Kaveri a. Thakoor

Principal Investigator

Dr. Kaveri Thakoor's Artificial Intelligence for Vision Science (AI4VS) laboratory is focused on transforming AI/deep learning systems into teammates for ophthalmologists by tackling key challenges currently inhibiting the translation of AI to the clinic, such as robustness, interpretability, and portability.  Dr. Thakoor earned her Ph.D. in Biomedical Engineering from Columbia University in the City of New York as a National Science Foundation Graduate Research Fellowship recipient.  Prior to that, she earned her B.S. with Honors in Chemistry from Stanford University and her M.S. in Computer Science from the University of Southern California.  Dr. Thakoor worked for two years as a research staff member on the Earthquake Early Warning algorithm development team at the California Institute of Technology Seismological Laboratory before joining Columbia.  She was awarded the 2022 Morton B. Friedman Memorial Prize for Doctoral Excellence by Columbia Engineering, and she received the 2022 Young Scientist Award for Graduate Students/Postdocs at the Northeast Bioengineering Conference.

Graduate Students

Ye Tian,  Ph.D. in BME

Ye is a first-year Ph.D. student in Biomedical Engineering. He worked on deep learning in glaucoma progression detection using OCT images and visual fields.

Ye is working on deploying diffusion models in OCT super-resolution for portable applications.

Michael Lau,  Ph.D. in CS

Michael is a recent MS Electrical Engineering graduate from Columbia University with an undergraduate degree in Computer Engineering from the University of Illinois at Urbana-Champaign. 

His interests lies in developing robust algorithms for healthcare applications.

Roshan Kenia,  M.S. in CS

Roshan is a first year Master’s student studying Computer Science whose focus is at the intersection of machine learning and biomedical imaging. He is currently working on using Vision Transformer (ViT) models to capture the eye-tracking of clinicians as they view OCT reports. He obtained his undergraduate degree in Computer Science from Stony Brook University. 

Roshan is interning at MIT Lincoln Laboratory.

Tri (Tom) Le,  M.S. in CS

Tom is a first year Master's student in Computer Science. He has a keen interest in applying Machine Learning to the healthcare sector, with a particular focus on biomedical imaging. 

He is currently researching and developing a system designed to assist novice ophthalmologists in analyzing Optical Coherence Tomography images. This system leverages the eye gaze patterns of seasoned clinicians, utilizing Mask Autoencoder and attention-based LSTM. Tom earned his Bachelor's degree in Computer Science from the University of Melbourne.

Mingyang Zang, Ph.D. in BME

Mingyang is a Ph.D. student in Dr. Andrew Laine's Lab, collaborating with AI4VS lab.

His research interest is computer aided image analysis including classification and segmentation. He is involved in projects on OCT report detection, Vision Field (VF) Mapping prediction, and Posterior Vitreous Opacities (PVOs) segmentation with Dr. Thakoor.


Undergraduate Students

Saanvi Aima, B.S. in CS & Stats

Saanvi is a CS-Statistics undergraduate student interested in the application of artificial intelligence in healthcare. She is working on the development of a multimodal text-image-gaze model. 

This would serve as a foundational tool to help radiologists identify regions of interest for glaucoma diagnosis using OCT reports. It would also help create a more interpretable and trustworthy automated diagnostic system.

Allison Cui, B.S. in CS

Allison is a CS undergraduate student interested in machine learning applications in the field of ophthalmology and medicine. 

She is learning image classification techniques in the context of eye-movement data to use deep learning methods to detect glaucoma progression and predict future visual fields. 

Angel (Leyi) Cui, B.A. in CS

Angel is a CS undergraduate student interested in applying machine learning to the field of healthcare and designing and developing useful and usable software to aid clinicians and improve healthcare software.

She is working on training experts’ eye fixation data to understand glaucoma results and developing helpful tools for clinicians and medical students to facilitate learning and diagnostic processes.

Sanmati Choudhary, B.S. in CS

Sanmati is a CS undergraduate working on using unsupervised machine learning/deep learning techniques (such as categorical embeddings) to extract meaningful clusters from expert clinician eye movement data.  By understanding the decision-making process of experts through their eye movements, her work has the potential to aid in medical education and in developing automated disease diagnosis systems.


OMAR MOUSSA, Ophthalmology Resident

Omar Moussa is an Ophthalmology Resident at Columbia University Irving Medical Center interested in use of AI for detection of Age-Related Macular Degeneration.  

Earlier, he contributed to the group's work on developing a multimodal deep learning system for detection of late stages of AMD and for finding common AMD ocular biomarkers for both human experts and AI.

Sophie Gu, Ophthalmology Resident

Sophie Gu grew up in the suburbs of Massachusetts. She studied biology and economics at Columbia University. She obtained her medical degree from Johns Hopkins Medical School. She is currently a PGY-3 resident at the Harkness Eye Institute at NYP-Columbia. 

In her free time, she enjoys traveling, eating, painting, and hanging out with her awesome coresidents.


Associate Professor of Electrical Engineering


Percy K. and Vida L.W. Hudson Professor of Biomedical Engineering and Professor of Radiology (Physics)

Lab: Heffner Biomedical Imaging Laboratory


Associate Professor of Ophthalmology

Website: Royce W.S. Chen, MD

Steve feiner

Professor of Computer Science

Website: Computer Graphics and User Interfaces Laboratory

Lab Alumni


Yifan Sun (09/2022 - 04/2023), M.S. in Financial Engineering, worked on incorporating expert clinician eye movement data (e.g. fixation location/duration) into training of Vision Transformer.

Anurag Sharma (09/2022 - 06/2023), M.S. in Biomedical Engineering, worked on developing approaches for extracting information from expert eye movement sequence data to train AI/deep learning systems.

Michelle Akerman (09/2022 - 06/2023), M.S. in Biomedical Engineering, worked on using unsupervised machine learning techniques to detect key features, such as fixations and saccades, in expert clinician eye movements.

Yifan Sun (09/2022 - 04/2023), M.S. in Financial Engineering, worked on incorporating expert clinician eye movement data (e.g. fixation location/duration) into training of Vision Transformer.

Shubham Kaushal (09/2022 - 12/2023), M.S. in Data Science, worked on incorporating expert clinician eye movement data (e.g. fixation order) into training of Vision Transformer models.

Alice Wang (05/2023 - 08/2023), B.S. in Computer Science, worked on incorporating expert clinician eye tracking data into Self-Supervised machine learning models and Vision Transformer models.


Joel Salas (Summer 2022), now at Rochester Institute of technology, worked on developing a PsychoPy experiment to assess the impact of incorporating predictive AI into the clinical workflow by measuring the impact on clinician's speed, confidence, and accuracy of glaucoma diagnosis.

Geoffrey Wu ( ? - September 2022), B.S. in CS, worked on developing multimodal deep learning algorithms using a combination of retinal fundus images, vessel segmentation images, and topological analysis of retinal vasculature to enable early detection of preeclampsia from vascular signatures of the disease found in the eye.