Real-Time Wrist Fracture Detection Using YOLO11

Abstract

More than 1.71 billion individuals globally are impacted by bone diseases, especially conditions related to the musculoskeletal system, with fractures being a major worry. Prompt diagnosis is crucial, however, overlooked fractures are frequent in emergency situations, resulting in care delays. AI, particularly deep learning (DL), is becoming increasingly popular for aiding radiologists in identifying fractures in medical images. This review examines how deep learning is used in bone imaging. We employed the YOLO11 model, an algorithm for detecting objects in real-time that can forecast both bounding boxes and class probabilities at once, to detect bone fractures in real-time. We analyzed a variety of bone fracture images from different sources. Sophisticated training methods were used to improve the model's precision and resilience. The YOLO11 model showed great efficiency in detecting and categorizing fractures, providing substantial promise for live clinical use and enhancing diagnostic speed in medical settings.

Ähnliche Arbeiten