Creating actionable insights in human health.

R1 R2 R3


What models are healthy?

We work on robust machine learning model that can efficiently and accurately model events from healthcare data, and investigate best practices for multi-source integration, and learning domain appropriate representations.

  • Ng, Nathan, Kyunghyun Cho, and Marzyeh Ghassemi. SSMBA: Self-Supervised Manifold Based Data Augmentation for Improving Out-of-Domain Robustness Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP). 2020.
  • Zhang, Haoran, Natalie Dullerud, Laleh Seyyed-Kalantari, Quaid Morris, Shalmali Joshi, and Marzyeh Ghassemi. An Empirical Framework for Domain Generalization in Clinical Settings. Proceedings of the Conference on Health, Inference, and Learning, pp. 279-290. 2021.
  • Roth, Karsten, Timo Milbich, Bjorn Ommer, Joseph Paul Cohen, and Marzyeh Ghassemi. Simultaneous Similarity-based Self-Distillation for Deep Metric Learning. International Conference on Machine Learning, pp. 9095-9106. PMLR, 2021.

What healthcare is healthy?

The labels we obtain from health research and health practices are all based on decisions made from humans, as part of a larger system. We work on auditing and improving model fairness, as well as understanding the trade-offs that other constructs such as privacy may dictate, are important parts of responsible machine learning in health.

  • Ghassemi, Marzyeh, Tristan Naumann, Peter Schulam, Andrew L. Beam, Irene Y. Chen, and Rajesh Ranganath. A Review of Challenges and Opportunities in Machine Learning for Health. AMIA Summits on Translational Science Proceedings 2020 (2020): 191.
  • Zhang, Haoran, Amy X. Lu, Mohamed Abdalla, Matthew McDermott, and Marzyeh Ghassemi. Hurtful Words: Quantifying Biases in Clinical Contextual Word Embeddings. Proceedings of the ACM Conference on Health, Inference, and Learning, pp. 110-120. 2020.
  • Suriyakumar, Vinith M., Nicolas Papernot, Anna Goldenberg, and Marzyeh Ghassemi. Chasing Your Long Tails: Differentially Private Prediction in Health Care Settings. Proceedings of the 2021 ACM Conference on Fairness, Accountability, and Transparency, pp. 723-734. 2021.

What behaviours are healthy?

A perfect model will fail if it is not used appropriately, and doesn’t conform well to the environment it will operate in. We work to define how models can interact with expert and non-expert users so that overall health practice and knowledge is actually improved.

  • Gaube, Susanne, Harini Suresh, Martina Raue, Alexander Merritt, Seth J. Berkowitz, Eva Lermer, Joseph F. Coughlin, John V. Guttag, Errol Colak, and Marzyeh Ghassemi. Do as AI say: susceptibility in deployment of clinical decision-aids. NPJ digital medicine 4, no. 1 (2021): 1-8.
  • Banerjee, Imon, Ananth Reddy Bhimireddy, John L. Burns, Leo Anthony Celi, Li-Ching Chen, Ramon Correa, Natalie Dullerud et al. Reading Race: AI Recognises Patient’s Racial Identity In Medical Images. arXiv preprint arXiv:2107.10356 (2021).