Sana
Tonekaboni

Ph.D. candidate at the University of Toronto

About Me

I'm a PhD Student in the Machine Learning Group at the University of Toronto and the Vector Institute. My supervisor is Anna Goldenberg and my research is in machine learning in healthcare. I'm interested in building solutions for barriers of adoption of AI in clinical settings, and my research spans from explainable AI to unsupervised representation learning for medical time series. I work in close collaboration with clinicians at the Hospital for Sick Children and the Laussen Labs research group and I am currently an Apple Scholars in AI/ML .

Download my CV

Announcements

  • Jan. 2022 — AISTATS Acceptance

    Our paper Decoupling Local and Global Representations of Time Series, done as part of my internship at Google was accepted to AISTATS 2022!

  • April. 2021 — Apple scholars in AI/ML

    I received the Apple Scholars in AI/ML PhD fellowship

  • Jan. 2021 — ICLR Acceptance

    Our paper Unsupervised Representation Learning for Time Series with Temporal Neighborhood Coding was accepted to ICLR 2021!

  • Sep. 2020 — NeurIPS Acceptance

    Our paper What went wrong and when? Instance-wise feature importance for time-series models was accepted to Neurips 2020!

  • Sep. 2019 — CIHR fellowship

    I received the health system impact fellowship from the Canadian Institute of Health Research (CIHR)

  • Aug. 2019 — QEII-GSST Award

    I received the Queen Elizabeth II Graduate Scholarship in Science & Technology

  • Nov. 2019 — Pan-Canadian Self-Organizing Conference on Machine Learning (poster chair)

    The inaugural Pan-Canadian Self-Organizing Conference on Machine Learning (PC-SOCMLx) is an event for Canadian graduate students in machine learning to meet each other and develop a research community (Supported by: Vector Institute, Mila, Amii, CIFAR, and Facebook)

  • Oct. 2019 — NeurIPS workshop acceptance

    Our paper "Individualized Feature Importance for Time Series Risk Prediction Models" was accepted to the Machine Learning for Health Workshop at NeurIPS, 2019

  • May. 2019 — MLHC acceptance

    Our paper What Clinicians Want: Contextualizing Explainable Machine Learning for Clinical End Use was accepted to MLHC 2019!

  • Nov. 2018 — Guest lecturer

    I gave a talk at Toronto rehabilitation institute research round, with the title: "Prediction models for longitudinal data"

    • Publications

    Conference and Journal

    • Decoupling Local and Global Representations of Time Series

      Sana Tonekaboni, Chun-Liang Li, Sercan O Arik, Anna Goldenberg, Tomas Pfister
      International Conference on Artificial Intelligence and Statistics (AISTATS), 2022.
      [paper][code]

    • How to validate Machine Learning Models Prior to Deployment: Silent trial protocol for evaluation of real-time models at the ICU

      Sana Tonekaboni, Gabriela Morgenshtern, Azadeh Assadi, Aslesha Pokhrel, Xi Huang, Anand Jayarajan, Robert Greer, Gennady Pekhimenko, Melissa McCradden, Fanny Chevalier, Mjaye Mazwi, Anna Goldenberg
      Conference on Health, Inference, and learning (CHIL), 2022.

    • Learning Unsupervised Representations for ICU Timeseries

      Addison Weatherhead, Robert Greer, Michael-Alice Moga, Mjaye Mazwi, Danny Eytan, Anna Goldenberg, Sana Tonekaboni
      Conference on Health, Inference, and learning (CHIL), 2022.

    • Unsupervised Representation Learning for TimeSeries with Temporal Neighborhood Coding

      Sana Tonekaboni, Danny Eytan, Anna Goldenberg
      International Conference on Learning Representations (ICLR), 2020.
      [paper][code]

    • What went wrong and when? Instance-wise feature importance for time-series models

      Sana Tonekaboni*, Shalmali Joshi*, Kieran Campbell, David Duvenaud, Anna Goldenberg
      Neural Information Processing Systems (NeurIPS), 2020.
      [paper] [code][poster]

    • What Clinicians Want: Contextualizing Explainable Machine Learning for Clinical End Use

      Sana Tonekaboni*, Shalmali Joshi*, Melissa McCradden, Anna Goldenberg
      Conference on Machine Learning for Healthcare (MLHC), 2019.
      [paper] [poster]

    • Prediction of Cardiac Arrest from Physiological Signals in the PediatricICU

      Sana Tonekaboni, Mjaye Mazwi, Peter Laussen, Danny Eytan, Robert Greer, Sebastian D. Goodfellow, Andrew Goodwin, Michael Brudno, Anna Goldenberg
      Conference on Machine Learning for Healthcare (MLHC), 2018.
      [paper]

    • Closed-Loop Neurostimulators: A Survey and A Seizure-Predicting Design Example for Intractable Epilepsy Treatment

      Hossein Kassiri, Sana Tonekaboni, M. Tariqus Salam, Nima Soltani, Karim Abdelhalim, Jose Luis Perez Velazquez, Roman Genov
      IEEE transactions on biomedical circuits and systems 11, 2017.
      [paper]

    Peer reviewed workshops

    • Individualized Feature Importance for Time Series Risk Prediction Models

      Sana Tonekaboni, Shalmali Joshi, Anna Goldenberg
      In Machine Learning for Health Workshop at NeurIPS, 2019
      [poster]

    • Five Pillars of Explainable Clinical Machine Learning

      Melissa McCradden, Sana Tonekaboni, Shalmali Joshi, Anna Goldenberg
      In Frontier of AI-Assisted Care (FAC) Scientific Symposium, 2019

    Awards

    Top