Improved YOLO-Based Pulmonary Nodule Detection with Spatial-SE Attention and an Aspect Ratio Penalty

• Published in: Sensors (Basel, Switzerland) Vol. 25; no. 14; p. 4245
• Main Authors: Song, Xinhang, Xie, Haoran, Gao, Tianding, Cheng, Nuo, Gou, Jianping
• Format: Journal Article
• Language: English
• Published: Switzerland MDPI AG 08.07.2025; MDPI
• Subjects: Accuracy; Algorithms; aspect ratio penalty; Boxes; channel attention; EAPIoU loss; Experiments; Fines & penalties; Humans; Localization; Lung Neoplasms - diagnosis; Lung Neoplasms - diagnostic imaging; Optimization; pulmonary nodule detection; Solitary Pulmonary Nodule - diagnosis; Solitary Pulmonary Nodule - diagnostic imaging; spatial attention; Tomography, X-Ray Computed - methods; YOLOV11; YOLOV11; pulmonary nodule detection; spatial attention; aspect ratio penalty; channel attention; EAPIoU loss

The accurate identification of pulmonary nodules is critical for the early diagnosis of lung diseases; however, this task remains challenging due to inadequate feature representation and limited localization sensitivity. Current methodologies often utilize channel attention mechanisms and intersecti...