Kaveri Thakoor, PhD

  • Assistant Professor of Ophthalmic Science (in Ophthalmology)
Profile Headshot

Profile Navigation

Overview

Kaveri Thakoor, Ph.D., is an Assistant Professor of Ophthalmic Science (in Ophthalmology) in the Department of Ophthalmology at the Columbia University Irving Medical Center.  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. 

Academic Appointments

  • Assistant Professor of Ophthalmic Science (in Ophthalmology)

Gender

  • Female

Credentials & Experience

Education & Training

  • BS, 2010 Stanford University, Stanford, CA
  • MS, 2015 Computer Science, University of Southern California, Los Angeles, CA
  • PhD, 2022 Biomedical Engineering, Columbia University, New York, NY

Honors & Awards

  • 2022: Biomedical Engineering Young Scientist Award for Graduate Students/Postdocs at NEBEC 2022
  • 2022: Morton B. Friedman Memorial Prize for Excellence
  • 2022: Three Minute Thesis Finalist
  • 2021: Selected to Submit Proposal for NIH Director’s Early Independence Award 
  • 2021: ISBI Three Minute Thesis Finalist
  • 2021: Top Candidate to Compete for Regeneron Prize for Creative Innovation
  • 2018: National Science Foundation Graduate Research Fellowship 
  • 2016: Felicitation from Mayor of Covina
  • 2015: 2nd Place Best Student Paper Award
  • 2013: Best Student Paper Award
  • 2006: Caltech Employees Federal Credit Union Scholarship
  • 2006: National Merit Commended Scholar
  • 2005: Senior Girl Scout Gold Award Recipient
  • 2004: Command Performance: Superior for Clarinet Performance 

Research

Dr. Kaveri Thakoor's Artificial Intelligence for Vision Science (AI4VS) Laboratory is focused on transforming Artificial Intelligence (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.  Towards robustness, Dr. Thakoor is focused on developing AI systems that can generalize to data collected at multiple locations, from multiple imaging modalities, or even from data of varying quality.  She has developed algorithms to detect glaucoma robustly from data collected at multiple locations and algorithms to detect Age-Related Macular Degeneration (AMD) from multiple imaging modalities (Optical Coherence Tomography, OCT, and OCT Angiography).  Towards portability, her research has resulted in proof-of-principle, generative adversarial network-based image processing techniques to super-resolve data collected from a low-cost, portable OCT device in order to enhance downstream AI performance even when using AI systems trained on high-quality, commercial data.  Towards interpretability, she has used expert eye movements to corroborate the mechanisms behind AI systems by comparing regions of interest used by AI to detect glaucoma from OCT images with OCT-image regions fixated most by medical experts during glaucoma diagnosis.  Her future interests lie in developing “eye tracking-informed” AI systems such that regions of interest ‘learned’ by the AI are constrained by expert eye movements, making such systems inherently more accurate, interpretable, and computationally efficient.  Furthermore, by embedding robust and interpretable AI systems into portable imaging devices, she seeks to make such technology more accessible to diverse populations.

