<![CDATA[Chronic kidney disease (CKD) is an increasing public health concern, with early detection being essential to prevent disease progression. Conventional screening methods are often invasive and require laboratory facilities, limiting their accessibility for adolescents. Methods: This research employed a Research and Development (R&D) design using the Waterfall and Prototype models to develop Sigaga, an artificial intelligence (AI)-based early screening application for kidney disease. The dataset included facial images to detect edema, lifestyle and health history data collected via standardized electronic questionnaires (WHO/CDC), and sleep quality assessed using the Pittsburgh Sleep Quality Index (PSQI). Convolutional Neural Networks (CNN) were applied to analyze facial images, while Random Forest/Gradient Boosting models processed questionnaire data. The results were integrated using ensemble learning. A pilot test was conducted with 100 adolescents in Karawang, Indonesia. Results: Most respondents were 18 years old (58%), female (88%), and of Sundanese ethnicity (55%). Questionnaire analysis showed that 42% of respondents were at moderate risk, strongly influenced by poor sleep quality (82%) and low water intake (87%]]>
Copyrights © 2026
