Skripsi
CARA-MATA : APLIKASI DETEKSI DINI KATARAK BERBASIS DEEP LEARNING MENGGUNAKAN ARSITEKTUR YOLOV11 PADA PERANGKAT MOBILE
Cataracts are the leading cause of blindness worldwide and in Indonesia, which are often treated too late due to limited access to medical services, especially in remote areas. Early detection is crucial to prevent more severe visual impairment. This study aims to develop a mobile application called CARA-Mata that utilizes Deep Learning technology with YOLOv11 architecture to detect cataracts early, quickly, and efficiently. The system development methods used include collecting an eye image dataset, training the YOLOv11 model, and designing an Android-based application. Testing was conducted in two main stages: model performance testing and application testing. The model training results showed excellent performance with a mAP@50 value of 98.7% and a mAP@50-95 of 95.0%, indicating high accuracy in classifying and mapping the position of cataracts. Functional testing using the Black Box Testing method showed that all application features ran according to the specified requirements and were valid. Furthermore, usability testing using the System Usability Scale (SUS) method resulted in an average score of 88.67, placing the application in the Excellent and Grade A categories. This shows that the CARA-Mata application is easy to use and suitable for implementation as a tool for early detection of cataracts in the community.
No other version available