Development of Intelligent Software for Early Detection of Stunting in Toddlers Based on Anthropometry

Abstract

Stunting describes chronic undernutrition status during growth and development since early life. This situation is represented by the z-score of height for age (TB/U) less than -2 standard deviations (SD). The current method used to determine the presence of stunting in toddlers is to use Health Card. The way to do this is by weighing toddlers every month, the weighing results are recorded in the Health Card, and between the weight points from last month's weighing results and this month's weighing results are connected by a line. The series of children's growth lines form a child's growth graph. This procedure is of course less effective. In addition, midwives can differ from one another in determining stunting status in toddlers. The research aims to conduct early detection of stunting in toddlers based on anthropometry using an intelligent system. This research includes 3 (three) main things. The first is the development of software for managing children's medical record data. The second is the development of software for early detection of stunting using the Support Vector Machine (SVM). The third is the development of graph software for toddler anthropometric data development. The results of this study are expected to provide a tool for posyandu cadres and midwives in early detection of stunting in toddlers.