BACKEND IMPLEMENTATION FOR DIABETES RISK PREDICTION SYSTEM

• المؤلف الرئيسي: Iqbal Alam Firmansyah, Muhammad
• التنسيق: Final Project
• اللغة: Indonesia
• الوصول للمادة أونلاين: https://digilib.itb.ac.id/gdl/view/85859
• المؤسسة: Institut Teknologi Bandung

Diabetes is a disease that has a major impact on global health, with the risk of serious complications and high mortality rates. To assist in the prevention and management of diabetes, the development of systems that can predict diabetes risk and calculate calorie intake is very important. This research aims to develop and implement a machine learning-based backend that can predict diabetes risk and develop a calorie calculation feature through integration with the Edamam API. In this study, the backend was designed to receive and process diabetes risk factor data such as age, gender, BMI, and other health factors. All this data is managed in a PostgreSQL database which ensures data security and accessibility. In addition, the development of the calorie calculation feature was carried out using machine learning to identify types of food from images sent by users, then food calorie information was retrieved through integration with the Edamam API. Backend functional testing is carried out using the Postman application to ensure that the system is able to meet all the specified requirements. In addition, performance testing was carried out using Locust to evaluate the system's ability to receive 50 images per hour. Test results show that the backend is able to carry out diabetes risk predictions and calorie calculations efficiently, and the system successfully receives and processes images with adequate RPS, namely 0.93, well above the minimum limit of 0.014 RPS.