Children often suffer wrist trauma in daily life, while they usually needradiologists to analyze and interpret X-ray images before surgical treatment bysurgeons. The development of deep learning has enabled neural networks to serveas computer-assisted diagnosis (CAD) tools to help doctors and experts inmedical image diagnostics. Since YOLOv8 model has obtained the satisfactorysuccess in object detection tasks, it has been applied to various fracturedetection. This work introduces four variants of Feature ContextsExcitation-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 themodel performance. Experimental results on GRAZPEDWRI-DX dataset demonstratethat our proposed YOLOv8+GC-M3 model improves the mAP@50 value from 65.78% to66.32%, outperforming the state-of-the-art (SOTA) model while reducinginference time. Furthermore, our proposed YOLOv8+SE-M3 model achieves thehighest mAP@50 value of 67.07%, exceeding the SOTA performance. Theimplementation of this work is available athttps://github.com/RuiyangJu/FCE-YOLOv8.