Research on Nutrition Deficiency Analysis and Diet Plan Recommendation System

The World Health Organization (WHO) has shown that a lack of or uneven intake of food contributes to roughly 9% of heart attack fatalities, 11% of ischemic heart disease deaths, and 14% of gastrointestinal cancer deaths globally. More than a billion individuals are anemic due to iron deficiency (anaemia), 0.25 billion children have vitamin deficiencies ranging from vitamin A to vitamin K inadequacy, and 0.7 billion are iodine deficient, making a total of roughly 0.25 billion people anaemic. The main objective of this paper is to provide dietary recommendations. This paper introduces a technology-driven solution that brings together artificial intelligence, data analytics, and nutritional science to offer personalized dietary insights in an efficient and user-friendly manner. By utilizing modern machine learning techniques, the system is capable of interpreting a wide range of user inputs and health indicators to provide targeted diet recommendations that are not only relevant but also adaptable to individual lifestyles. This approach enhances the accessibility of preventive nutrition strategies and supports early detection of deficiencies, ultimately empowering users to take control of their health in a more informed and proactive way.

LITERATURE REVIEW

1. 1.  Overview of Nutrition deficiency analysis and diet plan recommendation System

The system focuses on identifying nutrient imbalances in individuals and offering tailored dietary solutions to address these deficiencies. Nutrient deficiencies, those in iron, vitamin D, calcium, can lead to a range of health issues, including fatigue, weakened immunity, bone disorders. Traditional methods for diagnosing deficiencies, like blood tests and dietary surveys, are often costly and labor-intensive, making them inaccessible for many individuals. Recent advancements in technology have introduced automated systems, such as mobile applications and websites, to detect nutrient imbalances in real time and provide personalized diet recommendations. These tools offer a more efficient and accessible approach to managing nutritional health by tracking food intake and health biomarkers, ultimately promoting better long-term health outcomes. The development of systems holds the potential to significantly improve global health making personalized nutrition guidance available to a wider population.

1. 2. Techniques used in system

AIML techniques that are used in nutrition deficiency analysis and diet plan recommendation:

• Nutrition Deficiency Tracking and Visualization: It creates dynamic health graphs that track key nutrient levels (e.g., iron, calcium, vitamins). These graphs visualize deficiencies and improvements based on real-time data with the help of graphs helping users and healthcare providers monitor progress and adjust dietary plans.
• Predictive Analytics for Health Risks: Systems analyze trends in the health graph to forecast future nutrient deficiencies and associated health risks. By predicting potential issues (e.g., vitamin D deficiency leading to weakened immunity), these models enable early intervention and proactive dietary recommendations.
• Personalized Diet Recommendations: Based on the health graph and deficiency analysis, AI models suggest personalized diet plans tailored to address specific nutritional imbalances. These recommendations are updated dynamically, taking into account user feedback and health changes over time to ensure ongoing optimization.
• Integration of Multi-Source Health Data: The systems combine data from various sources, such as dietary logs, wearable health trackers, and medical records, into a unified health graph. This multi-modal approach provides a comprehensive view of an individual’s health and nutrient status, ensuring more accurate and holistic recommendations.
• Real-Time Monitoring and Dynamic Adjustments: Health graphs are updated in real time as users make dietary changes or receive new health data (e.g., blood test results). AI continuously monitors these updates, offering adaptive, data-driven recommendations to address deficiencies and improve health outcomes effectively.
  1. Data Sources

A diverse set of data sources is essential for Nutrition deficiency assessment and diet plan recommendation to function effectively:

• Dietary Intake Data: Capturing detailed information about what the user eats is essential for understanding nutrient intake. This data helps assess the user's overall nutrition and identify deficiencies in specific nutrients. A reliable food composition database is crucial for accurately assessing the nutritional content of the foods logged by the user.
• Medical History and Clinical Data: Data on chronic conditions (e.g., diabetes, hypertension) and medications is vital to tailor diet plans, as certain conditions and drugs affect nutrient absorption and metabolism.
• Genetic and Metabolic Data: Genetic data revealing how an individual metabolizes certain nutrients (e.g., vitamin D absorption, lactose intolerance) can be crucial for creating personalized diet plans that optimize nutrient utilization.
• Psychological and Behavioral Data: Understanding eating behaviors, meal timing, portion sizes, and emotional eating patterns is important for personalizing dietary changes in a way that suits the user’s lifestyle.

Fig 1. Nutrition based on multiple data aspects

METHODOLOGY

The proposed Nutrition Deficiency Analysis and Diet Plan Recommendation System begins by interacting with users—either individuals or healthcare professionals—seeking personalized dietary guidance. Users input vital information such as medical history, dietary habits, genetic and metabolic data, as well as psychological and behavioral patterns. Additionally, users can select symptoms like fatigue or brittle nails, which help indicate possible nutritional deficiencies. This data is then preprocessed through cleaning and feature extraction to ensure accuracy. Using a trained machine learning model, the system analyzes the input to detect deficiencies in vitamins, minerals, or macronutrients and generates a nutrient deficiency graph for visual representation. Based on this analysis, it retrieves relevant data from a backend SQLite3 database—including food nutrition values and medical guidelines—and provides a tailored diet plan that includes nutrient-rich food suggestions and meal planning strategies. Finally, the system gathers user feedback to optimize its recommendations through continuous learning, ensuring the model improves over time and delivers increasingly accurate and effective dietary guidance. The entire pipeline is developed in Python, trained using Jupyter Notebook or VS Code, and deployed using Flask for real-time user interaction.

