DICOM Flask App – Multi-Tab Lesion Detector 📌 Overview This Flask web app allows users to upload DICOM files, detect lesions using DeepLesion (Faster R-CNN), and classify them with ResNet50. The UI features a multi-tab layout with a sidebar for smooth navigation.

✅ Upload & Process DICOM Files ✅ Detect & Classify Lesions Automatically ✅ Multi-Tab UI with Sidebar Navigation ✅ Sample DICOM File Included for Testing

🛠 Installation & Setup 1️⃣ Clone the Repository bash Copy Edit git clone https://github.com/YOUR_GITHUB_USERNAME/DICOM-Flask-App.git cd DICOM-Flask-App 2️⃣ Install Dependencies bash Copy Edit pip install -r requirements.txt (If requirements.txt is missing, install manually:)

bash Copy Edit pip install flask torch torchvision pydicom numpy matplotlib opencv-python 3️⃣ Run the Flask App bash Copy Edit python app.py Then, open http://127.0.0.1:5000/ in your browser.

📂 Project Structure graphql Copy Edit DICOM-Flask-App/ │── static/ # CSS & processed images │ ├── style.css # UI Styling (Velvet Room Theme) │── templates/ # HTML Templates for Flask │ ├── index.html # Main UI (Tabs: Upload, Results) │── uploads/ # Stores uploaded DICOM files │── sample.dcm # Sample DICOM file for testing ✅ │── app.py # Flask Application │── requirements.txt # Dependencies │── README.md # This documentation 🚀 How to Use 1️⃣ Upload a DICOM File Go to http://127.0.0.1:5000/ Click "Upload DICOM", select a file, and click "Upload & Process" 2️⃣ View Results Click the "Results" tab to see detected lesions. Lesions are shown with bounding boxes and classified as Tumor, Cyst, Hemorrhage, or Inflammation. 📝 Sample DICOM File A sample DICOM file (se.dcm) is included in the repo for convenience.

If you don’t have a DICOM file, use this one for testing. 🔥 Future Upgrades 🔹 Grad-CAM Heatmaps – Highlight lesion focus areas. 🔹 DICOM Export – Save processed images back into DICOM format. 🔹 Automatic Report Generation – AI-generated text reports for findings.