Pengembangan Sistem Deployment Deteksi untuk Kista Ginjal pada Citra Ct Scan dengan Metode Yolo

Abstract

Kidney cysts are a medical condition characterized by the formation of fluid-filled sacs on the kidneys, where CT scan image analysis is crucial for diagnosis and management. This study aims to develop a YOLOv5-based object detection model to identify kidney cysts in CT scan images. The research methodology involved training the model with a public dataset from Kaggle and validating it using private clinical data, with manual annotation conducted by a radiographer to ensure data accuracy. The results indicate that the YOLOv5 model achieved high performance with a Mean Average Precision (mAP) of 99.3%, a precision of 97.4%, and a recall of 99.1%. The model was successfully integrated into a Flask-based application, facilitating real-time kidney cyst detection in clinical practice. Consequently, this study demonstrates that the use of YOLOv5 can effectively support medical diagnosis, enhancing the accuracy and speed of kidney cyst detection, and offering a practical and innovative diagnostic tool for healthcare professionals. These findings open up opportunities for applying similar deep learning technologies to other medical conditions, significantly contributing to technological advancements in healthcare.