Synopsis

Machine learning builds predictive models from the data. It is massive used on medical images these days, for a variety of applications ranging from segmentation to diagnosis.I will give an introductory tutorial to machine learning from a statistical point of view. I will introduce the methodology, the concepts behind the central models, the validation framework and a variety of caveats to look for.I will also discuss some applications to drawing conclusions from brain imaging, and use these applications to highlight various technical issues to have in mind when running machine learning models and interpreting their results.

Provisional outline:

Definitions and intuitions on machine learning

• Components of a models
• Fitting procedures
• Overfitting
• Regularization and priors

Model evaluation

• Cross-validation
• Impact of confounds and biases
• Statistical tests of model prediction accuracy

Unsupervised learning

• Clustering
• Linear decompositions: PCA, ICA, and dictionnary learning

A glance at a few models

• Linear models, non-sparse and sparse
• Random forests
• Gradient boosted trees

Learning on full-brain images

• Interpreting models weights
• Spatial regularizations for linear models

Learning on correlations in rest activity

• Dictionary learning for data reduction
• Vectorizing correlation matrices

Acknowledgements

References