VitaML: Vitamin Deficiency Detection Using Machine Learning

Abstract

Due to a lack of easily accessible and non-invasive diagnostic techniques, vitamin deficiencies are becoming a serious public health concern that frequently go undiagnosed. Conventional testing can be costly, time-consuming, and equipment-intensive. VitaML presents a machine learning-based framework that uses visible biomarkers, skin characteristics, and discoloration to detect vitamin deficiencies. VitaML seeks to develop a rapid, inexpensive, and user-friendly diagnostic tool by training classification models on a dataset of visual cues associated with particular deficiencies. This system can encourage preventive care, increase awareness of general nutritional health, and assist medical professionals in identifying problems early.