Medical imaging encompasses several techniques; however, the interpretation of the data and their use for identifying and detecting biomarkers of disease can be quite difficult. The use of standardized computational aids has been proven very effective in overcoming this difficulty. Deep learning (DL) is a sub-group of machine learning algorithms capable of automatically extracting discriminatory features from raw input. This capacity makes DL a powerful technique for supporting the diagnosis of several diseases and disorders, including autism spectrum disorder, Alzheimer’s disease, Parkinson’s disease.
Although feature extraction requires explainable methods for complex models, the potential of DL to be useful also to basic research - not necessarily related to diagnosis – is high. DL can be applied to several brain imaging methods, such as MRI, CT, PET, and more. The aim of this Research Topic is to highlight the potential of applying DL techniques to neuroimaging methods and applications. We welcome articles ranging from Original Research to Reviews to Methods to further elucidate innovations and applications of DL to imaging the developing-, adult-, and aging brain in health and disease.
Examples of topics include: classification and regression task related to diagnosis, prognosis, and treatment planning, interpretable methods for medical imaging, methods to overcome sparse label issues in medical imaging.
Medical imaging encompasses several techniques; however, the interpretation of the data and their use for identifying and detecting biomarkers of disease can be quite difficult. The use of standardized computational aids has been proven very effective in overcoming this difficulty. Deep learning (DL) is a sub-group of machine learning algorithms capable of automatically extracting discriminatory features from raw input. This capacity makes DL a powerful technique for supporting the diagnosis of several diseases and disorders, including autism spectrum disorder, Alzheimer’s disease, Parkinson’s disease.
Although feature extraction requires explainable methods for complex models, the potential of DL to be useful also to basic research - not necessarily related to diagnosis – is high. DL can be applied to several brain imaging methods, such as MRI, CT, PET, and more. The aim of this Research Topic is to highlight the potential of applying DL techniques to neuroimaging methods and applications. We welcome articles ranging from Original Research to Reviews to Methods to further elucidate innovations and applications of DL to imaging the developing-, adult-, and aging brain in health and disease.
Examples of topics include: classification and regression task related to diagnosis, prognosis, and treatment planning, interpretable methods for medical imaging, methods to overcome sparse label issues in medical imaging.
