MEG-Based Computational Biomarkers for Neurological and Neuropsychiatric Disorders

Neurological and neuropsychiatric disorders pose significant global health challenges, often characterized by complex and heterogeneous pathophysiologies that hinder accurate diagnosis, prognosis, and treatment monitoring. Traditional clinical assessments can lack objectivity and sensitivity, necessitating the development of robust, quantitative tools. Magnetoencephalography (MEG) offers a unique window into brain function, directly measuring neuronal current activity with exceptional temporal resolution and good spatial localization. Its non-invasive nature and ability to capture dynamic neural network interactions make it an invaluable tool for probing brain function. However, the sheer complexity and high dimensionality of MEG data require sophisticated computational approaches to extract relevant, disease-specific features, paving the way for data-driven biomarkers that reflect underlying brain states and pathological processes.This Research Topic aims to address the critical need for objective, sensitive, and specific biomarkers that can aid in the early detection, differential diagnosis, prognosis, and personalized treatment of neurological and neuropsychiatric disorders. Despite the rich information contained within MEG recordings, the development and clinical translation of robust MEG-based computational biomarkers remain an ongoing challenge. We seek to highlight cutting-edge research that leverages advanced computational neuroscience methodologies to transform raw MEG data into meaningful quantitative indicators of brain health and disease. By fostering interdisciplinary contributions, this topic will explore innovative signal processing techniques, network analyses, machine learning algorithms, and theoretical models applied to MEG data. The ultimate goal is to accelerate the discovery and validation of computational biomarkers that can objectively characterize disease states, track progression, predict treatment response, and provide deeper insights into the neuropathological mechanisms underpinning these complex conditions, thereby bridging the gap between fundamental research and clinical utility.This Research Topic seeks contributions focusing on the development, validation, and application of MEG-based computational biomarkers for neurological and neuropsychiatric disorders. We welcome original research, reviews, methodological papers, and opinion pieces that present novel computational approaches, demonstrate clinical relevance, or provide significant advancements in understanding disease mechanisms through MEG.Specific themes include, but are not limited to:- Advanced MEG signal processing and source localization techniques for biomarker extraction.- Machine learning and artificial intelligence applications for classifying and predicting neurological/neuropsychiatric conditions from MEG data.- Network-based analyses (e.g., functional connectivity, graph theory, dynamic network modeling) derived from MEG.- Computational modeling and theoretical frameworks explaining MEG abnormalities in disease.- MEG biomarkers for specific disorders such as Alzheimer’s disease, Parkinson’s disease, epilepsy, stroke, depression, schizophrenia, ADHD, and autism spectrum disorders.- Clinical validation, reproducibility, and translation of MEG biomarkers.- Integration of MEG with other neuroimaging modalities (e.g., fMRI, EEG) or clinical data for biomarker discovery. Topic editors: Jintao Sun, Moein Esghaei, Carlos S. Lima, Takefumi Ohki, Siamak Sorooshyari Source: https://www.frontiersin.org/research-topics/78171/meg-based-computational-biomarkers-for-neurological-and-neuropsychiatric-disorders