Brian W Locke, MD, MSCI
Pulmonary and critical care physician-scientist focused on respiratory failure, diagnostic evidence, clinical prediction, and reproducible research informatics.
Assistant Professor of Research, Department of Pulmonary and Critical Care Medicine, Intermountain Medical Center
My work connects respiratory measurement, Bayesian and statistical reasoning, and open clinical research workflows to practical questions in pulmonary and critical care medicine.
About
I work in the Shock Trauma ICU and Schmidt Chest Clinic and serve as core faculty in the University of Utah Pulmonary and Critical Care Fellowship. I trained in computer science, medicine, medical education, and clinical investigation. My career purpose centers on respiratory failure, diagnostic evidence, clinical prediction, and reproducible research informatics.
Current research and teaching interests include:
- Respiratory measurement, hypercapnia, and follow-up after hospitalization
- Diagnostic evidence, likelihood ratios, and clinical reasoning
- Clinical prediction, EHR/NLP, and research informatics
- Pulmonary vascular imaging and outcomes
- Medical education around evidence-based reasoning
Research Approach
My academic work emphasizes methods that make clinical evidence more interpretable, reusable, and useful at the bedside:
Respiratory Measurement
Using blood gas, bicarbonate, transcutaneous CO2, imaging, and EHR-derived signals to study acute and chronic respiratory failure.
Diagnostic Evidence
Applying likelihood ratios, Bayesian reasoning, and information-theoretic framing to clinical reasoning and diagnostic uncertainty.
Clinical Prediction and Informatics
Building EHR, NLP, and clinical AI workflows that support transparent prediction, phenotyping, and cohort construction.
Reproducible Statistical Workflows
Developing open Stata, Python, and R pipelines for clinical research, including causal inference, prediction modeling, and clear visualization.
Medical Education
Teaching evidence-based medicine, statistics, respiratory physiology, and diagnostic reasoning for trainees and clinicians.
Selected Current Work
Hypercapnic Respiratory Failure
Measuring disease burden, care variation, and post-discharge needs after hospitalization with hypercapnic respiratory failure.
Diagnostic Evidence and Prediction
Developing EHR-derived phenotypes and prediction models that make respiratory measurement more interpretable and reproducible.
Reproducible Research Workflows
Maintaining open statistical code, teaching materials, and research artifacts that support transparent clinical investigation.
National Abstracts and Presentations
Sharing recent work on respiratory failure, clinical prediction, pulmonary vascular imaging, and medical education.
Teaching and Curriculum
Developing teaching around evidence-based reasoning, pulmonary and critical care medicine, and clinical research methods.
Materials
Public code, teaching materials, reproducible analyses, and project artifacts are available in the materials index.