Optimising Labeling

Abstract

Osteophytes are bony protrusions that are key radiographic indicators of hip osteoarthritis (OA), but grading their severity in specific hip locations is a time consuming process that requires an expert. In many cases it is expensive to scale datasets with location annotated severity labelling by experts, whereas weak labels, containing only the global presence of osteophytes is much easier to attain. This paper investigates whether such weak global label can improve localized severity grading through a multitask deep learning framework.

We study a ResNet-18 based convolutional network that shares and updates its weights across two output heads, a global binary classification head and four regional ordinal heads for femur superior, femur inferior, acetabulum superior and acetabulum inferior. The model is trained under four supervision strategies: a strong-only configuration using only quadrant-level labels, a masked baseline that incorporates weakly labelled negatives via label propagation and ignores weak positives in the local loss, and two Multi-Instance Learning variants that use a Noisy-OR loss to propagate weak positive labels to the quadrants. We systematically vary the ratio of weak to strong labels and evaluate performance using quadratic weighted Cohen’s kappa as the primary metric.

Experiments show that the masked baseline with weak labels improves regional kappa score compared to the strong-only configuration, while MIL variants fail to out-perform the baseline and can degrade performance at higher weak-to-strong ratios. We further observe that selecting checkpoints by minimal joint validation loss underestimates achievable kappa score, due to faster convergence of the global task, whereas selecting by maximal kappa score yields substantially better localized grading. Overall the findings highlight the trade off between localization and classification performance in weakly supervised multitask learning pipelines for regional osteophytes grading in hip X-Rays.