Kayhan Batmanghelich

GitHub   LinkedIn   Twitter   GoogleScholar

Kayhan Batmanghelich

Assistant Professor
Department of Biomedical Informatics

kayhan@pitt.edu
Phone: 412-648-9037
My CV (please contact me)

Admin Support: Maria Bond (bond@pitt.edu)

Room 531
5607 Baum Blvd, Suite 500
Pittsburgh, PA 15206-3701

I am an Assistant Professor of the Department of Biomedical Informatics with a secondary appointment at the School of Computing and Information at the University of Pittsburgh. My research is at the intersection of medical vision (medical image analysis), machine learning, and bioinformatics. I develop algorithms to analyze and understand medical images, genetic data, and other electrical health records such as clinical reports. The main themes of research in my lab are about the main challenges of AI in healthcare: (1) Explainability, (2) Data Efficiency, (3) Multimodal Data Fusion and Causality. My lab works on Alzheimer’s Disease, Chronic Obstructive Pulmonary Disease (COPD), and Non-Alcoholic Fatty Liver Disease (NAFLD) projects. Our research is supported by funding from NIH, NSF, and industry awards.  Read More

Recent News

One paper is accepted to WACV!

Oct 11, 2022

Congrats to Sumedha and Nihal for their paper in WACV! They showed how our previous work on Counterfactual Explainer could be used to fix an overconfident BlackBox classifier!

Yingci’s manuscript is accepted to the oral oncology!

Sep 07, 2022

Happy for Yingci! Her manuscript has been accepted to Oral Oncology!

Counterfactual blackbox explanation paper is accepted to MedIA!

Sep 02, 2022

Congratulation to Sumedha and her team! Her manuscript on counterfactual model explanation paper is finally accepted for a special issue in MedIA about XAI! The very initial pre-print is here; the full version is coming out soon! I am thankful to Motahare for her amazing contribution to the paper.

An efficient 3D GANs is finally out!

Apr 26, 2022

Congratulations to Li and his team for their big paper in IEEE JBHI that makes volumetric GANs possible for high-resolution medical images. They made a huge effort, and I am proud of them!

Yanwu’s Paper about Adversarial Spatial Perturbation is accepted to CVPR!

Mar 02, 2022

Congratulations to Yanwu for his CVPR paper! His method uses the Maximal Spatial Perturbation idea that significantly enhances image-to-image translation!

Our paper about Knowledge Distillation is accepted to AAAI 21!

Dec 01, 2021

Congratulations to Ardavan, Li! Their paper is accepted to AAAI 21! The link and code will be posted soon.

Selected Publications

Probabilistic Modeling of Imaging, Genetics and the Diagnosis

K.N. Batmanghelich, A. Dalca, G. Quon, M. Sabuncu, P. Golland
IEEE Trans Med Imaging

Explanation by Progressive Exaggeration

S. Singla, B. Pollack, J. Chen, K. Batmanghelich

Eighth International Conference on Learning Representations (ICLR)

 

Subject2Vec: Generative-Discriminative Approach from a Set of Image Patches to a Vector

S. Singla, M. Gong, S. Ravanbakhsh, F. Sciurba, B. Poczos, K.  Batmanghelich
Medical Image Computing & Computer-Assisted Intervention (MICCAI)

Twin Auxiliary Classifiers GAN

M.*, Y. Xu*, Ch. Li, Kun Zhang, K. Batmanghelich (*: equal contribution)
NeurIPS 2019 [Spotlight 2.4%]

Lab Members

Tigmanshu Chaudhary

Master
Machine Learning Engineer

Junxiang Chen

PhD
Post Doctoral Researcher

Sumedha Singla

PhD Student (Pitt-CS)
Research Assistant

Ke Yu

PhD Student (Pitt-ISP)
Research Assistant

Li Sun

PhD Student (Pitt-ISP)
Research Assistant

Yanwu Xu

PhD Student (Pitt-ISP)
Research Assistant

Rick Chang

PhD Student (Pitt-Biostat)
Research Assistant

Shantanu Ghosh

PhD Student (Pitt-ISP)
Research Assistant

Nihal Murali

PhD Student (Pitt-ISP)
Research Assistant

Matthew Ragoza

PhD Student (Pitt-ISP)
Research Assistant

Maxwell Reynolds

PhD Student (Pitt-DBMI)
Research Assistant