AI & MS Discovery Grant: Artificial Intelligence Support for Treatment Decisions in the MS Clinic

Start Term
End Term
Funding Amount
$1,000,000
Affiliation(s)
McGill University
Geographic Region(s) / Province(s)
Quebec
Researcher(s)
Hot Topic(s)
Progressive MS
Research Priorities
Diagnosis
Progressive MS
Impact Goal(s)
Advance Treatment and Care

Summary:

  • Developing a clinical decision support tool for use by clinicians and people living with MS to make better, more personalized decisions on an individuals’ treatment plan
  • Utilizes artificial intelligence and large-clinical trial datasets, including MRI images and clinical data, from over 10,000 people with MS to learn from this data and make predictions on future lesion formation as viewed by MRI, future relapses, and future increases in disability, along with a measure of how certain one can be about these predictions
  • Learnings will be further tested and developed using real-world data from MS clinics and with input from patients and physicians

Project Description:

The goal of this research is to help MS physicians and people living with MS make better, personalized treatment decisions by providing more reliable predictions of an individual’s disease course, and how they are likely to respond to different disease modifying therapies (DMTs).

This project led by Dr. Douglas Arnold aims to utilize state-of-the-art artificial intelligence (AI) approaches, an AI approach called ‘deep learning’, and large amounts of clinical and imaging data collected from over 10,000 people with MS who participated in clinical trials over the last 20 years. Almost all attempts thus far to predict MS disease course and treatment response have used traditional clinical and MRI outcome measures made on relatively limited datasets, which has provided only weak predictors and have not been particularly helpful to clinicians or patients.

The researchers will use these learnings to develop a clinical decision support tool that would assist MS physicians and patients make informed treatment decisions by providing predictions of an individual patient’s prognosis (e.g., future lesions, relapses, and increases in disability). The clinical decision tool would be used to provide recommendations to individuals to start either high efficacy DMTs that come with a higher risk of side effects, or lower efficacy DMTs that come with a lower risk of side effects. The research team plans to test this clinical decision tool using real-world data from MS clinics at McGill and will seek feedback from patients and clinicians using the tool.

Potential Impact: Development of a reliable clinical decision support tool will empower both MS physicians and patients make better treatment choices earlier in the course of the disease. By understanding the specific risks, and likelihood of response to various DMTs, patients and their clinicians may opt for more effective treatments earlier in the course of the disease, in order to change the trajectory of their disease.

Project Status: In progress