Innovative Collaboration Enhances Symptom Monitoring for Multiple Sclerosis Patients

Researchers from the University of Pittsburgh and Carnegie Mellon University have developed a groundbreaking mobile application powered by artificial intelligence to assist individuals living with multiple sclerosis (MS) in tracking their symptoms more effectively. This innovative app passively collects health data from personal digital devices such as smartphones and fitness trackers, providing valuable insights into depression levels and symptom progression. The study, published in the Journal of Medical Internet Research, demonstrates that the machine learning model within the app can accurately predict depression burden two weeks in advance, with an accuracy range of 70-80%. This predictive ability allows patients and clinicians to intervene earlier, potentially mitigating severe mental health complications.

Multiple sclerosis affects approximately three million people worldwide, including around one million in the United States. MS involves the immune system attacking myelin, the insulating layer surrounding nerve fibers, which disrupts communication between the brain and body. Symptoms commonly include motor problems, vision changes, sensory issues like numbness, and cognitive disturbances. Depression is also prevalent among MS patients, impacting up to 50% of individuals with the condition and exacerbating symptoms such as fatigue, motor impairment, and memory problems.

The collaborative project leverages the strengths of both institutions: Birmingham-based UPMC offers clinical expertise, while CMU provides advanced machine learning capabilities. The app's passive data collection included step count, location, heart rate, sleep patterns, call and screen time, all de-identified to protect privacy. The collected data was processed through a machine learning model that accurately predicted patients' depression levels and forecasted symptoms for future two-week periods.

Participants appeared receptive to this technology, with full retention at three months and about half extending participation to six months. This approach offers a safe, non-intrusive way for patients to monitor their mental health from home, reducing reliance on frequent clinic visits. Experts highlight its potential to enable earlier detection of symptoms, leading to tailored treatment strategies that might slow disease progression or improve quality of life.

Despite the promising results, the researchers acknowledge limitations, including demographic biases, as most participants were women and 93% were white. They emphasize the importance of responsible AI development, considering biases and privacy safeguards. Moving forward, the team aims to customize and validate the model across diverse populations, including breast cancer survivors and adolescents with suicidal ideation, to broaden its applicability. Ultimately, the goal is to integrate such digital tools into routine care, enhancing early intervention and personalized treatment for MS and other chronic conditions.

This innovative fusion of healthcare and AI offers a new avenue for continuous symptom management, potentially transforming how MS and mental health are monitored and treated in the future.

source: https://medicalxpress.com/news/2025-07-collaboration-multiple-sclerosis-patients-track.html