Machine learning in neuroimaging: from research to clinical practice

Abstract

Neuroimaging is critical in clinical care and research, enabling us to investigate the brain in health and disease. There is a complex link between the brain’s morphological structure, physiological architecture, and the corresponding imaging characteristics. The shape, function, and relationships between various brain areas change during development and throughout life, disease, and recovery. Like few other areas, neuroimaging benefits from advanced analysis techniques to fully exploit imaging data for studying the brain and its function. Recently, machine learning has started to contribute (a) to anatomical measurements, detection, segmentation, and quantification of lesions and disease patterns, (b) to the rapid identification of acute conditions such as stroke, or (c) to the tracking of imaging changes over time. As our ability to image and analyze the brain advances, so does our understanding of its intricate relationships and their role in therapeutic decision-making. Here, we review the current state of the art in using machine learning techniques to exploit neuroimaging data for clinical care and research, providing an overview of clinical applications and their contribution to fundamental computational neuroscience.

An Umbrella Review of the Fusion of fMRI and AI in Autism

“…Artificial intelligence (AI) plays a fundamental role in the interpretation and processing of data from fMRI, providing sophisticated tools to analyze in depth the functioning of the human brain [12][13][14][15][16][17][18][19][20][21][22][23][24][25]. One high-impact area is advanced data analytics, where AI can identify complex patterns and correlations that would be difficult or impossible to identify manually [13,17,22].…”

“…This means that AI can help reveal subtle relationships between brain activity and certain stimuli or conditions, leading to a deeper understanding of cognitive and neural functions [16,20]. Furthermore, AI is essential for automating critical processes such as brain segmentation and mapping of brain regions [14,15]. This automation significantly speeds up the analysis process and ensures greater accuracy, allowing scientists to focus more on interpreting results rather than manipulating raw data [18,19,21].…”

“…Another powerful application is the prediction of brain responses to certain stimuli or tasks based on historical fMRI data [23]. AI can create predictive models that indicate how the brain might react in different situations, offering valuable insights for understanding the neural basis of different cognitive and behavioral activities [13,15,20]. Integrating multi-omics data are another important frontier where AI can contribute.…”

The Clinical Relevance of Artificial Intelligence in Migraine

“…Additionally, these models prove to be valuable in detection tasks, identifying regions of interest (ROI) within images or patterns within sequences. The segmentation of data is another field where ML showcases its efficacy, along with many other contexts [15,22]. Their application also seems to have been embedded in many neurological disciplines, including epilepsy, movement disorders, neuropsychiatric diseases, etc.…”

Diagnostic Machine Learning Applications on Clinical Populations using Functional Near Infrared Spectroscopy: A Review

“…However, these differences might not be valid for some individual cases due to huge variability across participants. At this point, the combination of neuroimaging approaches and ML techniques plays an important role in providing us some answers related to individual diagnoses rather than populations (Nenning & Langs, 2022). Previous reviews that cover a combination of ML techniques for the prediction of several diseases by using EEG (Craik et al, 2019), fMRI (de Filippis et al, 2019; Nakano et al, 2020) and PET (Duffy et al, 2019) showed that neuroimaging techniques and ML might have a future on individual diagnostic decisions.…”