring (RTM) ecosystem improved mobility outcomes and exercise adherence among people with multiple sclerosis (MS) in a 6-week pilot study.
• Participants demonstrated a 20.4% increase in 6-Minute Walk Test distance and a 14.9% reduction in Timed Up and Go time, alongside lower fatigue on exercise days.
• An artificial intelligence (AI)-enabled clinical decision support tool achieved 99.15% analytical accuracy and received high usability ratings from clinicians.

Advancements in remote therapeutic monitoring technologies may help address persistent barriers in MS care by integrating tele-exercise programming, wearable monitoring, and AI-enabled clinical decision support into a unified digital ecosystem. Investigators reported that a patient-centered approach improved engagement, mobility outcomes, and clinician usability in a phased research and development study involving people with MS and health care professionals.

Pilot Study Demonstrates Feasibility of Integrated RTM Ecosystem

Researchers designed a 6-week web-based tele-exercise program using an exploratory sequential mixed-methods approach that incorporated direct feedback from people with MS. Focus groups involving 16 participants identified key design requirements, including safety cues, exercise modification options, accountability structures, and preferred instructional formats.

These findings informed development of a cloud-based RTM ecosystem that integrated smartwatch-derived activity data, daily fatigue ratings, and clinical mobility assessments. The pilot phase enrolled 15 individuals with MS in a 6-week tele-exercise intervention.

Investigators reported no adverse events during the pilot study. Mean adherence reached 82.29% ± 8.1%, suggesting strong participant engagement. Physical activity measures showed significantly higher cadence on exercise days compared with nonexercise days, with steps per minute averaging 11.1 ± 11.6 vs 0.9 ± 0.6 (P < .01). Participants also reported significantly lower fatigue on exercise days (P = .036).

Clinical mobility outcomes improved during the intervention. The 6-Minute Walk Test distance increased by 20.4%, equivalent to an average gain of 188 ft (P = .001). Timed Up and Go performance improved by 14.9%, corresponding to a 2.19-second decrease in completion time (P = .048).

The study also evaluated an AI-enabled clinical decision support tool powered by a custom large language model designed to summarize multimodal datasets from disconnected digital platforms. Compared with ground-truth data, the system achieved 99.15% analytical accuracy. Clinicians participating in the evaluation phase (n = 5) reported strong usability ratings, including helpfulness scores of 4.20 ± 0.45 and ease-of-use scores of 4.40 ± 0.55. Technology acceptance scores also remained high for usefulness (4.20 ± 0.58) and intention to use (4.53 ± 0.64).

AI-Enabled Monitoring May Streamline Neurologic Rehabilitation

The findings highlight the growing role of integrated digital health infrastructure in MS rehabilitation and chronic disease management. Current RTM adoption in neurology remains limited because many digital tools require patients and clinicians to navigate multiple disconnected platforms that create workflow inefficiencies and data fragmentation.

By combining tele-exercise delivery, wearable monitoring, cloud-based integration, and AI-enabled data synthesis, the proposed ecosystem may help streamline care coordination while supporting more personalized rehabilitation strategies. For payers and managed care stakeholders, the results suggest that integrated RTM systems could potentially improve patient engagement and functional outcomes while reducing operational burdens associated with managing large volumes of remote patient data.

The patient-centered design process also underscores the importance of involving individuals with chronic neurologic conditions in technology development. Incorporating patient preferences related to safety, accountability, and accessibility may improve adherence and long-term usability, factors that remain critical for successful digital health implementation.

Although the pilot study was small and short-term, the results support further evaluation of AI-assisted RTM models in broader MS populations and potentially other chronic neurologic diseases.

Researchers Emphasize Importance of Patient-Centered Design

The investigators concluded that “patient-centered research and development strategies that directly engage people with MS can facilitate the translation of technological innovation into tailored, clinically meaningful tools for MS care.”

They also noted that cloud-based RTM integration and AI-enabled data synthesis may help “overcome pressing implementation challenges associated with managing large, complex datasets derived from multiple, disconnected platforms.”

Results Support Continued Evaluation of Digital MS Rehabilitation

The study demonstrates the feasibility of combining tele-exercise rehabilitation, wearable monitoring, and AI-enabled decision support into a patient-centered RTM ecosystem for multiple sclerosis care. Investigators reported improvements in mobility, adherence, and clinician usability, supporting continued exploration of integrated digital health models in neurologic rehabilitation.

Reference

Willingham B, Stowell J, Cartwright J, Collier G, Backus D. Translating innovation into impact: improving access and precision in multiple sclerosis care by integrating tailored remote monitoring with artificial intelligence-enabled decision support. Presented at: Consortium of Multiple Sclerosis Centers Annual Meeting; May 27-29, 2026; Charlotte, NC.