Working Papers

1. Auditing Human Labels with Concept-based Decompositions
   D. Shanmugam, M. Zhong, A. Field, E. Pierson
   Under review

2. Machine learning reveals hidden diagnoses among underserved patients
   D. Shanmugam*, B. Hardy, A. Wang, S. Divakaran, J. Guttag, E. Pierson**, M. Barnett**
   Under review

3. Improving Spontaneous Labor Prediction with Electronic Health Record Data
   M. Krishnamoorthy*, D. Shanmugam*, D. Tjandra, A. Peahl, C. Pancaro, E. Ziedan, A. Kowalski, J. Wiens
   Under revision at Nature Health Communications, 2026

Selected Publications

See Google Scholar for a complete list.

• The Trillion-Dollar Algorithm: Lessons From Machine Learning For MA Risk Adjustment
  M. Johnson*, D. Shanmugam*, D. Meyers, J. Wiens, E. Pierson
  Health Affairs Forefront, 2026

• A Roadmap for Responsibly Removing Race from Clinical Algorithms
  D. Shanmugam*, S. Balachandar*, A. Chouldechova, J. Diao, K. Ferryman, A. Manrai, S. Pfohl, N. Powe, I. D. Raji, E. Pierson
  Nature Health, 2026

• Evaluating multiple models using labeled and unlabeled data
  D. Shanmugam*, S. Sadhuka*, M. Raghavan, J. Guttag, B. Berger**, E. Pierson**
  NeurIPS, 2025

• Learning disease progression models that capture health disparities
  E. Chiang, D. Shanmugam, A. Beecy, G. Sayer, N. Uriel, D. Estrin, N. Garg, E. Pierson
  CHIL, 2025 · Best Paper Award for Impact & Society
  Code

• Test-time augmentation improves efficiency in conformal prediction
  D. Shanmugam, H. Lu, S. Swaminarayan, J. Guttag
  CVPR, 2025

• Generative Artificial Intelligence in Medicine
  D. Shanmugam, M. Agarwal, R. Movva, I. Y. Chen, M. Ghassemi, M. Jacobs, E. Pierson
  Annual Review of Biomedical Data Science, 2025

• Use large language models to promote health equity
  E. Pierson*, D. Shanmugam*, R. Movva*, J. Kleinberg* … & many others!
  NEJM AI, 2025

• Quantifying Inequality in Underreported Medical Conditions
  D. Shanmugam, K. Hou, E. Pierson
  npj Women’s Health, 2024
  Code

• Coarse race data conceals disparities in clinical risk score performance
  R. Movva*, D. Shanmugam*, K. Hou, P. Pathak, J. Guttag, N. Garg, E. Pierson
  MLHC, 2023 · Honorable Mention, Best Findings Paper at ML4H 2023
  Code

• Kaleidoscope: Semantically-grounded, context-specific ML model evaluation
  H. Suresh, D. Shanmugam, T. Chen, A. Bryan, A. D’Amour, J. Guttag, A. Satyanarayan
  Conference on Human Factors in Computing Systems, 2023

Preprints

“Nothing stinks like a pile of unpublished writing” - Sylvia Plath, The Bell Jar

• More than One Way to Train a Net: Predicting Ovulation using Multiple Proxies
  D. Shanmugam, S. Hill, J. Guttag

• When and Why Test-Time Augmentation Works
  D. Shanmugam, D. Blalock, G. Balakrishnan, J. Guttag