<div dir="ltr"><div><font face="georgia, serif" color="#000000">Dear Colleagues,</font></div><div><font face="georgia, serif" color="#000000"><br></font></div><div><font face="georgia, serif" color="#000000">We are writing to let you know that we are organizing a special issue "<span class="gmail-il">Graph</span> <span class="gmail-il">Learning</span> for <span class="gmail-il">Brain</span> Imaging" in Frontiers in Neuroscience (impact factor 4.7). We believe this is a timely special issue to showcase the new developments using <span class="gmail-il">graph</span> representation, deep <span class="gmail-il">learning</span> on <span class="gmail-il">graph</span>-structured data to address important <span class="gmail-il">brain</span> <span class="gmail-il">imaging</span> and computational neuroscience problems. </font></div><div><font face="georgia, serif"><br></font></div><div><p style="margin:0px;line-height:1.5em;outline:0px"><font face="georgia, serif" color="#000000"><b>Link</b>: <a href="https://www.frontiersin.org/research-topics/23683/graph-learning-for-brain-imaging" target="_blank">https://www.frontiersin.org/research-topics/23683/graph-<span class="gmail-il">learning</span>-for-<span class="gmail-il">brain</span>-<span class="gmail-il">imaging</span></a></font></p><p style="margin:0px;line-height:1.5em;outline:0px"><font face="georgia, serif" color="#000000"><br></font></p><p style="margin:0px;line-height:1.5em;outline:0px"><font face="georgia, serif"><span style="color:rgb(0,0,0);outline:0px"><b>Keywords</b></span><span style="color:rgb(0,0,0)">: <span class="gmail-il">Brain</span> Networks, Graph Neural Networks, <span class="gmail-il">Brain</span> <span class="gmail-il">Imaging</span>, <span class="gmail-il">Graph</span> Embedding, Multi-Modal <span class="gmail-il">Imaging</span>.</span></font></p><p style="margin:0px;line-height:1.5em;outline:0px"><span style="color:rgb(0,0,0)"><font face="georgia, serif"><br></font></span></p></div><div><font face="georgia, serif" color="#000000"><b>Topics</b>: We are looking for original, high-quality submissions on innovative research and developments in the analysis of <span class="gmail-il">brain</span> <span class="gmail-il">imaging</span> using <span class="gmail-il">graph</span> learning techniques. Topics of interest include (but are not limited to):<br style="outline:0px"><br style="outline:0px">• <span class="gmail-il">Graph</span> neural networks (GNN) for network neuroscience applications<br style="outline:0px">• <span class="gmail-il">Graph</span> neural network for <span class="gmail-il">brain</span> mapping and data integration<br style="outline:0px">• <span class="gmail-il">Graph</span> convolution network (GCN) for <span class="gmail-il">brain</span> disorder classification<br style="outline:0px">• (Dynamic) Functional <span class="gmail-il">brain</span> networks<br style="outline:0px">• <span class="gmail-il">Brain</span> networks development trajectories<br style="outline:0px">• Graphical model for <span class="gmail-il">brain</span> <span class="gmail-il">imaging</span> data analysis<br style="outline:0px">• Spatial-temporal <span class="gmail-il">brain</span> network modeling<br style="outline:0px">• <span class="gmail-il">Graph</span> embedding and <span class="gmail-il">graph</span> representation learning<br style="outline:0px">• Information fusion for <span class="gmail-il">brain</span> networks from multiple modalities or scales (fMRI, M/EEG, DTI, PET, genetics)<br style="outline:0px">• Generative <span class="gmail-il">graph</span> models in <span class="gmail-il">brain</span> <span class="gmail-il">imaging</span><br style="outline:0px">• <span class="gmail-il">Brain</span> network inference: scalable, online, and from non-linear relationships<br style="outline:0px">• Machine <span class="gmail-il">learning</span> over graphs: kernel-based techniques, clustering methods, scalable algorithms for <span class="gmail-il">brain</span> <span class="gmail-il">imaging</span><br style="outline:0px">• A few-shot <span class="gmail-il">learning</span> for <span class="gmail-il">learning</span> from limited <span class="gmail-il">brain</span> data<br style="outline:0px">• <span class="gmail-il">Graph</span> federated <span class="gmail-il">learning</span> for <span class="gmail-il">brain</span> <span class="gmail-il">imaging</span></font></div><div><font face="georgia, serif" color="#000000"><br></font></div><div><font face="georgia, serif" color="#000000"><b>Important Dates</b>: </font></div><div><font face="georgia, serif" color="#000000">Abstract: 30-Sep-2021 </font></div><div><font face="georgia, serif" color="#000000">Full paper: 