FITSCAN:AI CALORIE AND NUTRITION APP

Tired of traditional manual calorie tracking ? Just snap a photo and get instant calories, macros, nutrition insights & AI-generated meal plans—tailored to your goals and needs. One click. One app.

Comment

Inspiration

This project was inspired by my own fitness journey. For a long time, I struggled with weight loss, and when I finally committed to improving my health, I was advised to track my daily calorie intake. Although it sounded simple, the process quickly became frustrating. Manually recording every meal, estimating portions, and calculating calories felt tedious and difficult to maintain consistently. I found myself giving up multiple times. While brainstorming ideas for this project, I reflected on this experience and thought about how calorie tracking could be made easier for others. This led to the idea of building a tool that simplifies nutrition tracking, supports people on their wellness journey, and contributes to public good by helping individuals lead healthier lives.

IMPACT

This project addresses a real-world problem by simplifying calorie and nutrition tracking, which is a major barrier for many people trying to achieve health and fitness goals. Diet plays a critical role in outcomes such as weight loss, getting leaner, or gaining muscle, yet many individuals fail to stay consistent due to the inconvenience of manual tracking. By using AI-based food recognition, this web app reduces the effort required to log meals and provides all essential nutritional information in one place. To evaluate its impact, I shared the app with a few friends pursuing different health goals, including weight loss, muscle gain, and overall fitness improvement. They reported improved consistency in tracking their meals and better awareness of their nutrition. This demonstrates measurable impact through increased adherence, reduced friction, and improved engagement, showing how the app effectively meets a real-world need and supports sustainable healthy habits.

And i quote onE of my friends - “The relief of not having to search for the meals I ate in a traditional app database and especially not needing to measure every single ingredient for the dishes i make helped me finally stay consistent.”

What it does

FitScan is a complete one-stop nutrition solution designed to make calorie tracking and healthy eating effortless. Instead of searching for ingredients and manually entering food data, users can simply take a picture of their meal to instantly receive calorie and macro information. Traditional calorie-tracking methods are not only time-consuming but also prone to errors due to fragmented and inconsistent nutrition databases. FitScan overcomes these issues with a smarter, AI-driven approach.

Beyond calorie tracking, FitScan provides personalized meal plans and tailored nutritional guidance through its AI Meal Planner and AI Helper. Unlike many existing apps that offer generic advice, FitScan delivers recommendations customized to individual goals. Since diet plays a crucial role in overall health, FitScan empowers users to make better nutritional choices, making it a truly comprehensive solution for maintaining a healthy lifestyle.

How we built it

FitScan was developed as a modern web application using a React and TypeScript frontend, with shadcn/ui ensuring a polished and responsive user interface. The backend is powered by Supabase, which serves as a full-stack platform offering a PostgreSQL database, secure authentication, Row Level Security, and real-time data synchronization.

AI-powered food analysis was implemented using the Gemini API in combination with SmartBuckets to improve image understanding and accuracy. The application follows a component-based architecture with custom hooks for state management and API interactions. Vite was used to streamline the development workflow, and the database schema includes structured tables for user profiles, meals, and daily summaries that automatically calculate nutritional totals.

Challenges we ran into

This project was my first experience building a full-stack application, which involved a steep learning curve. One of the early challenges was implementing Supabase, particularly managing authentication and secure data handling across the application.

Another significant challenge was setting up AI-based image analysis. Achieving accurate calorie and macro detection required extensive fine-tuning so the system could recognize a wide variety of foods, including uncommon and mixed dishes.

The biggest challenge was integrating all the core features—the AI image calorie tracker, AI meal planner, nutrition analyzer, and AI helper—into a single cohesive system. Ensuring these components worked together seamlessly to deliver personalized responses based on each user’s profile required careful system design and repeated testing.

Accomplishments that we're proud of

1.)Successfully built a full-stack, AI-powered nutrition platform as a first experience with a system of this complexity.

2.)Implemented a unique photo-based calorie and macro tracking feature, eliminating tedious manual food logging.

3.)Fine-tuned AI image analysis to accurately recognize a wide range of foods, including uncommon and mixed meals.

4.)Integrated multiple AI components—image tracking, meal planning, nutrition analysis, and guidance—into a single unified platform.

5.)Delivered personalized nutrition insights dynamically tailored to each user’s profile and goals without human intervention.

6.)Created a true one-stop nutrition solution that simplifies tracking, planning, and guidance in one place.

What we learned

1.)How to design and build a full-stack application from the ground up.

2.)Practical experience using Supabase for authentication, data storage, and user management.

3.)Techniques for fine-tuning AI image analysis models for accurate food recognition and nutrition calculation.

4.)The importance of integrating multiple AI systems to deliver seamless and personalized user experiences.

5.)How simplifying real-world problems through automation leads to better, user-centric solutions.

What's next for FITSCAN:AI CALORIE AND NUTRITION APP

1.)Improve food recognition accuracy for regional, homemade, and mixed dishes.

2.)Integrate with wearable devices and health platforms like Google Fit, Fitbit, and Apple Health.

3.)Enhance personalization using long-term user data for better meal planning and calorie recommendations.

4.)Expand medical and dietary support, including conditions like diabetes, allergies, and specialized diets.

Built With

Share this project:

Updates