ScanInsight — Intelligent Multi-Disciplinary CT Diagnostics

ScanInsight is a next-generation, full-stack medical AI platform. It provides a unified, multi-disciplinary diagnostic analysis of DICOM CT scans across six major medical specialties: Neurology, Cardiology, Pulmonology, Oncology, Orthopedics, and Abdominal medicine.

🌟 Key Features

• Multi-Specialty Analysis: A single scan is concurrently analyzed by 6 specialized AI models.
  • 🧠 Neurology: Detects brain lesions, hemorrhages, midline shifts.
  • ❤️ Cardiology: Identifies coronary calcifications, aortic aneurysms.
  • 🫁 Pulmonology: Screens for nodules, consolidations, emphysema.
  • 🧬 Oncology: Cross-references findings to flag potential tumors.
  • 🦴 Orthopedics: Identifies fractures, bone density anomalies.
  • 🔍 Abdominal: Evaluates liver, kidneys, spleen for masses and cysts.
• Interactive Scan Viewer: A web-based DICOM viewer with tools to Pan, Zoom, Measure, Window/Level, and navigate slices, featuring AI-powered annotation overlays.
• Diagnostic Dashboard: Multi-specialty AI analysis results presented in a clear, actionable format.
• Cross-Disciplinary Correlation: AI-discovered relationships between findings across different medical specialties.
• Triage Dashboard: Priority-ranked findings with recommended actions and timelines.
• Automated Reporting: Generates comprehensive, unified multi-disciplinary diagnostic reports.
• Patient History Tracking: Review and manage previously generated diagnostic reports.

🛠️ Technology Stack

• Frontend: Vanilla HTML5, CSS3, and JavaScript. Responsive UI with dynamic routing.
• Backend: Python, FastAPI.
• Medical Imaging: PyDicom, NiBabel.
• Machine Learning: PyTorch, TorchVision, MONAI.

🚀 Getting Started

Prerequisites

• Python 3.8+
• Python (for serving the frontend)

Running the Backend

1. Navigate to the backend directory.
2. Activate your virtual environment (if applicable).
3. Install dependencies:

   pip install -r requirements.txt

4. Start the FastAPI server:

   python -m uvicorn main:app --reload --host 0.0.0.0 --port 8000

   The backend API will be available at http://localhost:8000 and Swagger docs at http://localhost:8000/docs.

Running the Frontend

1. Navigate to the root directory of the project.
2. Start a local HTTP server. For example, using Python:

   python -m http.server 8080

3. Open your browser and navigate to http://localhost:8080.

📂 Project Structure

• /backend - FastAPI backend application, AI inference scripts, and database models.
• /css - Stylesheets for the frontend UI.
• /js - Frontend logic, including API interaction, routing, and DICOM viewer functionality.
• index.html - The main entry point for the single-page application.