An open-source multi-agent system for automated end-to-end development of radiomics and deep learning models for medical imaging
mAIstro is an autonomous, open-source multi-agent system designed to orchestrate the full pipeline of medical imaging AI development - from exploratory data analysis (EDA) and radiomics feature extraction to training and deploying deep learning models.
Built around a team of specialized agents, mAIstro enables researchers and clinicians to interact with complex AI workflows using natural language prompts - no coding required.
🌐 LLM-Agnostic Design: mAIstro can operate with both open-source and commercial LLMs (e.g., GPT-4, Claude, DeepSeek, LLaMA, Qwen), providing flexibility across environments.
⭐ If you find mAIstro useful, please consider starring the repository to support the project and help others discover it!
- 🔎 Autonomous Exploratory Data Analysis (EDA)
- 🧬 Radiomics feature extraction (for CT, MRI, and multi-parametric imaging)
- ⚙️ nnU-Net Agent for segmentation model development and implementation
- ⚙️ TotalSegmentator Agent for full-body and organ-specific automatic segmentation
- 🩻 Image Classification Agent (ResNet, VGG16, InceptionV3 architectures)
- 📊 Feature Importance and Feature Selection
- 📈 Tabular data Classification and Regression Agents
- 🛠️ Modular tool-based architecture for extensibility
- 🧾 Integrated in a single user-friendly Jupyter Notebook
🚀 v1.0.4r released (04/02/2026) - mAIstro Desktop now available on Windows and Linux.
✨ Includes enhanced capabilities and performance improvements.
🧠 The latest versions of mAIstro tools are embedded in the mAIstro Desktop App.
Experience the power of mAIstro - your intelligent desktop assistant for building complete AI pipelines in medical imaging.
From radiomic feature extraction to the development, evaluation, and deployment of segmentation, classification, and regression AI models, everything happens right on your desktop by simply chatting with mAIstro.
👉 Learn more and get download & installation instructions
👉 Instructions to set up Docker and run mAIstro
Want to try mAIstro without setting up anything locally?
You can now run the full framework interactively on Google Colab
What’s included:
All necessary requirements are automatically installed
The full mAIstro_workspace folder is downloaded, including:
✅Experiment data
✅Predefined folders and structure
✅Ready-to-run example prompts pointing to the correct locations
🔐 What you need to do: The only manual step is to provide your API key for the LLM of your choice (e.g., OpenAI, Claude, DeepSeek, etc.).
A pre-configured cell is provided with options for multiple LLM providers – just paste your key and you're ready to go!
This makes it easy to explore and test the full functionality of mAIstro on any device, using just your browser.
Watch the tutorial on Google Drive
The following tools are part of the mAIstro multi-agentic system. Each tool is described in detail in its own documentation file:
- Radiomics Tool
- Exploratory Data Analysis Tool
- Feature Selection Tool
- nnUNet Training Tool
- nnUNet Inference Tool
- TotalSegmentator Tool
- PyCaret Classification Training Tool
- PyCaret Classification Inference Tool
- PyCaret Regression Training Tool
- PyCaret Regression Inference Tool
- Medical Image Classification Tools
This project is licensed under the Apache License 2.0.
You are free to use, modify, and distribute this software under the terms of the license.
If you use mAIstro in your research, please cite:
Eleftherios Tzanis, Michail E. Klontzas (2025). mAIstro: An open-source multi-agent system for automated end-to-end development of radiomics and deep learning models for medical imaging. European Journal of Radiology Artificial Intelligence, 4, 100044. https://doi.org/10.1016/j.ejrai.2025.100044