Brian Carlson

Brian Carlson
Research Associate Professor
Molecular and Integrative Physiology
Medical School
University of Michigan
Molecular and Integrative Physiology
2800 Plymouth Road NCRC B10/A126
Ann Arbor, Michigan 48109
[email protected]
Available to mentor
Brian Carlson
Brian Carlson
Research Associate Professor
  • About
  • Qualifications
  • Center Memberships
  • Research Overview
  • Recent Publications
  • Manage Your Profile

  • About

    Dr. Carlson is a Research Associate Professor in the Department of Molecular and Integrative Physiology. His lab has focused over the past decade on using physics and engineering based approaches to understand and describe physiological systems in health and disease.

    Dr. Carlson earned PhD in Mechanical Engineering in 2001 where his thesis focused on alterations in blood flow patterns within the microvasculature in sickle cell disease. After completing his PhD he conducted postdoctoral research at the University of Arizona working with Dr. Timothy Secomb and at the University of Washington working with Dr. James Bassingthwaighte. Before joining the University of Michigan, he was Assistant Professor of Physiology at the Medical College of Wisconsin.

    Qualifications

    • Postdoctoral Research Fellow
      University of Washington, Bioengineering, Seattle, United States
      2005 - 2007
      Postdoctoral Fellowship
    • Postdoctoral Research Fellow
      University of Arizona, Physiology, Tucson, United States
      2002 - 2005
      Postdoctoral Fellowship
    • PhD
      University of California, Berkeley, California, United States
      1988 - 2001
    • BS
      University of California, Berkeley, California, United States
      1980 - 1986

    Center Memberships

    • Center Member
      Weil Institute for Critical Care Research
    • Center Member
      Samuel and Jean Frankel Cardiovascular Center

    Research Overview

    Dr. Carlson’s specific research interests center on:

    (1) Patient-specific modeling. Clinical data used to diagnose heart failure can be analyzed using machine learning tools to find out which patients have a similar upper level phenotype possibly leading to a more precise patient-specific treatment in the clinic. We have taken this one step further and are now using cardiovascular systems models combined with the clinical data to learn a bit more about what these patient phenotypes have in common at a lower mechanistic level in the cardiovascular system. Heart failure with preserved ejection fraction (HFpEF) patients who are difficult to consistently treat have been successfully classified into three different subphenotypes and current work is focused on how these groups may be treated more efficiently. This methodology has also been used to track recovery trajectory after heart transplantation.

    (2) Cardiovascular dynamics. Sympathetic tone plays an important role in the acute response of the cardiovascular system to changes in pressure and blood volume. Understanding how left ventricular pressure and volume changes with respect to changing sympathetic tone is of great interest and whole body cardiovascular models along with experimental studies are being used to quantify this relationship.

    (3) Stem cell derived cardiomyocytes. Pluripotent stem cells can be differentiated into cardiomyocytes to provide a source of patient specific cells for pharmacological testing. How cardiomyocyte-like these cells are is still an open question. Theoretical work coupled with experimental work with collaborators at University of Wisconsin-Madison and Washington University at St. Louis are helping to understand how these cells change in their electrophysiological function and force generation capabilities immediately after differentiation and during cell maturation in culture.

    (4) Blood flow regulation. Mechanical (pressure and shear stress), neuronal (sympathetic tone), metabolic (ATP release) and humoral (circulating small molecules, e.g. Angiotensin II) stimuli are integrated locally in the resistance vessels of the vasculature to govern blood flow. Research on this topic focuses on utilizing experimental and theoretical methods to define the mechanisms transducing these stimuli into changes in vessel caliber and hence blood flow.

    (5) Semantic-based methods of model description and sharing. Computational models are developed with different standards and conventions in labs all over the world. In order to share, reuse and merge these models within the community, a method of attaching unambiguous biological and physics-based descriptors to components of the model is needed. Semantic-based methodologies to solve this problem are being developed with collaborators at the University of Washington and Medical College of Wisconsin.

    Recent Publications

    See All Publications
    • Preprint
      Physiology-Informed Digital Twin-AI Framework Predicts Pacing Therapy Response in HFpEF
      Gu F, Infeld M, Schenk NA, Wan H, Krishnan MJ, Cyr JA, Sturgess VE, Wittrup E, Ježek F, Carlson BE, van Loon T, Hua X, Tang Y-D, Kayvan N, Hummel SL, Lumens J, Meyer M, Beard DA. 2026 Mar 11; medRxiv, DOI:10.64898/2026.03.06.26347199
    • Journal Article
      Identification of digital twins to guide interpretable AI for diagnosis and prognosis in heart failure
      Gu F, Meyer AJ, Ježek F, Zhang S, Catalan T, Miller A, Schenk NA, Sturgess VE, Uceda D, Li R, Wittrup E, Hua X, Carlson BE, Tang YD, Raza F, Najarian K, Hummel SL, Beard DA. Npj Digital Medicine, 2025 Dec 1; 8 (1): DOI:10.1038/s41746-025-01501-9
    • Journal Article
      Machine learning-based phenotyping and assessment of treatment responses in heart failure with preserved ejection fraction
      Li R, Liu Y, Zhao Z, Zhang C, Dong W, Qi Y, Gao J, Gu F, Carlson BE, Beard DA, Hummel SL, Chen B, Fan H, Gu X, Hua X, Tang YD. Eclinicalmedicine, 2025 Oct 1; 88: DOI:10.1016/j.eclinm.2025.103462
    • Proceeding / Abstract / Poster
      Abstract Wed136: Integration of Mechanistic Fontan Circulatory Models with Interpretable Machine Learning Classifiers
      Schenk N, Egbe A, Carlson B, Beard D. Circulation Research, 2025 Oct 9; 137 (Suppl_1): DOI:10.1161/res.137.suppl_1.wed136
    • Proceeding / Abstract / Poster
      Physiology-Driven Digital Twin for Pacing Response Prediction and Mechanistic Insights in Heart Failure with Preserved Ejection Fraction
      Gu F, Krishnan MJ, Meyer AJ, Zhang S, Ježek F, Carlson BE, Kayvan N, Hummel SL, Beard DA. Physiology, 2025 May 14; 40 (S1): 2084 DOI:10.1152/physiol.2025.40.s1.2084
    • Journal Article
      The importance of incorporating ventricular–ventricular interaction (VVI) in the study of pulmonary hypertension
      Colunga A, Carlson BE, Olufsen MS. Mathematical Biosciences, 2024 Sep 1; 375: DOI:10.1016/j.mbs.2024.109242
      PMID: 38944112
    • Journal Article
      Integrated modeling and simulation of recruitment of myocardial perfusion and oxygen delivery in exercise
      Sturgess VE, Tune JD, Figueroa CA, Carlson BE, Beard DA. Journal of Molecular and Cellular Cardiology, 2024 Jul 1; 192: 94 - 108. DOI:10.1016/j.yjmcc.2024.05.006
      PMID: 38754551
    • Preprint
      Identification of Digital Twins to Guide Interpretable AI for Diagnosis and Prognosis in Heart Failure
      Gu F, Meyer A, Ježek F, Zhang S, Catalan T, Miller A, Schenk N, Sturgess V, Uceda D, Li R, Wittrup E, Hua X, Carlson B, Tang Y-D, Raza F, Najarian K, Hummel S, Beard D. 2024 Nov 13; medRxiv, DOI:10.1101/2024.11.11.24317106