Interpretable Rheumatoid Arthritis Scoring via Anatomy-aware Multiple Instance Learning

TL;DR

This paper introduces an interpretable, anatomy-aware multiple instance learning approach for predicting rheumatoid arthritis severity from hand radiographs, achieving state-of-the-art accuracy comparable to expert radiologists.

Contribution

It presents a novel two-stage pipeline that extracts disease-relevant regions and uses attention-based multiple instance learning for interpretable scoring.

Findings

Achieved a Pearson's correlation coefficient of 0.945 in score prediction.

Ensemble learning improved accuracy further.

The model identified relevant anatomical structures for RA progression.

Abstract

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.