Label Augmentation Method for Medical Landmark Detection in Hip Radiograph Images

TL;DR

This paper presents a label augmentation technique for medical landmark detection in hip radiographs, demonstrating improved sample efficiency and performance over traditional methods through a curriculum training approach.

Contribution

Introduces a label-only augmentation scheme with a curriculum training strategy for enhanced landmark detection in medical images.

Findings

Outperforms traditional data augmentation methods.

Achieves higher accuracy with fewer training samples.

Validated on six radiograph datasets with expert annotations.

Abstract

This work reports the empirical performance of an automated medical landmark detection method for predict clinical markers in hip radiograph images. Notably, the detection method was trained using a label-only augmentation scheme; our results indicate that this form of augmentation outperforms traditional data augmentation and produces highly sample efficient estimators. We train a generic U-Net-based architecture under a curriculum consisting of two phases: initially relaxing the landmarking task by enlarging the label points to regions, then gradually eroding these label regions back to the base task. We measure the benefits of this approach on six datasets of radiographs with gold-standard expert annotations.