Pediatric Wrist Fracture Detection Using Feature Context Excitation Modules in X‐ray Images
學年 114
學期 1
出版（發表）日期 2025-12-20
作品名稱 Pediatric Wrist Fracture Detection Using Feature Context Excitation Modules in X‐ray Images
作品名稱（其他語言）
著者 Rui‐Yang Ju, Chun‐Tse Chien, Enkaer Xieerke, and Jen‐Shiun Chiang
單位
出版者
著錄名稱、卷期、頁數 IET Image Processing, vol. 20, issue 1
摘要 Children often suffer wrist trauma in daily life and typically require radiologists to analyse and interpret X-ray images before undergoing surgical treatment. The development of deep learning has enabled neural networks to serve as computer-aided diagnosis (CAD) tools, assisting doctors and experts in medical image diagnostics. Since the you only look once version-8 (YOLOv8) model has achieved satisfactory success in object detection tasks, it has been applied to various fracture detection tasks. This work introduces four variants of feature contexts excitation-YOLOv8 (FCE-YOLOv8) model, each incorporating a different FCE module (i.e., modules of squeeze-and-excitation (SE), global context (GC), gather-excite (GE), and Gaussian context transformer (GCT)) to enhance the model performance. Experimental results on the GRAZPEDWRI-DX dataset demonstrate that our proposed YOLOv8 + GC-M3 model improves the mAP val 50 value from 65.78% to 66.32%, outperforming the state-of-the-art (SOTA) model while reducing inference time. Furthermore, our proposed YOLOv8 + SE-M3 model achieves the highest mAP val 50 value of 67.07%, exceed- ing the SOTA performance. The implementation of this work is publicly available at https://github.com/RuiyangJu/FCE-YOLOv8 .
關鍵字 Yolo v8, wrist Fracture detection, attention mechanism, modules of squeeze-and-excitation (SE), global context (GC), gather-excite (GE), Gaussian context transformer (GCT)
語言 en_US
ISSN
期刊性質 國外
收錄於 SCI
產學合作
通訊作者 江正雄
審稿制度 否
國別 GBR
公開徵稿
出版型式 ,電子版
SDGS 優質教育