The application of a deep learning method for kidney stone detection by CT scans: A YOLOv8 approach

Kidney stone is a common urological condition that causes a significant healthcare burden. Early detection is crucial for optimal treatment planning. Traditionally, urologists rely on manual analysis of Computed Tomography (CT) scans for kidney stone detection which is a time-consuming and potentially error-prone process. This study investigated the application of a deep learning approach, specifically YOLOv8, for automated kidney stone detection in CT scans. We evaluated the performance of YOLOv8, with the YOLOv8x model demonstrating exceptional performance, achieving a near-perfect F1 score of 0.99 in detecting kidney stones on CT scans. This approach improves diagnostic accuracy by providing radiologists and urologists with a computer-aided diagnosis (CAD) tool for faster and more accurate kidney stone detection by CT scans.