Selected Publications

Journal/Magazine/Book Articles

  1. Thakoor, K.A., Yao, J., Bordbar, D., Moussa, O., Lin, W., Sajda, P., Chen, R. “A Multimodal Deep Learning System to Distinguish Late Stages of AMD and to Compare Expert vs. AI Ocular Biomarkers.” Scientific Reports, 12(1), p.1?11, 2022.
  2. Thakoor, K.A., Li, X., Tsamis, E., Zemborain, Z.Z., De Moraes, C.G., Sajda, P., Hood, D.C. “Strategies to improve convolutional neural network generalizability and reference standards for glaucoma detection from OCT scans.” Translational Vision Science and Technology, 10(4), pp.16?16, 2021.
  3. Koorathota, S., Thakoor, K., Hong, L., Mao, Y., Adelman, P., Sajda, P. A Recurrent Neural Network for Attenuating Non?cognitive Components of Pupil Dynamics. Frontiers in Psychology, 12, p.12, 2021.
  4. Thakoor, K., Koorathota, S., Hood, D., Sajda, P. “Robust and Interpretable Convolutional Neural Networks to Detect Glaucoma in Optical Coherence Tomography Images.” IEEE Transactions on Biomedical Engineering, 68(8), pp. 2456?2466, August 2021. Early Access: https:// ieeexplore.ieee.org/document/9286420, 8 December 2020.
  5. Tsamis, E., Bommakanti, N., Sun, A., Thakoor, K., De Moraes, C.G., and Hood, D.C. “An Automated Method for Assessing Topographical Structure?Function Agreement in Abnormal Glaucomatous Regions.” Translational Vision Science and Technology, 9(4), 2020 .
  6. Thakoor, K., Andrews, J., Hauksson, E. and Heaton, T. “From Earthquake Source Parameters to Ground Motion Warnings near You: The ShakeAlert Earthquake Information to Ground?Motion (eqInfo2GM) Method.” Seismological Research Letters, 90(3), pp.1243?1257, 2019 .
  7. Thakoor, K.A., “Cognitive Mechanisms Underlying the Classification of Reduced Dimensionality, Information Rich Image Representations.” IEEE Intelligent Systems, 31(2), pp.9?20, 2016 .
  8. Thakoor, K., Mante, N., Zhang, C. et al. “A System for Assisting the Visually Impaired in Localization and Grasp of Desired Objects,” Springer International Publishing, Switzerland; L. Agapito, et al. (Eds): Lecture Notes in Computer Science (LNCS), ECCV 2014 Workshops, Part III, Volume 8927, pp. 643?657, 2015.
  9. Thakoor, K.A. “Tinkering with Quantum Dots – Strange Electron Behavior in a Semiconductor Nanostructure: The Kondo Effect.” Stanford Scientific Magazine, Vol. 6, Issue 1, Autumn 2007.

Conference Papers

  1. Thakoor, K., Koorathota, S., Hood, D., Sajda, P. “Robust and Interpretable Convolutional Neural Networks to Detect Glaucoma in Optical Coherence Tomography Images.” IEEE International Conference on Image Processing (ICIP), 2021.
  2. Thakoor, K., Bordbar, D., Yao, J., Moussa, O., Chen, R., Sajda, P. “Hybrid 3D?2D Deep Learning for Detection of Neovascular Age?Related Mac? ular Degeneration Using Optical Coherence Tomography B?Scans and Angiography Volumes.” International Symposium on Biomedical Imaging (ISBI), 2021.
  3. Koorathota, S., Thakoor, K., Adelman, P., Mao, Y., Liu, X., Sajda, P. “Sequence Models in Eye Tracking: Predicting Pupil Diameter During Learning.” In ACM Symposium on Eye Tracking Research and Applications, pp. 1?3, 2020.
  4. Thakoor, K.A., Li, X., Tsamis, E., Sajda, P. and Hood, D.C. “Enhancing the Accuracy of Glaucoma Detection from OCT Probability Maps using Convolutional Neural Networks”. In 41st Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), pp. 2036?2040, 2019.
  5. Drori, I., Dwivedi, I., Shrestha, P., Wan, J., Wang, Y., He, Y., Mazza, A., Krogh?Freeman, H., Leggas, D., Sandridge, K., Nan, L., Thakoor, K., Josh, C., Goenka, S., Chen, K., Pe’er, I. “High quality prediction of protein Q8 secondary structure by diverse neural network architectures.” Neural Information Processing 2018 Workshop on Machine Learning for Molecules and Materials, 8 December 2018.
  6. Thakoor, K.A., “Classification of Biological Images and Natural Scenes via Reduced Dimensionality, Information Rich Representations.” In Proceedings of the 28th International Conference on Computer Applications in Industry and Engineering, pp. 12?14, San Diego, CA, USA, 2015.
  7. Thakoor, K.A., “Neural Basis for Enhanced Classification Accuracy of Reduced Dimensionality, Information Rich Representations of Images.” In Proceedings of the 37th International Conference of the IEEE Engineering in Medicine and Biology Society, Milan, Italy, Aug. 25?29, 2015.
  8. Thakoor, K.A., “Reduced Dimensionality, Information Rich Visual Representations for Scene Classification.” In 2015 IEEE Signal Processing and Signal Processing Education Workshop (SP/SPE), pp. 43?48, 2015, 2nd Place Best Student Paper Award.
  9. Thakoor, K.A., Marat, S., Nasiatka, P.J., et al. “Attention Biased Speeded Up Robust Features (AB?SURF): A Neurally?Inspired Object Recogni? tion Algorithm for A Wearable Aid for the Visually?Impaired.” In 2013 IEEE International Conference on Multimedia and Expo Workshops (ICMEW), pp. 1?6, 2013, Best Student Paper Award.

