<![CDATA[Stunting is a serious health problem that affects children's growth and development, especially in areas with limited access to early detection. This research aims to develop a TensorFlow Lite-based “CENTING” Android application to detect stunting risk quickly and accurately. The prototyping method is used with the stages of identifying user needs, making initial prototypes, testing, and refinement based on the feedback of health workers and parents, until the application is ready to be implemented optimally. The dataset contains 121,000 child growth data from public sources, with variables such as age, gender, height, and nutritional status to detect stunting traits early. The data was processed and split 80:20 for training and testing, resulting in a detection accuracy of 98%. The selection of TensorFlow Lite is based on its advantage in response speed on mobile devices. The results showed that the CENTING application functioned optimally with a user acceptance score of 89.5%. The app supports self-detection, prevention education, and offline access, relevant for network-limited areas. These findings accelerate stunting intervention efforts and support government programs in reducing stunting prevalence.]]>
Copyrights © 2024
