Kimaya Kulkarni

Kimaya Kulkarni

I’m a Research Data Scientist at UCLA Health, working in the Medical Imaging Informatics (MII) Group with Dr. William Hsu, where I build robust AI systems for medical imaging. My work spans deep learning, harmonization, sensor fusion, and fairness in healthcare AI.

During my MS in Electrical & Computer Engineering at UCLA, I conducted research on remote healthcare and computational imaging with Prof. Achuta Kadambi.

Previously, I interned at Rivian working on image signal processing and object detection for autonomous systems.

Email  /  CV  /  Google Scholar  /  LinkedIn  /  GitHub

Kimaya Kulkarni
Research

I’m interested in computer vision, multimodal learning, medical imaging, fairness, and robustness in AI. My projects often explore how to make machine learning models generalize well across real-world healthcare data.
CTNorm CT-Norm: A Toolkit to Characterize and Harmonize Variability in CT
SPIE Medical Imaging, 2025
code / paper

A modular medical imaging toolkit for assessing and harmonizing imaging variability across datasets to boost AI model reliability and generalization.
EquiPleth Equitable Heart Rate Estimation using Camera and Radar Fusion
SIGGRAPH & ICCP, 2022
project page

Developed a contactless heart rate monitoring system using camera-radar fusion with AI that improves fairness across skin tones.
Diverse RPPG Diverse RPPG: Fair Heart Rate Estimation across Skin Tones
arXiv
project page

Designed deep learning models robust to skin-tone bias using demographically diverse data, improving the fairness of rPPG estimation models.

Website adapted from Jon Barron’s academic site. Template inspiration from his GitHub repository.