DREAM Challenge to automate assessment of radiographic damage from rheumatoid arthritis

UAB is helping develop automated technology to assess joint damage from rheumatoid arthritis.

DREAM2S. Louis Bridges Jr., M.D., Ph.D., and his team have been awarded $100,000 by the National Institutes of Health National Center for Data to Health Awards Program and $200,000 by Bristol-Myers Squibb Company to implement a DREAM Challenge. The challenge is to develop an automated method using machine learning approaches to quickly and accurately quantify the degree of radiographic joint damage associated with rheumatoid arthritis, or RA.

DREAM — Dialogue on Reverse Engineering Assessment and Methods — Challenges, are community-based collaborative competitions that engage diverse communities of scientists. These include the brightest minds in statistics, machine learning and computational biology to competitively solve a specific problem in biomedicine in a given time period.

RA, an autoimmune disease in which the body’s immune system attacks its own joints, can causes devastating joint pain and deformity. Radiographs of the hands and feet are used to assess joint damage and adjust medications to prevent further damage. The current process of quantifying joint damage is by manually inspecting radiographic images to subjectively estimate the degree of joint space narrowing and bony erosions in affected joints. This process is time-consuming and expensive, and there is often variability among the evaluators’ scores.  

Bridges is confident the RA2-DREAM Challenge will lead to a tool that will be highly useful to researchers and rheumatologists, and will lead to better ways for physicians to tailor therapies to prevent joint damage and disability in patients with RA.  

“Our goal for this project is to create a robust, publicly available method to accurately and quickly quantify joint damage from RA,” explained Bridges, director of the Division of Immunology and Rheumatology at the University of Alabama at Birmingham. “We’re excited for the potential of this damage assessment tool to improve outcomes for patients with RA.”

UAB will be providing existing large sets of radiographs and validated damage scores from previous studies for participants to use for algorithm development. The RA2-DREAM Challenge is sponsored by Sage Bionetworks. In addition to a team from UAB, additional collaborators from the University of Colorado and Icahn Mount Sinai School of Medicine are involved.   

The challenge is now open and will end March 19, 2020. Winners will be announced in July 2020, and a total of $50,000 will be awarded to the winning team.