Abstract
In this chapter, we provide a step-by-step tutorial on the implementation of a standard supervised machine learning pipeline using Python programming language. We use a toy dataset with neuroimaging-based data (i.e., gray matter volume and thickness from different brain regions extracted with FreeSurfer) to classify patients with schizophrenia and healthy controls using a Support Vector Machine. Both the toy dataset and the source code are available to download. This tutorial is aimed at the machine learning novice and assumes minimal programming experience. Instructions on how to install Python and the necessary libraries are provided. In the last section, we discuss how the analysis pipeline can be improved and how the sample code can be adapted to other neuroimaging-based data and other data modalities that are used in brain disorders research.
| Original language | English |
|---|---|
| Title of host publication | Machine Learning |
| Subtitle of host publication | Methods and Applications to Brain Disorders |
| Publisher | Elsevier |
| Pages | 343-370 |
| Number of pages | 28 |
| ISBN (Electronic) | 9780128157398 |
| ISBN (Print) | 9780128157398 |
| DOIs | |
| Publication status | Published - 2020 |
Keywords
- Classification
- Cross-validation
- Machine learning
- Neuroimaging
- Permutation test
- Python
- Sample code
- Schizophrenia
- Support vector machine
- Tutorial