X-Reporto

Table of Contents

• Overview
• Built Using
• System Design
• Get Started
• Contributors
• License

Overview

X-Reporto is a web-based application designed to support radiologists by automating the reporting process for chest X-ray images. The tool enhances image quality to correct X-ray device defects, generates detailed saliency maps to highlight key findings, and produces comprehensive, template-based reports tailored to each anatomical region and identified disease. By streamlining the diagnostic workflow, X-Reporto reduces report generation time and improves diagnostic accuracy, ultimately aiding in faster and more precise medical interventions.

Built Using

System Design

Repositories

• AI Repository: Contains all the AI modules for the X-ray analysis project, including object detection, classifiers, heatmap generation, and denoising models.
• Frontend Repository: Includes the user interface for the X-ray application, designed using React and Ant Design. It handles the presentation and interaction aspects of the application.
• Backend Repository: Hosts the server-side code and APIs for the X-ray project, built with Flask/FastAPI. It manages data processing, model inference, and interactions between the frontend and AI components.
• ML Repository: Contains all the ML modules for the X-ray analysis project, including object detection, feature extraction and denoising models.

Get Started

Native

AI Server

1. Clone the Repository

   git clone https://github.com/BasmaElhoseny01/X-Reporto-AI.git
   cd X-Reporto-AI

2. Install Dependencies

   pip install -r requirements.txt

3. Download Model Weights

   # Download X-Reporto Models
   curl -o /models/object_detector.pth https://path-to-your-models/object_detector_best.pth
   curl -o /models/region_classifier.pth https://path-to-your-models/region_classifier_best.pth
   curl -o /models/LM.pth https://path-to-your-models/LM_best.pth
   
   # Download Heat Map Models
   curl -o /models/heat_map.pth https://path-to-your-models/heat_map_best.pth

4. Run AI Server

   python -m src.app

Backend Server

1. Clone the Repository

   git clone https://github.com/BasmaElhoseny01/X-Reporto-Backend.git
   cd X-Reporto-Backend

2. Install Dependencies

   pip install -r requirements.txt

3. Start the Backend Server

   # Command to start the backend server

Client Server

1. Clone the Repository

   git clone https://github.com/BasmaElhoseny01/X-Reporto-Frontend.git
   cd X-Reporto-Frontend

2. Install Dependencies

   npm install

3. Start the Client Server

   npm start

Contributors

Ahmed Sabry

Ahmed Hosny

Zeinab Moawed

Basma Elhoseny

License

This software is licensed under the MIT License. See License for more information ©Basma Elhoseny.

---

This project is sponsored by