Deep Learning for Multimodal Neuroimaging

Dear Colleagues,

Deep learning (DL) is a family of machine learning methods that has gained considerable attention in the scientific community, showing its performance in speech, visual recognition, and recently in brain neuroimaging. DL differs from conventional machine learning methods due to its ability to learn the optimal representation from the raw data through consecutive nonlinear transformations, achieving increasingly higher levels of abstraction and complexity. Given its ability to detect abstract and complex patterns, DL has been applied in neuroimaging studies of neurodegenerative, neurodevelopmental, psychiatric, and neurological disorders characterized by subtle and diffuse alterations.

Deep learning has recently been used to analyze neuroimages, such as structural magnetic resonance imaging (MRI), functional MRI, and positron emission tomography (PET), and it has achieved significant performance improvements over traditional machine learning computer-aided diagnosis of brain disorders and diseases. The principal aim of this Special Issue is to collect studies related to various types of deep neural networks and network architectures, their comparison, and validation applied to non-invasive multimodal neuroimaging techniques, including (but not limited to):

• Electroencephalography (EEG);
• Magnetoencephalography (MEG);
• Magnetic resonance imaging (MRI);
• Functional MRI (fMRI);
• Diffusion MRI (dMRI);
• Diffusion tensor imaging (DTI);
• Positron emission tomography (PET);
• Functional near-infrared spectroscopy (fNIRS);
• Utilizing cutting-edge artificial intelligence neurodiagnostic schemes;
• Neurotherapeutic approaches of non-invasive brain stimulation;
• Transcranial magnetic stimulation (TMS);
• Transcranial electric stimulation (TES);
• Temporal interference stimulation (TIS). 

Moreover, DL can investigate the discriminative power of neuroimaging data related to (but not limited to) the following brain disorders and diseases:

• Autism spectrum disorder;
• Schizophrenia;
• Bipolar disorder;
• Dyslexia;
• Dyscalculia;
• Dementia (including Alzheimer's disease and preclinical stages);
• Parkinson’s disease;
• Traumatic brain injury.

We thus welcome contributions to new architectures, learning strategies, and testing of DL. Below, we summarize potential research thematic areas: 

• Predicting the brain evolution trajectory;
• Population network integration and fusion;
• Computer-aided prognostic methods (e.g., for brain diseases);
• Biomarker identification;
• Comprehensive surveys on deep learning algorithms for multimodal neuroimaging dedicated to a brain disease/disorder;
• Interpretable deep learning.

Special care will be given to graph neural networks (GNN). This Special Issue will also attract papers related to one of the most popular GNNs, geometric deep learning (GDL). GDL applies convolutional layers to learn the network topology of the input graph or brain network. GNNs have recently been used for the analysis of different types of the human connectome, such as structural, functional, and morphological networks extracted respectively from diffusion tensor imaging (DTI)/diffusion magnetic resonance imaging (dMRI), functional magnetic resonance imaging (fMRI), and T1-w MRI data. 

Reviews of the applications of deep learning methods for neuroimaging-based brain disorder analysis will also be accepted.

We invite original research articles to share at least metadata from open or not neuroimaging datasets and the related code for reproducibility of the findings. All the open websites and services for uploading code and metadata will be accepted.

Dr. Stavros I. Dimitriadis
Guest Editor

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 100 words) can be sent to the Editorial Office for announcement on this website.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Algorithms is an international peer-reviewed open access monthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 1600 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.

• brain diseases
• multimodal neuroimaging techniques
• deep learning
• artificial intelligence
• brain connectome