KCDetector

A mobile application developed to detect creatinine levels, supporting early diagnosis of kidney diseases. The app is designed for resource-limited healthcare settings and leverages machine learning and computer vision for efficient and accessible testing.

Collaboration

• Dr. Sudip Chattopadhyay | AIIMS Kalyani
• Oishila Bandyopadhyay | IIIT Kalyani

Features

• Creatinine Level Detection: Uses advanced image processing to assess creatinine levels from captured images.
• Accessibility: A lightweight Android app optimized for resource-limited environments.
• User-Friendly Interface: Built with Flet for a simple, intuitive experience.
• Machine Learning Model: Developed using TensorFlow and OpenCV for real-time, accurate predictions.

Tech Stack

• Backend: Flask
• Frontend: Flet (for Android)
• ML Model: TensorFlow, OpenCV

Project Structure

The project has two main folders:

• backend: Contains the Flask application serving API endpoints.
• frontend: Flet-based interface for Android, handling user interaction and displaying predictions.

Installation

To run the project locally, follow these steps:

Backend Setup

1. Clone the repository:

   git clone https://github.com/ProPython007/KCDetector.git
   cd KCDetector

2. Set up a virtual environment and install dependencies:

   cd backend
   python -m venv env
   source env/bin/activate
   pip install -r requirements.txt

3. Run the Flask server:

   python app.py

Frontend Setup

1. Navigate to the frontend directory:

   cd frontend

2. Install required Flet packages:

   pip install -r requirements.txt

3. Run the frontend application on an Android device (instructions available in the app or project documentation).

Usage

1. Launch the backend server.
2. Start the frontend app on an Android device.
3. Capture an image for creatinine level assessment, and receive a real-time result from the model.

Preview