Patch-based feature mapping with generative adversarial networks for auxiliary hip fracture detection

Shang-Lin Chung; Chi-Tung Cheng; Chien-Hung Liao; I-Fang Chung

doi:10.1016/j.compbiomed.2024.109627

Patch-based feature mapping with generative adversarial networks for auxiliary hip fracture detection

Comput Biol Med. 2025 Jan 9:186:109627. doi: 10.1016/j.compbiomed.2024.109627. Online ahead of print.

Authors

Shang-Lin Chung¹, Chi-Tung Cheng², Chien-Hung Liao², I-Fang Chung³

Affiliations

¹ Institute of Biomedical Informatics, National Yang Ming Chiao Tung University, Taipei, Taiwan.
² Department of Trauma and Emergency Surgery, Chang Gung Memorial Hospital, Linkou, Chang Gung University, Taoyuan, Taiwan.
³ Institute of Biomedical Informatics, National Yang Ming Chiao Tung University, Taipei, Taiwan. Electronic address: ifchung@nycu.edu.tw.

PMID: 39793347
DOI: 10.1016/j.compbiomed.2024.109627

Abstract

Background: Hip fractures are a significant public health issue, particularly among the elderly population. Pelvic radiographs (PXRs) play a crucial role in diagnosing hip fractures and are commonly used for their evaluation. Previous research has demonstrated promising performance in classification models for hip fracture detection. However, these models sometimes focus on the images' non-fracture regions, reducing their explainability. This study applies weakly supervised learning techniques to address this issue and improve the model's focus on the fracture region. Additionally, we introduce a method to quantitatively evaluate the model's focus on the region of interest (ROI).

Methods: We propose a new auxiliary module called the patch-auxiliary generative adversarial network (PAGAN) for weakly supervised learning tasks. PAGAN can be integrated with any state-of-the-art (SOTA) classification model, such as EfficientNetB0, ResNet50, and DenseNet121, to enhance hip fracture detection. This training strategy incorporates global information (the entire PXR image) and local information (the hip region patch) for more effective learning. Furthermore, we employ GradCAM to generate attention heatmaps, highlighting the focus areas within the classification model. The intersection over union (IOU) and dice coefficient (Dise) are then computed between the attention heatmap and the fracture area, enabling a quantitative assessment of the model's explainability.

Results and conclusions: Incorporating PAGAN improved the performance of the classification models. The accuracy of EfficientNetB0 increased from 93.61 % to 95.97 %, ResNet50 improved from 90.66 % to 94.89 %, and DenseNet121 saw an increase from 93.51 % to 94.49 %. Regarding model explainability, the integration of PAGAN into classification models led to a more pronounced attention to ROI. The average IOU improved from 0.32 to 0.54 for EfficientNetB0, from 0.28 to 0.40 for ResNet50, and from 0.37 to 0.51 for DenseNet121. These results indicate that PAGAN improves hip fracture classification performance and substantially enhances the model's focus on the fracture region, thereby increasing its explainability.

Keywords: Explainable AI; Generative adversarial network; Hip fracture detection; Weakly supervised learning.