Interpretable Rheumatoid Arthritis Scoring via Anatomy-aware Multiple Instance Learning
This work addresses the problem of automating radiographic damage quantification for Rheumatoid Arthritis in clinical practice, representing a strong domain-specific advancement.
The authors tackled the inefficiency of manual Sharp/van der Heijde scoring for Rheumatoid Arthritis by developing an interpretable two-stage pipeline using dual-hand radiographs, achieving state-of-the-art performance with a Pearson's correlation coefficient of 0.945 and root mean squared error of 15.57, comparable to experienced radiologists.
The Sharp/van der Heijde (SvdH) score has been widely used in clinical trials to quantify radiographic damage in Rheumatoid Arthritis (RA), but its complexity has limited its adoption in routine clinical practice. To address the inefficiency of manual scoring, this work proposes a two-stage pipeline for interpretable image-level SvdH score prediction using dual-hand radiographs. Our approach extracts disease-relevant image regions and integrates them using attention-based multiple instance learning to generate image-level features for prediction. We propose two region extraction schemes: 1) sampling image tiles most likely to contain abnormalities, and 2) cropping patches containing disease-relevant joints. With Scheme 2, our best individual score prediction model achieved a Pearson's correlation coefficient (PCC) of 0.943 and a root mean squared error (RMSE) of 15.73. Ensemble learning further boosted prediction accuracy, yielding a PCC of 0.945 and RMSE of 15.57, achieving state-of-the-art performance that is comparable to that of experienced radiologists (PCC = 0.97, RMSE = 18.75). Finally, our pipeline effectively identified and made decisions based on anatomical structures which clinicians consider relevant to RA progression.