Image-Based Nutritional Analysis

I built this web app for a special lady in my life.

Clinical care is only one part of a person’s health—nutrition and physical activity are critical determinants of long-term wellbeing. Considering the stress and rigor required for daily exercise, dieting should be the easiest part of maintaining health. Yet for many people, tracking meals and nutritional metrics is a daily burden that discourages consistency. I realized that machine learning and computer vision could remove much of that friction and make nutritional tracking feel effortless.

This project is a computer vision–powered web application that takes in a photo of a plate of food and returns a nutritional analysis of the meal. The system identifies foods in the image, classifies them by type, and estimates key nutritional components such as calories, carbohydrates, proteins, fats, vitamins, and minerals. By making meal analysis fast and accessible, the goal is to help users make more informed decisions and build healthier habits over time.

Looking forward, aggregating user history could help surface long-term trends and highlight how eating patterns correlate with wellbeing. With richer data, the same pipeline could support personalized recommendations and low-cost nutritional guidance—bridging the gap between nutrition and technology in a user-friendly way.

GitHub: cv-nutritional-analysis