Toronto scientist Rahul Krishnan gets big award to study artificial intelligence in health care.

Amazon just awarded a Toronto-based researcher $85,000 to study the consequences of implementing artificial intelligence in health care.

“I think there needs to be a conversation about how to mitigate the potential negative harms that [these tools] may come with,”

Rahul Krishnan, an assistant professor exploring computational medicine at the University of Toronto, told CTV News Toronto.

Amazon Research Awards were handed to 79 academic researchers last week to study topics from sustainability to automated reasoning.

The 33-year-old Toronto recipient began taking an interest in computational medicine years ago when he was studying electrical and computer engineering at the University of Toronto, beginning in 2008.

“This is well before machine learning was in the public eye in any way,” Rahul Krishnan said. “By the time I completed my PhD (at the Massachusetts Institute of Technology), machine learning exploded.”

Rahul Krishnan award comes at a time when artificial intelligence has captured the public’s attention, with programs such as ChatGPT and MidJourney made accessible to anyone with an internet connection.

Just last week, Microsoft announced plans to integrate GPT-3 language models into Epic, an electronic health record (EHR) software used in North American hospitals.

“There are obviously consequences with this decision and so what we would like to do is study some of these consequences,”

Rahul Krishnan said.

“I think it’s a really important question to study because medicine is a field that is not static, what constitutes the standard of care is something that is dynamic and changes over time.”

‘PLAYING CATCH-UP’ – Toronto scientist Rahul Krishnan gets big award to study artificial intelligence in health care

As popularity spikes, questions on the future of artificial intelligence have surfaced, and many are asking how we will regulate the tools.

“I think our regulation, particularly in the context of health care, has been playing catch-up over the years,”

Rahul Krishnan said.

“There is really a need for regulation to catch-up so we’re not caught flat footed if these tools do have unintended consequences when deployed in health care.”

One such danger could include “internal biases, baked into the model,” that have the potential to harm patients by perpetuating prejudiced decision-making.

Another research topic he’s pursuing investigates how these biases manifest in machine learning models, in an effort to create tools to help doctors make decisions faster or in a more informed manner.

As an example, he pointed to disparities that exist in the kidney transplant waitlist. Black people are four times more likely to develop kidney failure than white people in the United States, yet they are less likely to receive a lifesaving transplant, according to 2020 findings in the National Library of Medicine (NLM).

“Our findings support previous work examining the effects of discrimination and medical mistrust on referral,”

the research states.

One of the challenges in Canada is a lack of sufficient data available on subgroup identities, Rahul Krishnan said.

“I think it’s naive to say there is no disparity. I believe those disparities do exist.”

His goal with this research is to identify disparities existing in medical data, work to improve outcomes for those patients, and then assess how to apply the methods they adapt to other discriminated groups.

Across the board, Rahul Krishnan said he aims to understand the strength of tools in artificial intelligence, alongside their limitations.

“I think that there is both an enormous amount of opportunity here and an enormous amount of care that needs to be given.”

Toronto scientist Rahul Krishnan gets big award to study artificial intelligence in health care

Rahul G. Krishnan

Assistant Professor
Dept. of Computer Science
Dept. of Laboratory Medicine and Pathobiology
The University of Toronto
CIFAR AI Chair at Vector Institute

Google ScholarBio

Bio

Rahul G. Krishnan is an Assistant Professor of Computer Science and Medicine (Laboratory Medicine and Pathobiology). He is a CIFAR AI Chair at the Vector Institute and a member of the Temerty Center for Artificial Intelligence in Medicine. He works on leveraging tools from developing algorithms for probabilistic inference, and applied machine learning to problems in healthcare such as modeling disease progression and risk stratification. Previously, he was a Senior Researcher at Microsoft Research New England. He received his MS from New York University and his PhD in Electrical Engineering and Computer Science from MIT in 2020.

Research – Toronto scientist Rahul Krishnan gets big award to study artificial intelligence in health care

My goal is to develop machine learning algorithms to create a learning healthcare system, where digitized clinical and biological data are used to improve clinical care while improving our understanding of human & disease biology.

My research interests lie in the following topics:

  • Deep learning: Unsupervised and self-supervised learning algorithms for extracting predictive patterns from noisy, high-dimensional data.
  • Causal inference: Developing methods for estimating causal effects to identify good interventional policies from high-dimensional, time-varying observational data.
  • Reliable machine learning: Developing guardrails for the reliable deployment of machine learning models.

Updates– Toronto scientist Rahul Krishnan gets big award to study artificial intelligence in health care

  • Please see the Joining page for information on how to join the group.
  • Please see the Teaching page for the most recent updates if you’re enrolled in one my courses in Fall 2022.

