AI-Driven Food Calorie Intake Estimation and Diet Recommendation System

Abstract

The modern life has enhanced the pace of emergence of chronic ailments such as type-2 diabetes, hypertension and thyroid afflictions owing to the unapproved eating patterns and absence of physical activities. The conventional health recommendation systems are founded on the firmly formed rule-based tendencies or shallow machinelearning systems that are incapable of mirroring the non-linear multidimensional factualities of human physiology. The given paper outlines the Food Calorie Intake Estimation and Diet Recommendation System based on Artificial Neural Networks (ANN), Contextual Personalization (MCP), and Large Language Models (LLMs) to predict health trends and give them guidance, respectively, activated by conversational and readable texts. The age, weight, height, Body Mass Index (BMI), blood-glucose, and hormone parameters, as well as exercise program and nutrition preference as parameters provided by the users, are entered into a 3-layered pipeline to produce individualized mealplans and fitness programs. The precision of the recommendations is numerically pegged on determinate formulas of Basal Metabolic rate (BMR) and Total Daily Energy expenditure (TDEE), and MCP context templates are used to make sure that a generative model does not yield physiologically implausible outcomes. The unit, integration, functional and scenario based testing has demonstrated a greater plan relevance and user compliance as compared to the old rule based baselines. The system is structured in a modular way thus enabling its future growth including integration with wearables, medical-condition-specific planning, and mobile deployment.