Speaker: Taghi Khaniyev, Bilkent University
Date & Time: October 15, 2021, Friday, 13:30
Place: EE – 01
Title: Prescriptive approaches in healthcare operations and network neuroscience domains
Abstract: In the first part of this talk, I will be speaking about a joint work with Massachusetts General Hospital and MIT Sloan School of Management on predicting inpatient discharges. We propose a prescriptive approach to identify a list of dischargeable patients and prescribe associated interventions for their timely transition out of the hospital. The proposed approach starts with representing patients’
clinical and administrative barriers to discharge in a clinically interpretable way, followed by building a neural network model to predict discharge likelihood of a given patient within 24 hours.
Finally, using the trained neural network coefficients as parameters, we employ a mixed integer programming model to identify the minimal subset of barriers that needs to be resolved, i.e., prescribed interventions, in order to lift a patient’s discharge likelihood above a given threshold. We show the effectiveness of the proposed approach in a series of retrospective analyses conducted at the Massachusetts General Hospital.
In the second part, I will be speaking about some of my ongoing research projects and potential future research topics motivated by the prescriptive approach described in the first part. Specifically, I will briefly introduce two potential applications of the proposed approach to
(i) a natural language processing (NLP) task of highlighting context-relevant text from an unstructured clinical note, and (ii) a clinical decision-making problem related to the optimal (partial) removal of tumorous tissue in the patients with low-grade glioma.
Bio: Taghi Khaniyev is an assistant professor at Bilkent University, Department of Industrial Engineering since September 2021. 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 focuses on developing and implementing data-driven analytical tools with the interplay of machine learning and optimization that can predict the discharge likelihood of patients and prescribe personalized interventions for the timely discharge of patients. Prior to joining MIT/MGH, he acquired his PhD in Management Science at the University of Waterloo. 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 and European Journal of Operations Research, 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).