30-Dec-2021</font></div><div><font face="georgia, serif" color="#000000"><br></font></div><div><font face="georgia, serif" color="#000000"><b>Background:</b></font></div><div><p style="margin:0px;line-height:1.5em;outline:0px"><font face="georgia, serif" color="#000000">Unprecedented collections of large-scale <span class="gmail-il">brain</span> <span class="gmail-il">imaging</span> data, such as MRI, PET, fMRI, M/EEG, DTI, etc. provide a unique opportunity to deepen our understanding of the <span class="gmail-il">brain</span> working mechanisms, improve prognostic predictions for mental disorders, and tailor personalized treatment plans for <span class="gmail-il">brain</span> diseases. Recent advances in machine <span class="gmail-il">learning</span> and large-scale <span class="gmail-il">brain</span> <span class="gmail-il">imaging</span> data collection, storage, and sharing lead to a series of novel interdisciplinary approaches among the fields of computational neuroscience, signal processing, deep <span class="gmail-il">learning</span>, <span class="gmail-il">brain</span> <span class="gmail-il">imaging</span>, cognitive science, and computational psychiatry, among which <span class="gmail-il">graph</span> <span class="gmail-il">learning</span> provides a valuable means to address important questions in <span class="gmail-il">brain</span> <span class="gmail-il">imaging</span>.<br style="outline:0px"><br style="outline:0px"><span class="gmail-il">Graph</span> <span class="gmail-il">learning</span> refers to designing effective machine <span class="gmail-il">learning</span> and deep <span class="gmail-il">learning</span> methods extracting important information from graphs or exploiting the <span class="gmail-il">graph</span> structure in the data to guide the knowledge discovery. Given the complex data structure in different <span class="gmail-il">imaging</span> modalities as well as the networked organizational structure of the human <span class="gmail-il">brain</span>, novel <span class="gmail-il">learning</span> methods based on graphs inferred from <span class="gmail-il">imaging</span> data, <span class="gmail-il">graph</span> regularizations for the data, and <span class="gmail-il">graph</span> embedding of the recorded data, have shown great promise in modeling the interactions of multiple <span class="gmail-il">brain</span> regions, information fusion among networks derived from different <span class="gmail-il">brain</span> <span class="gmail-il">imaging</span> modalities, latent space modeling of the high dimensional <span class="gmail-il">brain</span> networks, and quantifying topological neurobiomarkers. The goal of this Research Topic is to synergize the start-of-the-art discoveries in terms of new computational <span class="gmail-il">brain</span> <span class="gmail-il">imaging</span> models and insights of <span class="gmail-il">brain</span> mechanisms through the lens of <span class="gmail-il">brain</span> networks and <span class="gmail-il">graph</span> <span class="gmail-il">learning</span>.<br style="outline:0px"><br style="outline:0px"></font></p><p style="margin:0px;line-height:1.5em;outline:0px"><font face="georgia, serif" color="#000000"><br></font></p><p style="margin:0px;line-height:1.5em;outline:0px"><font face="georgia, serif">--On Behalf of all the Guest Editors</font></p><p style="margin:0px;line-height:1.5em;outline:0px"><font face="georgia, serif">Feng Liu, Stevens Institute of Technology, Hoboken, NJ, USA</font></p><p style="margin:0px;line-height:1.5em;outline:0px"><font face="georgia, serif">Yu Zhang, Lehigh University, <span style="color:rgb(2,2,2)">Bethlehem, PA, USA</span></font></p><h3 style="margin:0px;line-height:18px;color:rgb(35,31,32);padding:0px;outline:0px"><span style="font-weight:normal"><font size="2" face="georgia, serif">Jordi Solé-Casals, Universitat de Vic - Universitat Central de Catalunya, Barcelona, Spain</font></span></h3><h3 style="margin:0px;line-height:18px;color:rgb(35,31,32);padding:0px;outline:0px"><font size="2" face="georgia, serif" style="font-weight:normal">Islem Rekik, Istanbul Technical University, Istanbul, Turkey</font></h3><h3 style="margin:0px;line-height:18px;color:rgb(35,31,32);padding:0px;outline:0px"><font size="2" face="georgia, serif" style="font-weight:normal">Yehia Massoud, King Abdullah University of Science and Technology, Thuwal, Saudi Arabia</font></h3></div><div><font size="2" face="georgia, serif"><br></font></div><div><font size="2" face="georgia, serif">Best regards</font></div><div><font size="2" face="georgia, serif">Feng Liu</font></div><font color="#888888"><div><font size="2" face="georgia, serif"><br></font></div></font></div>