Imaging + Machine Learning

Multimodal imaging (MRI/CT), radiomics, and machine learning to connect imaging signatures with tissue-level phenotypes and clinical outcomes.

Key methods:

Glossary: Radiomics · Interpretable ML · Multi-omics

Key outputs

• Dataset/tool: Selected code + tools live on my GitHub (https://github.com/madavid128) and on the Projects page under Software & tools.

• Paper: Search the Publications page for imaging/ML/radiomics work:

  Machine learning radiomics imaging
• Figure: Representative images appear in Images.

What I work on

• Imaging-derived biomarkers and radiomics features
• Interpretable models that link imaging to biology
• Reproducible pipelines for translational deployment

Key questions

• What imaging signatures track tissue-level phenotypes and outcomes?
• How can models stay interpretable enough for translational decision-making?
• Which workflows generalize across datasets and sites?

Methods I use

• Radiomics (glossary) and multimodal imaging feature extraction (MRI/CT where available)
• Interpretable machine learning (glossary) and careful validation
• Reproducible pipelines (versioned code, containers, and traceable metadata)

Related publications

Images