Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{194566,
        author = {Samarth Anil Katakdhond and Kaiwalya Kailas Mane and Niketan Purushottam Pandit and Harshvardhan Santosh Mastud and Chaitanya Bajirao Kopnar and Sohan Sampat Kale},
        title = {FitTrack Pro: An AI-Powered Full-Stack Fitness Intelligence Platform with Local LLM Integration and Multi-Dimensional Health Analytics},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {12},
        number = {10},
        pages = {5028-5036},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=194566},
        abstract = {The proliferation of mobile health applications has transformed personal fitness management, yet existing solutions frequently exhibit critical limitations in data privacy, computational accuracy, and contextual personalization. This research presents FitTrack Pro, a comprehensive full-stack fitness intelligence platform that addresses these challenges through innovative architectural decisions and scientifically validated algorithms. The platform integrates a locally-hosted Large Language Model (LLM) via Ollama with the Gemma 3:12B architecture, ensuring complete data sovereignty while delivering context-aware AI coaching through real-time database injection mechanisms. A corrected machine learning insight engine resolves three significant computational errors identified in conventional calorie balance prediction systems, including double-counting exercise calories and weight change sign inversion. The system architecture employs a modular Flask-based backend with Blueprint organisation, MySQL database with SQLAlchemy ORM, and a responsive vanilla JavaScript frontend. Multi-factor authentication with OTP verification, Server-Sent Events for real-time AI streaming, and a multi-dimensional gamification engine distinguish this implementation from contemporary fitness applications. Experimental evaluation demonstrates improved user engagement metrics and scientifically accurate metabolic calculations, positioning FitTrack Pro as a significant contribution to intelligent health monitoring systems.},
        keywords = {Fitness tracking, Large Language Models, Local AI inference, Health analytics, Machine learning, Calorie balance prediction, RESTful API, Server-Sent Events, Gamification, Data privacy},
        month = {March},
        }