About the Project

Growth stunting in children is an important health issue globally; it has both short and long-lasting effects that are physical and cognitive. It is crucial to detect stunting because early intervention is crucial however detection at any time is important to allow for intervention methods. This project aims to detect growth stunting in children aged 5-10 using mobile cameras and machine learning algorithms.

The objective of the project is to develop an android mobile application with a system that can analyse visual data from, a mobile camera to detect signs of growth stunting. The system will use two machine learning algorithms for feature extraction and pattern recognition to identify signs of growth stunting.

The system will first collect data about the child’s health once the mobile camera has scanned the child. These data will include weight-to-age, height-to-age, and visual malnutrition signs such as thin arms and legs, swelling in the belly and face, dry, inelastic skin, rashes, and lesions. Then, the support vector description algorithm (SVDD) will be used to identify children who don’t fall within the typical range of a non-stunted child to reduce the list of children who might be stunted. After that, eXtreme gradient boosting (xgbTree) will be used on the flagged children to decide if the child is stunted based on the visual signs s of growth stunting.