About this Research Topic
There is a critical demand for noninvasive and safe neuroimaging biomarkers that facilitate the early detection of neurological and psychological disorders, monitor disease progression, and objectively assess treatment outcomes. Over the past decades, various network analysis approaches, including principal component analysis, independent component analysis, and graph theoretical analysis, have been developed for functional brain imaging such as fluorodeoxyglucose positron emission tomography (FDG PET) and resting state functional MRI (rs-fMRI). These methods aim to identify disease-related functional brain networks and explore changes in the brain's organization in these disorders. Additionally, techniques like diffusion tensor imaging, fiber tractography and connectome-based analysis have been applied to diffusion MRI to uncover disease-related structural brain networks. While these analytic approaches for PET and MRI offer opportunities to identify biomarkers for disease diagnosis and monitoring, their performance may be influenced by factors such as collection site, scanner vendor, or acquisition sequence.
Recent advances in artificial intelligence (AI) have revolutionized the study of disease-related brain networks in neuroimaging, particularly revealing patterns that elude traditional methods. In this context, deep learning models can be employed on neuroimaging data to identify and validate specialized brain networks predicting diagnostic categories or specific clinical features in individual patients. This research topic aims to establish reliable and unbiased neuroimaging biomarkers for neurological and psychological disorders across large multi-center datasets. The focus is on applying deep learning neural networks within an explainable AI framework to neuroimaging data. The approach involves characterizing disease-related networks and employing graph theoretical analysis to explore their organization. Through in- and out-of-sample testing, the study will examine the performance of disease-related networks across independent datasets and leverage explainable AI to enhance user understanding and trust in the results and outputs generated by machine learning/deep learning algorithms. This is crucial for ensuring transparency and understating in the interpretation of complex AI-generated insights.
The Research Topic scope encompasses machine learning/deep learning and explainable AI in brain imaging (PET and MRI) for humans and experimental animal models. Its applications extend to neurological disorders such as Parkinson's disease and related disorders, dementia syndromes (Alzheimer's disease, frontotemporal dementia, diffuse Lewy body disease), and other conditions including tremor, dystonia, and tic disorder. The Research Topic will also cover psychiatric conditions including schizophrenia, bipolar disorder, obsessive-compulsive disorder, and autism spectrum disorder. Key aspects of the research topic involve utilizing machine learning/deep learning neural network approaches to identify and validate reliable imaging biomarkers for these and other brain disorders. Additionally, the study incorporates the application of explainable AI to visually represent disease-related network, fostering a comprehensive understanding of the diseases. The Research Topic further engages in employing graph theoretical analysis to explore changes in the brain's organization within disease-related networks and comprehend their underlying mechanisms. The investigation probes the longitudinal progression of disease-related networks, utilizing them as a tool to assess treatment outcomes.
Recent advances in artificial intelligence (AI) have revolutionized the study of disease-related brain networks in neuroimaging, particularly revealing patterns that elude traditional methods. In this context, deep learning models can be employed on neuroimaging data to identify and validate specialized brain networks predicting diagnostic categories or specific clinical features in individual patients. This research topic aims to establish reliable and unbiased neuroimaging biomarkers for neurological and psychological disorders across large multi-center datasets. The focus is on applying deep learning neural networks within an explainable AI framework to neuroimaging data. The approach involves characterizing disease-related networks and employing graph theoretical analysis to explore their organization. Through in- and out-of-sample testing, the study will examine the performance of disease-related networks across independent datasets and leverage explainable AI to enhance user understanding and trust in the results and outputs generated by machine learning/deep learning algorithms. This is crucial for ensuring transparency and understating in the interpretation of complex AI-generated insights.
The Research Topic scope encompasses machine learning/deep learning and explainable AI in brain imaging (PET and MRI) for humans and experimental animal models. Its applications extend to neurological disorders such as Parkinson's disease and related disorders, dementia syndromes (Alzheimer's disease, frontotemporal dementia, diffuse Lewy body disease), and other conditions including tremor, dystonia, and tic disorder. The Research Topic will also cover psychiatric conditions including schizophrenia, bipolar disorder, obsessive-compulsive disorder, and autism spectrum disorder. Key aspects of the research topic involve utilizing machine learning/deep learning neural network approaches to identify and validate reliable imaging biomarkers for these and other brain disorders. Additionally, the study incorporates the application of explainable AI to visually represent disease-related network, fostering a comprehensive understanding of the diseases. The Research Topic further engages in employing graph theoretical analysis to explore changes in the brain's organization within disease-related networks and comprehend their underlying mechanisms. The investigation probes the longitudinal progression of disease-related networks, utilizing them as a tool to assess treatment outcomes.
Keywords: machine learning, deep learning, explainable artificial intelligence, neuroimaging, network analysis organization
Important Note: All contributions to this Research Topic must be within the scope of the section and journal to which they are submitted, as defined in their mission statements. Frontiers reserves the right to guide an out-of-scope manuscript to a more suitable section or journal at any stage of peer review.