Research

Publications

A Learning Based Hypothesis Test for Harmful Covariate Shift
T. Ginsberg, Z. Liang, R. Krishnan
(Code)
International Conference on Learning Representations (ICLR) 2023

Neural Differential Equations with Orthogonal Polynomial Projections
E. de-Brouwer, R. Krishnan
International Conference on Learning Representations (ICLR) 2023

Partial Identification with Implicit Generative Models
V. Belazadeh-Meresht, V. Syrgkanis, R. Krishnan
(Code)
Neural Information Processing Systems (NeurIPS) 2022

HiCu: Leveraging Hierarchy for Curriculum Learning in Automated ICD Coding
W. Ren, R. Zeng, T. Wu, T. Zhu, R. Krishnan
(Code)
Machine Learning for Healthcare (MLHC) 2022

Large Images as Long Documents: Hierarchical ViTs with Self-Supervised Pretraining in Gigapixel Image Pyramids
R. Chen, C. Chen, Y. Li, T. Chen, A. Trister, R. Krishnan*, F. Mahmood*
(Code)
Computer Vision and Pattern Recognition (CVPR), 2022
(*: equal contribution)

Oral Presentation

Hierarchical Optimal Transport for Comparing Histopathology Datasets
A. Yeaton, R. Krishnan, R. Mieloszyk, D. Alvarez-Melis, G. Huynh
(Code)
Medical Imaging with Deep Learning (MIDL), 2022

Using Time-Series Privileged Information for Provably Efficient Learning of Prediction Models
R. Karlsson, M. Willbo, Z. Hussain, R. Krishnan, D. Sontag, F. Johansson
(Code)
Artificial Intelligence and Statistics (AISTATS), 2022

Clustering Interval-Censored Time-Series for Disease Phenotyping
I. Chen, R. Krishnan, D. Sontag
(Code)
Association for the Advancement of Artificial Intelligence (AAAI), 2022

Mitigating bias in estimating epidemic severity due toheterogeneity of epidemic on-set and data aggregation
R. Krishnan, S. Cenci, L. Bourouiba
In Press, Annals of Epidemiology, 2021

Neural Pharmacodynamic State Space Modeling
(Code) (Data)
Z. Hussain*, R. Krishnan*, D. Sontag
International Conference on Machine Learning (ICML), 2021
(*: equal contribution)

Max-Margin learning with the Bayes factor
R. Krishnan, A. Khandelwal, R. Ranganath, D. Sontag
Uncertainty in Artificial Intelligence (UAI), 2018

Representation Learning Approaches to Detect False Arrhythmia Alarms from ECG Dynamics
E. Lehman, R. Krishnan, X.Zhao, R. Mark, L. Lehman
Machine Learning for Healthcare (MLHC), 2018

Variational Autoencoders for Collaborative Filtering (Code)
D. Liang, R. Krishnan, M. Hoffman, T. Jebara
World Wide Web Conference (WWW), 2018

On the challenges of learning with inference networks on sparse, high-dimensional data
(Code)
R. Krishnan, D. Liang, M. Hoffman
Artificial Intelligence and Statistics (AISTATS), 2018

Structured Inference Networks for Nonlinear State Space Models
(Code)
R. Krishnan, U. Shalit, D. Sontag
Association for the Advancement of Artificial Intelligence (AAAI), 2017

Oral Presentation

Barrier Frank-Wolfe for Marginal Inference
(Code)
R. Krishnan, S. Lacoste-Julien, D. Sontag
Neural Information Processing Systems (NeurIPS), 2015

Preprints– Toronto scientist Rahul Krishnan gets big award to study artificial intelligence in health care

Deep Kalman Filters
(Code)
R. Krishnan, U. Shalit, D. Sontag
Presented at Advances in Approximate Bayesian Inference & Black Box Inference (AABI) Workshop, NeurIPS, 2015

Peer-reviewed workshop papers

Learning predictive checklists from continuous medical data
Y. Makhija, E. de Brouwer, R. Krishnan
Machine Learning for Healthcare Workshop (ML4H), NeurIPS 2022

Self-Supervised Vision Transformers Learn Disentangled Representations in Histopathology
R. Chen, R. Krishnan
(Code)
Learning Meaningful Representations of Life Workshop (LMRL), NeurIPS 2021

Mixture-of-experts VAEs can disregard unimodal variation in surjective multimodal data paper
J. Wolff, T. Klein, M. Nabi, R. Krishnan, S. Nakajima
Bayesian Deep Learning Workshop (BDL), NeurIPS 2021

Inference and Introspection in Deep Generative Models of Sparse Non-Negative Data
R. Krishnan, M. Hoffman
Advances in Approximate Bayesian Inference & Black Box Inference (AABI) Workshop, NeurIPS, 2016

Disney Research Award

Early Detection of Diabetes from Health Claims
R. Krishnan, N. Razavian, Y. Choi, S. Nigam, S. Blecker, A. Schmidt, D. Sontag
Machine Learning in Healthcare Workshop, NeurIPS, 2013

Teaching

Fall 2022: CSC2541H – Topics in Machine Learning: Introduction to Causality

Fall 2022: CSC311H1- Introduction to Machine Learning

Fall 2021: CSC2541H – Topics in Machine Learning: Machine Learning for Health

Fall 2021: CSC311H1 – Introduction to Machine Learning

Graduate Student Researchers

Vahid Belazadeh Meresht
PhD Student, CS

Michael Cooper
PhD Student, CS, Co-supervised with Michael Brudno

Jerry Ji
PhD Student, CS, Co-supervised with Anna Goldenberg

Ian Shi
PhD Student, CS, Co-supervised with Quaid Morris

Tom Ginsberg
MS Student, CS

Asic Chen
MS Student, CS

Aslesha Pokhrel
MScAC Student, CS

Vahid Zehtab
MScAC Student, CS

Keerat Guilani
MScAC Student, CS

Undergraduate Student Researchers

Alumni

Edward de Brouwer
Visiting PhD student from KU Leuven, Belgium

Jacob Si
BSc in CS ->
MSc at UCLA

Taewoo Kim
MScAC in CS ->
Layer6