deepprostate 1.5.0

AI-powered medical imaging analysis for prostate MRI

DeepProstate

AI-Powered Prostate MRI Analysis Platform

---

• 🧠 AI Segmentation: Prostate gland, TZ/PZ zones and csPCa detection via nnUNet v2
• 👁️ Multi-view Viewer: Synchronized axial, sagittal and coronal views (Single & Quad layout)
• ✏️ Manual Editing: Brush, eraser and flood-fill tools with correct orientation in all planes
• 📊 Quantitative Analysis: Volume calculations and radiomics metrics
• 🔄 Format Support: DICOM, NIfTI (.nii / .nii.gz), MHA, NRRD
• 🌐 Internationalization: Spanish and English UI
• 🖥️ Cross-platform: Windows, macOS, Linux — standalone executable or PyPI package
• 🛡️ Medical Compliance: HIPAA-compliant logging and audit trails

---

Installation

Requirements

• Python: 3.9+
• RAM: 8 GB+ recommended
• GPU: NVIDIA GPU with CUDA (optional but highly recommended for AI inference)

From PyPI (Recommended)

pip install deepprostate

From Source

git clone https://github.com/Marquita-oss/DeepProstate.git
cd DeepProstate
pip install -e .

GPU Support (Recommended for AI Analysis)

# CUDA 11.8
pip install torch torchvision --index-url https://download.pytorch.org/whl/cu118

# CUDA 12.1
pip install torch torchvision --index-url https://download.pytorch.org/whl/cu121

---

Standalone Executables

Pre-built installers are available on the Releases page — no Python installation required.

Platform	File
Windows	DeepProstate-1.5.0-windows-setup.exe
macOS	DeepProstate-macos-arm64.dmg
Linux	DeepProstate-linux-x86_64.tar.gz

---

Quick Start

deepprostate        # launch GUI (PyPI install)

Basic Workflow

1. Load Patient Data — Patient Browser → "Load DICOM Folder" or "Load Single File"
2. Load AI Models — AI Analysis panel → "Load AI Models Path" → select nnUNet models folder
3. Run AI Analysis — select image → choose analysis type → "Run AI Analysis"
4. Review & Refine — Single/Quad view, Manual Editing tools, 3D viewer
5. Export — quantitative metrics via the Quantitative Analysis panel

---

AI Models

Model	Input	Output
Prostate Gland	T2W	Complete prostate mask
Zonal Anatomy	T2W	TZ and PZ masks
csPCa Detection	T2W + ADC + HBV	Cancer lesion masks

models/
├── Task500_ProstateGland/nnUNetTrainer__nnUNetPlans__3d_fullres/
├── Task501_ProstateTZPZ/nnUNetTrainer__nnUNetPlans__3d_fullres/
└── Task502_csPCa/nnUNetTrainer__nnUNetPlans__3d_fullres/

---

Project Structure

deepprostate/
├── core/                        # Domain layer (entities, services, repositories)
├── use_cases/                   # Application layer (orchestrators)
├── frameworks/infrastructure/   # Infrastructure layer
│   ├── coordination/            # Workflow orchestration
│   ├── ui/                      # PyQt6 UI components
│   └── utils/                   # Helpers (resources, logging, i18n)
└── resources/                   # Icons, images, translations, config

---

License & Disclaimer

MIT License — see LICENSE.

⚠️ Research use only. Not FDA-approved. Not intended for clinical diagnostic decisions.

---

Citation

@software{deepprostate2026,
  title   = {DeepProstate: AI-Powered Prostate MRI Analysis Platform},
  author  = {Marca, Ronald and Salas, Rodrigo and Ponce, Sebastian and Caprile, Paola and Besa, Cecilia},
  year    = {2026},
  version = {1.5.0},
  url     = {https://github.com/Marquita-oss/DeepProstate}
}

---

Support

• Issues: GitHub Issues
• Email: [e-mail]

---

Acknowledgments

• nnUNet Team — self-configuring medical image segmentation framework
• PyQt6 — cross-platform UI framework
• VTK — 3D visualization toolkit
• Medical Imaging Community — feedback and testing

---

Made with ❤️ for the Medical Imaging Community