Minimizing and quantifying uncertainty in AI-informed decisions: Applications in medicine.

Journal: Proceedings of the National Academy of Sciences of the United States of America
Published Date: Aug 26, 2025

Abstract

AI is now a cornerstone of modern dataset analysis. In many real world applications, practitioners are concerned with controlling specific kinds of errors, rather than minimizing the overall number of errors. For example, biomedical screening assays may primarily be concerned with mitigating the number of false positives rather than false negatives. Quantifying uncertainty in AI-based predictions, and in particular those controlling specific kinds of errors, remains theoretically and practically challenging. We develop a strategy called multidimensional informed generalized hypothesis testing (MIGHT) which we prove accurately quantifies uncertainty and confidence given sufficient data, and concomitantly controls for particular error types. Our key insight was that it is possible to integrate canonical cross-validation and parametric calibration procedures within a nonparametric ensemble method. Simulations demonstrate that while typical AI based-approaches cannot be trusted to obtain the truth, MIGHT can be. We apply MIGHT to answer an open question in liquid biopsies using circulating cell-free DNA (ccfDNA) in individuals with or without cancer: Which biomarkers, or combinations thereof, can we trust? Performance estimates produced by MIGHT on ccfDNA data have coefficients of variation that are often orders of magnitude lower than other state of the art algorithms such as support vector machines, random forests, and Transformers, while often also achieving higher sensitivity. We find that combinations of variable sets often decrease rather than increase sensitivity over the optimal single variable set because some variable sets add more noise than signal. This work demonstrates the importance of quantifying uncertainty and confidence-with theoretical guarantees-for the interpretation of real-world data.

Authors

  • Samuel D Curtis
    Department of Pharmacology and Molecular Sciences, Johns Hopkins University School of Medicine, Baltimore, MD 21205.
  • Sambit Panda
    Department of Biomedical Engineering, Johns Hopkins University, Baltimore, MD 21218.
  • Adam Li
    Department of Neurosurgery, University of Rochester Medical Center, Rochester, NY, USA.
  • Haoyin Xu
    Department of Biomedical Engineering, Johns Hopkins University, Baltimore, MD 21218.
  • Yuxin Bai
    Department of Biomedical Engineering, Johns Hopkins University, Baltimore, MD 21218.
  • Itsuki Ogihara
    Department of Biomedical Engineering, Johns Hopkins University, Baltimore, MD 21218.
  • Eliza O'Reilly
    Department of Applied Mathematics and Statistic, Johns Hopkins University, Baltimore, MD 21205.
  • Yuxuan Wang
    Department of Maxillofacial and Otorhinolaryngology Oncology, Tianjin's Clinical Research Center for Cancer, National Clinical Research Center for Cancer, Tianjin Medical University Cancer Institute and Hospital, Tianjin Medical University, Tianjin, China.
  • Lisa Dobbyn
    Department of Oncology, the Sidney Kimmel Cancer Center, Johns Hopkins University School of Medicine, Baltimore, MD 21205.
  • Maria Popoli
    Department of Oncology, the Sidney Kimmel Cancer Center, Johns Hopkins University School of Medicine, Baltimore, MD 21205.
  • Janine Ptak
    Department of Oncology, the Sidney Kimmel Cancer Center, Johns Hopkins University School of Medicine, Baltimore, MD 21205.
  • Nadine Nehme
    Department of Oncology, the Sidney Kimmel Cancer Center, Johns Hopkins University School of Medicine, Baltimore, MD 21205.
  • Natalie Silliman
    Department of Oncology, the Sidney Kimmel Cancer Center, Johns Hopkins University School of Medicine, Baltimore, MD 21205.
  • Jeanne Tie
    Division of Personalized Oncology, Walter and Eliza Hall Institute of Medical Research, Parkville, VIC 3052, Australia.
  • Peter Gibbs
    Department of Radiology, Memorial Sloan Kettering Cancer Center, New York, NY, USA; Department of Medical Physics, Memorial Sloan Kettering Cancer Center, New York, NY, USA.
  • Lan T Ho-Pham
    BioMedical Research Center, Pham Ngoc Thach University of Medicine, Ho Chi Minh City 72510, Vietnam.
  • Bich N H Tran
    Saigon Precision Medicine Research Center, Ho Chi Minh City 72512, Vietnam.
  • Thach S Tran
    Saigon Precision Medicine Research Center, Ho Chi Minh City 72512, Vietnam.
  • Tuan V Nguyen
  • Ehsan Irajizad
    Department of Biostatistics, The University of Texas MD Anderson Cancer Center, Houston, TX 77030.
  • Michael Goggins
    Department of Oncology, the Sidney Kimmel Cancer Center, Johns Hopkins University School of Medicine, Baltimore, MD 21205.
  • Christopher L Wolfgang
    Department of Surgery, The Sol Goldman Pancreatic Cancer Research Center, Johns Hopkins University School of Medicine, Baltimore, Maryland.
  • Tian-Li Wang
    Department of Pathology, Johns Hopkins University School of Medicine, Baltimore, MD 21205.
  • Ie-Ming Shih
    Department of Pathology, Johns Hopkins University School of Medicine, Baltimore, MD 21205.
  • Amanda Fader
    Department of Gynecology and Obstetrics, Johns Hopkins Medical Institutions, Baltimore, MD 21287.
  • Anne Marie Lennon
    Department of Radiology, Johns Hopkins University, 1800 Orleans Street, Baltimore, MD, 21287, USA; Department of Medicine, Johns Hopkins University, 1800 Orleans Street, Baltimore, MD, 21287, USA; Department of Surgery, Johns Hopkins University, 1800 Orleans Street, Baltimore, MD, 21287, USA; Sidney Kimmel Comprehensive Cancer Center, Department of Oncology, Johns Hopkins University, 1800 Orleans Street, Baltimore, MD, 21287, USA; Ludwig Center and Howard Hughes Medical Institute at the Sidney Kimmel Cancer Center, Johns Hopkins University, 1800 Orleans Street, Baltimore, MD, 21287, USA. Electronic address: felipe.oviedo@microsoft.com.
  • Ralph H Hruban
    Sol Goldman Pancreatic Cancer Research Center, Department of Pathology, Johns Hopkins University School of Medicine, Baltimore, Maryland.
  • Chetan Bettegowda
    Department of Neurosurgery, Johns Hopkins University School of Medicine, Baltimore, Maryland.
  • Lucy Gilbert
    Department of Oncology, McGill University Health Centre, Montreal, QC H4A 3J1, Canada.
  • Kenneth W Kinzler
    Sidney Kimmel Comprehensive Cancer Center, Johns Hopkins University School of Medicine, Baltimore, Maryland.
  • Nickolas Papadopoulos
    Department of Oncology, the Sidney Kimmel Cancer Center, Johns Hopkins University School of Medicine, Baltimore, MD 21205.
  • Bert Vogelstein
    Department of Molecular Biology and Genetics, Johns Hopkins School of Medicine, Baltimore, Maryland.
  • Joshua T Vogelstein
    Johns Hopkins University, Baltimore, Maryland.
  • Christopher Douville
    Department of Oncology, the Sidney Kimmel Cancer Center, Johns Hopkins University School of Medicine, Baltimore, MD 21205.

Keywords

External Resources

Popular Topics
Recent Journals