Research

We aim to make use of observational data to gain insight into human health, and better prevent and treat disease. Our core research areas are methods for causal inference and explanation from complex data, discovery from observational biomedical data, nutrition, and decision-making.

Causality

How can we learn how things work when we can only observe, and not experiment on a system? This is the problem faced in many biomedical sciences, and is complicated by the fact that causes are not only composed of individual variables but complex entities like conjunctions (exercise and meals interact to affect blood glucose) and durations (smoking for decades is likelier to cause cancer than a single cigarette), and relationships happen at many timescales (weight loss has a gradual impact on glucose, eating affects it sooner). We develop new algorithms that can automatically extract these complex relationships from large-scale time series data.

  • L. Gomez and S. Kleinberg Causal Inference for Time Series Datasets with Partially Overlapping Variables, < Journal of Biomedical Informatics, 2025. [html]
  • M. Zheng, and S. Kleinberg. Using Domain Knowledge to Overcome Latent Variables in Causal Inference from Time Series, Machine Learning for Healthcare, 2019. [pdf]
  • S. Kleinberg. Causality, Probability, and Time, Cambridge University Press, 2012. [my book page] [Cambridge] [Amazon US]

Biomedical informatics

We aim to improve human health by helping clinicians and patients make better use of complex medical data. Our methods have been applied to automatically classify consciousness in ICU patients just from physiological data, predict glycemic responses to meals in people with diabetes, and identify risk factors for congestive heart failure. We have developed new technology along the way, including better methods for handling missing data and approaches for directly modeling the uncertainty in measurements.

  • Y. Shen, E. Choi, and S. Kleinberg. Predicting Postprandial Glycemic Responses With Limited Data in Type 1 and Type 2 Diabetes, Journal of Diabetes Science and Technology, 2025. [html]
  • L. A. Gomez, Q. Shen, K. Doyle, A. Vrosgou, A. Velazquez, M. Megjhani, S. Ghoshal, D. Roh, S. Agarwal, S. Park, J. Claassen, and S. Kleinberg. Classification of Level of Consciousness in a Neurological ICU Using Physiological Data, Neurocritical Care, 38(1) 118-128, 2023. [html]
  • T. T. Yavuz, J. Claassen, and S. Kleinberg, Lagged Correlations among Physiological Variables as Indicators of Consciousness in Stroke Patients, AMIA Annual Symposium, 2019. [pdf] [code on github]

Nutrition and dietary monitoring

Nutrition is a core part of maintaining health and preventing disease, yet it is difficult to obtain long-term dietary data and challenging to make use of machine learning methods with this complex data. We aim to change that, with low-burden methods of tracking diet and new methods to improve study design and variable representation.

  • J. D. Pleuss, A. L. Deierlein, and S. Kleinberg. Estimating Days Needed for Dietary Assessment in Pregnancy: A Modeling Study, The American Journal of Clinical Nutrition, 2025. [html]
  • S. Kleinberg, J. D. Pleuss, and A. L. Deierlein. Food Records Show Daily Variation in Diet During Pregnancy: Results From the Temporal Research in Eating, Nutrition, and Diet during Pregnancy (TREND-P) Study, The Journal of Nutrition, 2024. [html]
  • M. Mirtchouk, C. Merck, and S. Kleinberg. Automated Estimation of Food Type and Amount Consumed from Body-worn Audio and Motion Sensors, UbiComp, 2016. [pdf]

Decision-making

All of our work is motivated by helping people make better choices: helping clinicians decide on treatments, and helping patients better manage their chronic diseases. Thus we also work on understanding how people use information to make decisions in everyday situations.

  • E. Korshakova and S. Kleinberg. Evaluating Causal and Non-Causal Text Messages to Promote Physical Activity in Adults: A Randomized Pilot Study, JMIR Formative Research, 2025. [html]
  • S. Kleinberg and J. Marsh. Less is More: Information Needs, Information Wants, and What Makes Causal Models Useful, CRPI, 2023. [html]
  • M. Zheng, J. K. Marsh, J. V. Nickerson, and S. Kleinberg, How Causal Information Affects Decisions, Cognitive Research: Principles and Implications (CRPI) 5(6), 2020. [pdf]