A Deep Dive into Smart-Based Malnutrition Detection Using DenseNet-121 Deep Learning Algorithm

Nutrition plays a crucial role in the lives of children in society about their survival, growth and development. Undernutrition is now becoming common in the modern world. The research work is focused on children under five years and predominantly in the developing world. The data mining approach is used to forecast the malnutrition level of children below five years of age. Identification of underweight children is to enhance timely care under malnutrition. By incorporating an innovation in image analyzing feature in the framework; Deep Neural Networks (CNNs) specifically DenseNet-121 is incorporated and Transfer Learning which can help healthcare providers assess incoming cases to decide whether children are malnourished or not with high degrees of accuracy. In addition, it saves the time of healthcare professionals in a way that they don’t need to spend much time making such assessments manually which are time consuming. This not only increases the quality of the healthcare delivery systems but also optimizes the final result for children prone to malnutrition. Further, these systems can be installed in many environments such that many populations gain timely, accurate nutritional assessment.