AI Nutritionist: Intelligent Dietary Analysis Using Google Gemini Pro Vision

Abstract

AI Nutritionist is an innovative application that leverages Google's Gemini Pro Vision API to provide real-time nutritional analysis of food images. The system offers personalized dietary recommendations and detailed nutritional breakdowns, making healthy eating more accessible through artificial intelligence.

Introduction

In an era where nutrition awareness is crucial for public health, AI Nutritionist bridges the gap between technology and dietary knowledge. This application demonstrates the practical application of computer vision and large language models in nutritional analysis.

Technical Architecture

Technology Stack

• Frontend: Streamlit
• AI Model: Google Gemini Pro Vision
• Image Processing: PIL (Python Imaging Library)
• Environment Management: python-dotenv
• Backend: Python

Implementation Details

Environment Setup and Configuration

import streamlit as st
import os
import google.generativeai as genai
from PIL import Image
from dotenv import load_dotenv

# Load environment variables and configure API
load_dotenv()
genai.configure(api_key=os.getenv("GOOGLE_API_KEY"))

Image Processing Function

def input_image_setup(uploaded_file):
    if uploaded_file is not None:
        bytes_data = uploaded_file.getvalue()
        image_parts = [
            {
                "mime_type": uploaded_file.type,
                "data": bytes_data
            }
        ]
        return image_parts
    else:
        raise FileNotFoundError("No file uploaded")

AI Response Generation

def get_gemini_response(input, image, prompt):
    model = genai.GenerativeModel('gemini-pro-vision')
    response = model.generate_content([input, image[0], prompt])
    return response.text

Prompt Engineering

The system uses a specialized prompt template for nutritional analysis:

input_prompt = """
You are an expert nutritionist analyzing food items from the image.
Calculate total calories and provide details for each food item:
1. Item 1 - no of calories
2. Item 2 - no of calories
...

Please also indicate:
- Overall health rating
- Macronutrient ratio (protein/carbs/fats)
- Fiber content
- Sugar content
- Other important nutritional elements
"""

Streamlit Interface Implementation

st.set_page_config(page_title="Nutritionist Generative AI Doctor")
st.header("Nutritionist Generative AI Doctor")

# User input controls
input = st.text_input("Input Prompt: ", key="input")
uploaded_file = st.file_uploader("Choose an image...", 
                                type=["jpg", "jpeg", "png"])

# Image display
if uploaded_file is not None:
    image = Image.open(uploaded_file)
    st.image(image, caption="Uploaded Image.", 
            use_column_width=True)

# Analysis trigger
submit = st.button("Tell me the total calories")

if submit:
    image_data = input_image_setup(uploaded_file)
    response = get_gemini_response(input_prompt, 
                                 image_data, 
                                 input)
    st.subheader("The Response is")
    st.write(response)

Key Features

1. Image Analysis

• Real-time food image processing
• Multi-format image support (JPG, JPEG, PNG)
• High-resolution image handling

2. Nutritional Analysis

• Calorie calculation
• Macronutrient breakdown
• Health rating assessment
• Dietary recommendations

3. User Interface

• Intuitive image upload
• Real-time analysis results
• Clear nutritional information display
• User-friendly design

4. AI Integration

• Google Gemini Pro Vision API integration
• Advanced machine learning algorithms
• Natural language processing
• Accurate food recognition

System Output

The application provides detailed nutritional information including:

1. Total calorie count
2. Individual item breakdown
3. Macronutrient ratios
4. Health assessment
5. Dietary recommendations

Performance Considerations

1. Image Processing

   • Optimal image size requirements
   • Format compatibility
   • Processing speed optimization

2. API Response Time

   • Efficient request handling
   • Response optimization
   • Error handling

3. User Experience

   • Minimal loading times
   • Intuitive interface
   • Clear result presentation

Future Improvements

1. Meal planning integration
2. Dietary restriction recognition
3. Recipe recommendations
4. Nutritional goal tracking
5. Multi-language support
6. Batch image processing
7. Historical analysis features

Security and Privacy

• Secure API key management
• User data protection
• Image data handling
• Privacy-compliant analysis

Conclusions

AI Nutritionist demonstrates the successful integration of advanced AI technology in dietary analysis. The system provides accurate, real-time nutritional information while maintaining user privacy and security.

App Link : https://huggingface.co/spaces/AbhayBhaskar/AI-Nutritionist