Machine learning in neuroimaging: Progress and challenges

Abstract

It has been a privilege to serve the Neuroimage community as a Handling Editor, with primary focus on machine learning, over the past six years. As I step down, I was asked to provide my thoughts on the evolution of the field over the past 10–15 years, as well as its potential future challenges and directions. It is exciting to reflect back on the spectacular growth of this field over the past decade (Fig. 1), and to ponder on some of the challenges faced currently by the field, and which are likely to become central points of investigation in the upcoming decade. Open in a separate window Fig. 1. Publications obtained from Pubmed with the following search query: (MRI” OR “Magnetic Resonance Imaging” OR Structural Magnetic Resonance Imaging OR Functional Magnetic Resonance Imaging OR Diffusion Tensor Imaging OR FDG-PET” OR Amyloid-PET” OR “multimodal OR neuroimaging) AND (“brain”) AND (“Machine learning” OR “pattern classification”), on September 5, 2018. This plot reflects the exponential growth of adoption of these methods in neuroimaging.