Demographic-aware fine-grained visual recognition of pediatric wrist pathologies

TL;DR

This paper presents a demographic-aware hybrid model for pediatric wrist pathology recognition from X-rays, addressing challenges of developmental variability and improving accuracy through fine-grained recognition and demographic integration.

Contribution

It introduces a hybrid convolution-transformer model that incorporates demographic data and progressive masking, enhancing pediatric wrist pathology detection beyond traditional methods.

Findings

Hybrid model outperforms CNNs in accuracy.

Demographic fusion improves diagnostic performance.

Pretraining on fine-grained data enhances transferability.

Abstract

Pediatric wrist pathologies recognition from radiographs is challenging because normal anatomy changes rapidly with development: evolving carpal ossification and open physes can resemble pathology, and maturation timing differs by sex. Image-only models trained on limited medical datasets therefore risk confusing normal developmental variation with true pathologies. We address this by framing pediatric wrist diagnosis as a fine-grained visual recognition (FGVR) problem and proposing a demographic-aware hybrid convolution--transformer model that fuses X-rays with patient age and sex. To leverage demographic context while avoiding shortcut reliance, we introduce progressive metadata masking during training. We evaluate on a curated dataset that mirrors the typical constraints in real-world medical studies. The hybrid FGVR backbone outperforms traditional and modern CNNs, and demographic fusion yields additional gains. Finally, we show that initializing from a fine-grained pretraining source improves transfer relative to standard ImageNet initialization, suggesting that label granularity, even from non-medical data, can be a key driver of generalization for subtle radiographic findings.