This paper explores the application of statistical machine learning methods to neuroimaging data, focusing on the use of the scikit-learn Python library. Scikit-learn is a versatile tool that includes a wide range of supervised and unsupervised learning algorithms, making it suitable for various neuroimaging tasks. The authors provide detailed examples of how to prepare neuroimaging data for analysis, apply supervised learning techniques to link brain images with behavioral or clinical observations, and use unsupervised learning to uncover hidden structures in resting-state functional MRI data. They also discuss the interpretation of results and the importance of proper data preprocessing. The paper highlights the benefits of using scikit-learn for neuroimaging, emphasizing its simplicity and modularity, which make it accessible to both machine learning experts and neuroscientists. Additionally, the authors mention the development of nilearn, a project aimed at simplifying the use of scikit-learn for neuroimaging, and provide code snippets for generating figures and performing analyses. Overall, the paper demonstrates how machine learning can be effectively applied to neuroimaging data to address complex scientific questions.This paper explores the application of statistical machine learning methods to neuroimaging data, focusing on the use of the scikit-learn Python library. Scikit-learn is a versatile tool that includes a wide range of supervised and unsupervised learning algorithms, making it suitable for various neuroimaging tasks. The authors provide detailed examples of how to prepare neuroimaging data for analysis, apply supervised learning techniques to link brain images with behavioral or clinical observations, and use unsupervised learning to uncover hidden structures in resting-state functional MRI data. They also discuss the interpretation of results and the importance of proper data preprocessing. The paper highlights the benefits of using scikit-learn for neuroimaging, emphasizing its simplicity and modularity, which make it accessible to both machine learning experts and neuroscientists. Additionally, the authors mention the development of nilearn, a project aimed at simplifying the use of scikit-learn for neuroimaging, and provide code snippets for generating figures and performing analyses. Overall, the paper demonstrates how machine learning can be effectively applied to neuroimaging data to address complex scientific questions.