Speaker: Taghi Khaniyev, Bilkent University
Date & Time: August 3, 2022, Wednesday 12:00.
Abstract: Connectome is the representation of the structural connectivity (the physical wiring) and functional connectivity (interactions between different brain regions) of an organism’s nervous system as a mathematical network (a set of nodes and the connections between them). Recent advances in medical imaging technologies that can probe the neural circuitry of human brains allowed the construction of connectivity networks from fiber (white matter) densities between different regions of interest (ROIs) in the human brain. One of the clinical areas where network neuroscience has contributed significantly is the analysis of the impact of damages to the brain regions. In this work, we assume we are given the structural connectome of a patient with brain tumor where a number of regions are marked as resectable. Our goal is to identify a maximal volume of contiguous resectable regions whose removal will cause not more than a predetermined magnitude of disruption in the network’s global efficiency. Due to the intricate nature of the networks, it is not realistic to assume that the impact of removing multiple nodes from the network together will be equal to the sum of the marginal impacts of removing them individually. We formulate this problem as a mixed integer linear programming problem, which is closely linked to the critical node/edge problem in the literature. The main challenge, after formulating the problem, is to solve real-life size problem instances with thousands of nodes for which we propose algorithmic approaches.
Bio: Taghi Khaniyev is an assistant professor at Bilkent University, Department of Industrial Engineering since 2021. He is also affiliated with UMRAM (National Magnetic Resonance Research Center) and Aysel Sabuncu Brain Research Center. He acquired his PhD in Management Science at the University of Waterloo in 2018. Prior to joining Bilkent, he was a postdoctoral fellow at MIT Sloan School of Management working in collaboration with Massachusetts General Hospital (MGH) on hospital operations management. His research in the healthcare domain focuses on developing and implementing data-driven analytical tools with the interplay of machine learning and optimization to predict outcomes of interest and prescribe personalized interventions for facilitating favorable outcomes. His main research interests are hospital operations management, deep neural networks, data-driven optimization, structure detection and decomposition in mathematical programs, and brain connectivity networks. His research has been published in prestigious scientific journals such as INFORMS Journal on Computing, European Journal of Operations Research, and Frontiers in Neuroscience; an algorithm he developed for decomposition and parallel processing of large-scale optimization problems have been adopted by the software company SAS Inc., the machine learning tool he developed for discharge prediction has become an integral part of the Capacity Coordination Center’s workflow at MGH (Harvard), a surgery duration prediction model he developed was used by Lucile Packard Children’s Hospital (Stanford), and his paper on the network optimization approach for identifying the hub regions in the human brain won the best student paper award by Canadian Operations Research Society (CORS).