Design and Evaluation of a Deep Learning Aided Approach for Kidney Stone Detection in CT scan Images

A lot of patients confess to emergency departments with agonizing pain due to kidney stones, which are a prevalent illness across the world. Numerous imaging techniques are used to diagnose kidney stone disease. Specialists are needed for the full diagnosis and analysis of these photographs. The practical methods that may be employed as supplemental tools to aid physicians in their diagnosis are computer-aided diagnostic systems. This study suggests using coronal computed tomography (CT) scans to automatically detect kidney stones (having stones or not), using the deep learning (DL) approach, which has lately made major strides in the field of artificial intelligence. Different cross-sectional CT pictures were taken for each subject, resulting in a total of 12446 images being used. In the model’s training phase, 1600 CT images were employed, of which 80% were used for training and 20% for validation. 346 pictures that weren’t included in the deep model’s training were used to determine test performance values once model training was complete. This detection is constructed using the automated model, its accuracy was 99.82%.