Pranav Kulkarni

prof_pic.jpg

6116 Executive Blvd, Suite 200
North Bethesda, MD 20852
pranavk [at] umd [dot] edu

I am a PhD student at the University of Maryland, College Park, advised by Heng Huang. I am also a Graduate Research Assistant at the University of Maryland University of Maryland Institute for Health Computing (UM-IHC), where I bridge expertise in machine learning and computer vision with imaging informatics to transform everyday clinical practice and improve patient outcomes.

My research is focused on how AI can reason over multiphase and multimodal clinical evidence, and do so reliably enough to support real-world clinical decision-making. Currently, I am working on multimodal foundation models and medical image synthesis with CT imaging.

In my free time, I enjoy hiking, gardening, and reading about history, biogeography, and urban planning!

Recent News

Oct 2026 Co-organized the AMPLIFAI Challenge at MICCAI’26 and delivered an invited keynote, “Multiphase Reasoning is a Foundation for Building Trustworthy Clinical AI.”
Jun 2026 Awarded an NVIDIA Academic Grant under the Data Science CFP for project titled, “Curie8: Accelerating Data Curation for Next-Generation Medical Image Analysis.”
Jun 2026 Interviewed for a student spotlight at UM-IHC.
Feb 2026 One paper accepted at CVPR’26 as highlight.
Jan 2026 One paper accepted at IEEE ISBI’26.
Sep 2025 Elected as the graduate student representative for the CS Education Committee.
Jul 2025 Started my PhD at University of Maryland under Heng Huang.
Jul 2025 One paper accepted at ICCV CVAMD’25.
May 2025 A review article that I co-authored has been published in Radiology.
Apr 2025 Received UM-IHC Travel Award to support my travel for MIDL’25.

Selected Publications

  1. AMPLIFAI: A Multiphase CT Dataset for Benchmarking Clinical Reasoning in LI-RADS Assessment of Liver Lesions
    Pranav Kulkarni, Nikhil Shah, Amritansh Suryavanshi, Jana Delfino, James Tonascia, Jade Wong-You-Cheong, Barton Lane, Joseph Chirico, Jeffrey D. Hirsch, Ang Li, Heng Huang, and Florence X. Doo
    arXiv preprint arXiv:2608.14778. Aug 2026
  2. CheXMark: Robust Dataset Ownership Verification for Chest Radiographs
    Pranav Kulkarni, Junfeng Guo, Florence X. Doo, and Heng Huang
    arXiv preprint arXiv:2602.09284. Jul 2026
  3. Multi-Crit: Benchmarking Multimodal Judges on Pluralistic Criteria-Following
    Tianyi Xiong, Yi Ge, Ming Li, Zuolong Zhang, Pranav Kulkarni, Kaishen Wang, Qi He, Zeying Zhu, Chenxi Liu, Ruibo Chen, Tong Zheng, Yanshuo Chen, Xiyao Wang, Renrui Zhang, Wenhu Chen, and Heng Huang
    IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Jun 2026