Fundusnap

Fundusnap is a comprehensive medical imaging solution designed to help healthcare professionals detect and analyze diabetic retinopathy through fundus images. The project consists of three main components: a mobile application, a backend API service, and an AI model for offline image classification.

A modern cross-platform mobile application built with Flutter that provides:

• High-quality fundus image capture
• Video recording capabilities
• Secure storage of medical images
• AI-powered analysis of diabetic retinopathy
• Intelligent chatbot assistant for medical insights

Tech Stack:

• Framework: Flutter (SDK ^3.8.0)
• Language: Dart
• State Management: Flutter Bloc
• Navigation: Go Router
• Key Dependencies:
  • Camera: camera: ^0.11.1
  • Video Player: video_player: ^2.9.5
  • Secure Storage: flutter_secure_storage: ^9.2.4
  • Image Picker: image_picker: ^1.1.2
  • HTTP Client: dio: ^5.8.0+1

A robust backend service that handles:

• User authentication and management
• Image analysis and processing
• AI-powered chat interactions
• Secure storage of medical data
• HIPAA-compliant data handling

Tech Stack:

• Runtime: Node.js
• Framework: Express.js
• Database: MongoDB with Mongoose
• Authentication: JWT
• Storage: Azure Blob Storage
• AI Services:
  • Microsoft Azure Custom Vision API
  • OpenRouter API with Microsoft's Phi Mini Instruct model
• Email Service: Nodemailer

An offline-capable image classification system that serves as a fallback solution for:

• Offline image analysis
• Poor network connectivity scenarios
• Primary API unavailability

Tech Stack:

• Deep Learning Framework: FastAI
• Base Model: ResNet34 (pretrained)
• Data Augmentation: Albumentations
• Loss Function: Focal Loss
• Performance Metrics:
  • Overall Accuracy: 81%
  • Macro Average F1-Score: 0.81
  • Weighted Average F1-Score: 0.81

The entire system is designed with security and compliance in mind:

• HIPAA-compliant data storage and handling
• Secure authentication using JWT
• Encrypted data transmission
• Secure storage of medical images
• Regular security updates and patches

Each component has its own repository with detailed setup instructions. Please refer to the individual README files in each repository for specific setup and installation steps.

This project is licensed under the MIT License.

Fundusnap Developers dev@fundusnap.faizath.com