Weakly Supervised Universal Fracture Detection in Pelvic X-rays

TL;DR

This paper introduces a two-stage weakly supervised method for detecting pelvic and hip fractures in X-ray images, achieving high accuracy and outperforming some physicians without requiring detailed annotations.

Contribution

The novel two-stage approach uses weakly supervised ROI mining and localized classification, improving fracture detection accuracy without needing region-specific labels.

Findings

Achieved an AUC of 0.975 on 4,410 PXRs.

Performed comparably or better than emergency physicians in a reader study.

Outperformed state-of-the-art fracture detection methods.

Abstract

Hip and pelvic fractures are serious injuries with life-threatening complications. However, diagnostic errors of fractures in pelvic X-rays (PXRs) are very common, driving the demand for computer-aided diagnosis (CAD) solutions. A major challenge lies in the fact that fractures are localized patterns that require localized analyses. Unfortunately, the PXRs residing in hospital picture archiving and communication system do not typically specify region of interests. In this paper, we propose a two-stage hip and pelvic fracture detection method that executes localized fracture classification using weakly supervised ROI mining. The first stage uses a large capacity fully-convolutional network, i.e., deep with high levels of abstraction, in a multiple instance learning setting to automatically mine probable true positive and definite hard negative ROIs from the whole PXR in the training data. The second stage trains a smaller capacity model, i.e., shallower and more generalizable, with the mined ROIs to perform localized analyses to classify fractures. During inference, our method detects hip and pelvic fractures in one pass by chaining the probability outputs of the two stages together. We evaluate our method on 4 410 PXRs, reporting an area under the ROC curve value of 0.975, the highest among state-of-the-art fracture detection methods. Moreover, we show that our two-stage approach can perform comparably to human physicians (even outperforming emergency physicians and surgeons), in a preliminary reader study of 23 readers.