Peer-Rreviewed Abstracts

  1. Thakoor, K., Leshno, A., La Bruna, S., Tsamis, E., De Moraes, C.G., Sajda, P., Harizman, N., Liebmann, J., Cioffi, G.A., Hood, D.C. “Evaluation of a Deep Learning Model on a Real?World Clinical Glaucoma Dataset.” ARVO Annual Meeting, 2022.
  2. Thakoor, K., Bordbar, D., Yao, J., Moussa, O., Lin, W., Scherbakova, I., Diaconita, V., Sajda, P., Chen, R. “A Hybrid Deep Learning System to Distinguish Late Stages of AMD and to Compare Expert vs. Machine AMD Risk Features.” Investigative Ophthalmology and Visual Science, 62(8), pp.2146?2146, 2021.
  3. Li, X., Tsamis, E., Thakoor, K.A., Zemborain, Z., De Moraes, C.G., Hood, D.C. “Evaluating the Transferability of Deep Learning Models that Distinguish Glaucomatous from Healthy OCT Circumpapillary Disc Scans.” Investigative Ophthalmology and Visual Science, 61(7), pp.4548? 4548, 2020.
  4. Thakoor, K., Tsamis, E., De Moraes, C.G., Sajda, P., Hood, D.C. “Impact of Reference Standard, Data Augmentation, and OCT Input on Glau? coma Detection Accuracy by CNNs on a New Test Set.” Investigative Ophthalmology and Visual Science, 61(7), pp.4540?4540, 2020.
  5. Tsamis, E., Bommakanti, N., Sun, A., Thakoor, K., De Moraes, C.G., and Hood, D.C. “An automated method for assessing topographical structure?function agreement in abnormal regions in glaucoma.” Investigative Ophthalmology and Visual Science, 60(9), pp.6141?6141, 2019.
  6. Thakoor, K.A., Zheng, Q., Nan, L., Li, X., Tsamis, E.M., Rajshekhar, R., Dwivedi, I., Drori, I., Sajda, P. and Hood, D.C., “Assessing the Ability of Convolutional Neural Networks to Detect Glaucoma from OCT Probability Maps.” Investigative Ophthalmology and Visual Science, 60(9), pp.1464?1464, 2019.
  7. Adebiyi A., et al. (4th co?author of 4). “Feedback measures for a wearable visual aid designed for the visually impaired..” Investigative Ophthalmology and Visual Science, 54 (15), pp. 2764?276, 2013.
  8. LeMieux, M., Barman, S., Mark, R., Opatkiewicz, J., Thakoor, K.A., Bao, Z., “Nanotube Network Transistors on Functional Surfaces.” Materials Research Society Symposium P: Carbon Nanotubes and Related Low?Dimensional Materials., 60(9), 24?28 March, 2008.
  9. Rajan, S.K., Thakoor, K.A., Yang, T.J., Gao, F., “VEGF Levels Aid Prediction of Chemotherapy Induced Myeloid Toxicity and Presence of Extranodal Disease in Diffuse Large B?Cell Lymphoma (DLBCL).” Blood, 106: 11 (2), A4725, November, 2005.

For a complete list of publications, please visit Google Scholar