FLOW CHART

Working

The system operates by first collecting comprehensive user data through an intuitive interface, capturing details such as medical background, dietary intake, symptoms, and lifestyle habits. This information is then preprocessed to handle missing values and extract meaningful features that influence nutritional health. Using a machine learning model trained on relevant datasets, the system analyzes the user's profile to identify potential nutrient deficiencies and generate a visual nutrient deficiency graph. It then consults a backend SQLite3 database containing food nutrition data and clinical guidelines to recommend a personalized, balanced diet plan. These recommendations are tailored to the user's deficiencies and preferences, promoting better health outcomes. As users follow the plan and provide feedback, the system refines its suggestions using continuous learning, ensuring increasingly accurate and effective dietary guidance over time.

System Requirement

Software Requirement

1) Python Version IDE

2) VS Code

Modules Used

1) Flask

2) PyTorch

IMPLEMENTATION & RESULT

Implementation

Implementation Step by Step

Step 1: Run the Python File for Output

Fig. Shows Running the Code for Output

Step 2: User Authentication and Data Input Interface

Fig. Shows Login Interface for User Data Input

In this paper, we focus on analyzing and predicting deficiencies in key vitamins and minerals that are essential for maintaining good health. These include Iron, Vitamin A, Vitamin B1, B2, B3, B6, B12, Vitamin C, D, E, K, Folate, and Zinc. Each nutrient plays a critical role in the body, such as improving vision (Vitamin A), supporting nerve function (B12), enhancing immunity (Vitamin C), maintaining bone strength (Vitamin D), and aiding in blood formation (Iron, Folate). Based on the user's symptoms and lifestyle factors, the system identifies potential deficiencies and provides personalized diet recommendations to restore nutrient balance and improve overall well-being.

Fig. Shows User Registration Form for New Account Creation

Fig. Shows User Login Interface

Step 3: User-Guided Symptom Assessment Dashboard

Fig. Shows the User Dashboard

Fig. shows User interface displaying symptom selection and personal data input fields

Fig. i)

Fig. ii)

Fig. i) & ii) shows a web interface for selecting symptoms

Step 4: Prediction Analysis and Performance Metrics

Fig. Shows User Interface for Selecting Symptoms Related to Nutritional Deficiencies

Fig. shows a web interface displaying identified nutritional deficiencies with percentage likelihood

Fig. iii)

Fig. iv)

Fig. iii) & iv) shows a web interface displaying a personalized daily meal schedule and recommended foods based on identified nutritional needs and better nutrient absorption tips

Following the generation of the diet plan, the system provides supplementary information to maximize the benefits of the dietary recommendations. This image illustrates the "Absorption Tips" section, offering advice such as consuming Riboflavin-rich foods with meals and ensuring adequate intake of other B vitamins. This step highlights the system's comprehensive approach to dietary guidance, going beyond just suggesting foods to also educating users on how to optimize nutrient uptake for better health outcomes.

Fig. shows a web interface displaying reported symptoms and a disclaimer about machine learning-based predictions

RESULT

The Nutrition Deficiency Analysis and Diet Plan Recommendation System leverages comprehensive user data, including medical history, dietary intake, genetic factors, and lifestyle behaviors, to assess and address potential nutrient deficiencies. By analyzing symptoms such as fatigue, hair loss, and poor wound healing, the system identifies specific deficiencies like iron or vitamin D. It then generates personalized diet plans tailored to individual health profiles, recommending nutrient-rich foods and structured meal plans to optimize nutritional balance and overall well-being. Through user feedback and continuous learning, the system refines its recommendations, ensuring effective and tailored dietary guidance for improved health outcomes.

Fig. shows the graph of common vitamin deficiencies and deficiency distribution across age groups

Fig. shows a visual representation of symptom correlation with vitamin deficiencies and monthly trends of new cases

Fig. shows a web dashboard showing user history, predicted vitamin deficiency (Vitamin B2)

The user profile section indicates that "opwpoe" has been a member since April 17, 2025, and has taken a total of one test, with the last test conducted on April 19, 2025. The "Prediction History" clearly shows that on April 19, 2025, the system identified a Vitamin B2 (Riboflavin) deficiency for this user, based on the reported symptoms: Dry Eyes, Memory Loss, Fast Heart Rate, Headache, and Easy Bruising. This feature allows users to easily access their past diagnostic outcomes within the system.

Fig. Shows Prediction history interface displaying a previous diagnosis of Vitamin B2 (Riboflavin) deficiency along with associated symptoms and a personalized dietary plan generated based on the prediction