Welcome to the AI for Medical Imaging Bootcamp, a self-paced introduction to building intelligent systems that analyze, interpret, and learn from medical images.
This bootcamp combines foundational AI concepts with hands-on modules that guide you from data preprocessing to segmentation, biomarker discovery, and report generation.
The goal of this bootcamp is to help you:
- Understand the core principles of AI in medical imaging.
- Learn how to design and implement end-to-end imaging workflows.
- Gain hands-on experience with open-source tools such as PyRadiomics, MERLIN, nnU-Net, and TotalSegmentator.
- Build confidence to contribute to research in the BioIntelligence Lab or launch your own imaging AI projects.
| Phase | Focus | Modules | Description |
|---|---|---|---|
| 🐍 Module 0 (Optional) | Python Foundations | — | For beginners: complete the Kaggle Python Course. |
| ⚙️ Core Bootcamp | Fundamentals of AI for Medical Imaging | 1 – 4 | Learn data science, image processing, ML, DL, and segmentation. |
| 🧬 Application Bootcamp | Imaging Biomarkers & Beyond | 5 | Apply AI techniques to real-world medical imaging workflows. |
These modules build your foundation in medical imaging AI — from basic data science to segmentation model training.
Each module can be opened directly in Google Colab — just click and start learning.
| Module | Description | Link |
|---|---|---|
| 1. Data Science Foundations | Learn Python, NumPy, and Pandas for data wrangling, visualization, and exploratory analysis. | Open in Colab → |
| 2. Image Processing | Explore image filtering, enhancement, and region-based operations using OpenCV and SimpleITK. | Open in Colab → |
| 3. Machine Learning | Build and evaluate classical ML models using scikit-learn on imaging datasets. | Open in Colab → |
| 4. Introduction to Deep Learning | Understand CNN architectures and how deep networks process images using Keras and fast.ai. | Open in Colab → |
This module aims to connect foundational knowledge to a real-world application.
| Module | Learning Goal | Link |
|---|---|---|
| 5. Imaging Biomarkers — From Data to Prediction | Build a complete imaging biomarker pipeline: cohort discovery → data download → segmentation → feature extraction → predictive modeling. | Open in Colab → |
BioIntelligence Lab — UTHealth Houston
👨🏫 Dr. Vishwa S. Parekh
📧 [email protected]
🌐 BioIntelligence-Lab GitHub