Kayhan Batmanghelich, PhD
- Assistant Professor, Department of Biomedical Informatics
Kayhan Batmanghelich is an Assistant Professor of the Department of Biomedical Informatics and Intelligent Systems Program with secondary appointments in the Computer Science Department at the University of Pittsburgh and an adjunct faculty in the Machine Learning Department at the Carnegie Mellon University. He received his Ph.D. from the University of Pennsylvania (UPenn) under the supervision of Prof. Ben Taskar and Prof. Christos Davatzikos. He spent three years as a postdoc in Computer Science and Artificial Intelligence Lab (CSAIL) at MIT, working with Prof. Polina Golland. His research is at the intersection of medical vision, machine learning, and bioinformatics. His group develops machine learning methods that address interesting challenges of AI in medicine, such as explainability, learning with limited and weak data, and integration of medical image data with other biomedical data modalities. His research is supported by awards NIH and NSF, as well as industry-sponsored projects.
- S. Singla, B. Pollack, J. Chen, K. Batmanghelich, Explanation by Progressive Exaggeration. Eighth International Conference on Learning Representations (ICLR), 2020.
- Y. Xu†, M. Gong†, C. Li, K. Zhang, K. Batmanghelich, Twin Auxiliary Classifiers GAN. Conference on Neural Information Processing Systems (NeurIPS), 2019.
- J. Schabdach, S. Wells, M. Cho, N. Batmanghelich, A Likelihood-Free Approach for Characterizing Heterogeneous Diseases in Large-Scale Studies. International Conference on Information Processing in Medical Imaging (IPMI), LNCS, pp 170-183, 2017.
- N. Batmanghelich, A. Dalca, G. Quon, M. Sabuncu, P. Golland, Probabilistic Modeling of Imaging, Genetics and the Diagnosis. IEEE Transactions on Medical Imaging (TMI), pp 1765-1779, 2016.
- N. Batmanghelich, B. Taskar, C. Davatzikos, Generative-Discriminative Basis Learning for Medical Imaging. IEEE Transactions on Medical Imaging (TMI), 31(1), pp 51-69, 2012.
I am interested in research direction for medical imaging in the following directions:
- Explainable and Interpretable AI
- Learning with limited data (weakly/semi-/un-supervised learning)
- Bayesian Probabilistic Reasoning and Causal Inference
- Learning from Multi-Modal data (Combination of imaging and genetic/genomics and text)
- Application domain: Alzheimer’s disease, Chronic Obstructive Pulmonary Disease (COPD), Non-Alcoholic Fatty Liver Disease (NAFLD)
Ongoing Research Support
SAP Batmanghelich (PI) 04/01/2018 – 04/30/2021
Deep Multi-Domain Learning: A Framework to Incorporate Weak Labels to the Attention Models
1R01HL141813-01 Batmanghelich (PI) 05/01/2018 – 03/31/2023
An Integrative Radiogenomic Approach to Design Genetically-Informed Image Biomarker for Characterizing COPD
NSF DMS-1839332 Batmanghelich (PI) 10/01/2018-09/30/2021
TRIPODS+X:RES: Collaborative Research: Learning with Expert-In-The-Loop for Multimodal Weakly Labeled Data and an Application to Massive Scale Medical Imaging
Completed Research Support
Pfizer Batmanghelich (PI) 10/31/2016 – 10/31/2017
Developing Statistical Method to Jointly Model Genotype and High Dimensional Imaging Endophenotype
1R21 MH109819-01A1 Del Re (PI) 09/26/2016 – 08/31/2018
Ventricles, Corpus Callosum, Symptoms & MIR137 in Large N Study of Schizophrenia
University of Pittsburgh (CMRF) Batmanghelich (PI) 07/01/2017 – 06/30/2018
Machine Learning Approach to Characterize COPD using Heritable Image Phenotype
T15 LM007059-31S1 Hochheiser (PI) 09/01/2017 – 06/30/2018
Pittsburgh Biomedical Informatics Training Program – Supplement