A Crowdsourcing Approach to Develop Machine Learning Models to Quantify Radiographic Joint Damage in Rheumatoid Arthritis
A Crowdsourcing Approach to Develop Machine Learning Models to Quantify Radiographic Joint Damage in Rheumatoid Arthritis
Abstract To develop machine learning methods to quantify joint damage in patients with rheumatoid arthritis (RA), we developed the RA2 DREAM Challenge, a crowdsourced competition that utilized existing radiographic images and “gold-standard” scores on 674 sets of films from 562 patients. Training and leaderboard sets were provided to participants to develop methods to quantify joint space narrowing and erosions. In the final round, participants submitted containerized codes on a test set; algorithms were evaluated using weighted root mean square error (RMSE). In the leaderboard round, there were 173 submissions from 26 teams in 7 countries. Of the 13 submissions in the final round, four top-performing teams were identified. Robustness of results was assessed using Bayes factor and validated using an independent set of radiographs. The top-performing algorithms, which consisted of different styles of deep learning models, provided accurate and robust quantification of joint damage in RA. Ultimately, these methods lay the groundwork to accelerate research and help clinicians to optimize treatments to minimize joint damage.
Sun Dongmei、Mason Michael、Chung Verena、Ericson Lars、Guan Yuanfang、Stolovitzky Gustavo、Bridges S. Louis Jr.、Nguyen Thanh M.、Wang Jelai、RA2 DREAM Challenge Community、Dimitrovsky Isaac、Olar Alex、Allaway Robert J.、Chen Jake Y.、Pataki Balint Armin、Costello James C.、Israel Ariel、Yu Thomas V、Gulko Percio S.、Li Hongyang、Frazier Mason B.、Guinney Justin
University of Alabama at BirminghamSage BionetworksSage BionetworksCatskills ResearchDepartment of Computational Medicine and Bioinformatics, University of MichiganIBM T J Watson Research Center, IBM||Sema4University of Alabama at Birmingham||Division of Rheumatology, Department of Medicine, Hospital for Special SurgeryUniversity of Alabama at BirminghamUniversity of Alabama at BirminghamWRQ ResearchE?tv?s Lor¨￠nd University - Department of Complex Systems in PhysicsSage BionetworksUniversity of Alabama at BirminghamE?tv?s Lor¨￠nd University - Department of Complex Systems in PhysicsDepartment of Pharmacology, University of Colorado Anschutz Medical CampusLeumit Health Services, Tel-Aviv, Israel and Medil, solutions for digital medicineSage BionetworksDivision of Rheumatology, Department of Medicine, Icahn School of Medicine at Mount SinaiDepartment of Computational Medicine and Bioinformatics, University of MichiganUniversity of Alabama at BirminghamSage Bionetworks
医学研究方法基础医学临床医学
Sun Dongmei,Mason Michael,Chung Verena,Ericson Lars,Guan Yuanfang,Stolovitzky Gustavo,Bridges S. Louis Jr.,Nguyen Thanh M.,Wang Jelai,RA2 DREAM Challenge Community,Dimitrovsky Isaac,Olar Alex,Allaway Robert J.,Chen Jake Y.,Pataki Balint Armin,Costello James C.,Israel Ariel,Yu Thomas V,Gulko Percio S.,Li Hongyang,Frazier Mason B.,Guinney Justin.A Crowdsourcing Approach to Develop Machine Learning Models to Quantify Radiographic Joint Damage in Rheumatoid Arthritis[EB/OL].(2025-03-28)[2025-04-24].https://www.medrxiv.org/content/10.1101/2021.10.25.21265495.点此复制
评论