Rishikesan Kamaleswaran
Duke University, NC·Biomedical Engineering
Patients in intensive care units (ICUs) can deteriorate quickly. Conditions like sepsis, the body's dangerous overreaction to infection, and acute respiratory distress syndrome (ARDS) are often recognized too late.
- Fields
Biomedical Engineering
Medicine & Clinical Research
Computer Science
- Topics
sepsis
acute respiratory distress syndrome
critical care
- Grants
- 3 known grants, through Jul 2031
What they work on
- Discover 'physiomarkers' in continuous ICU monitor data, including high-fidelity waveforms, that predict the onset of sepsis in patients who develop secondary infections after ICU admission.
- Combine physiomarkers with electronic medical record indicators in machine learning models that assess fluid (volume) responsiveness and identify new subtypes of sepsis.
- Develop and validate models that combine a small set of blood protein biomarkers with clinical data to diagnose acute respiratory distress syndrome (ARDS) early in ICU patients.
- Identify resilience markers and endotypes (biological subgroups) linked to ARDS recovery by integrating time-course transcriptomic, metabolomic, proteomic and lipidomic data with machine learning.
- Develop and refine the Trustworthy AI Implementation and Governance (TRIG) Framework for AI in aging and Alzheimer's disease research, using literature synthesis, a Delphi expert-consensus process and pilot testing.
- Help funded pilot projects audit AI models for performance and fairness, apply interpretability methods such as SHAP and LIME, and follow TRIPOD reproducibility standards.
More about the research
Hospitals already collect huge amounts of data from monitors, blood tests and medical records, but much of it goes unused in clinical decisions. Across three projects, the lab applies machine learning to physiological signals and molecular data for earlier diagnosis and to reveal hidden patient subtypes. It also develops standards for checking that medical artificial intelligence (AI) is trustworthy.
Why it matters
Earlier warning of sepsis and ARDS could help ICU teams start proven supportive treatments sooner, when each hour of delay raises the risk of death. Resilience markers could reveal new treatment targets for ARDS, which has no drug therapy. A standardized framework for testing AI fairness and reliability could make AI tools for older adults and people with dementia safer to use in research and care.
Known grants
From NIH and NSF research grants. Labs often have other funding we can't see.
Clinical Implementation and Data Science
NIH · Aug 2026 – Jul 2031
Resilience Endotype-derivation through Parsimonious Analysis and causal Inference of Recovery from Acute Respiratory Distress Syndrome (REPAIR-ARDS)
NIH · Sep 2025 – Jun 2030
Sepsis Physiomarkers for Appropriate Risk Knowledge of monitored patients in the ICU (SPARK-ICU)
NIH · Sep 2021 – Jun 2027
Closest labs
Li-Wei H Lehman
Massachusetts Institute of Technology, MA·Medicine & Clinical Research
Estimate what would happen under different treatment strategies for neuromuscular blockade and steroids in ARDS by emulating target trials in intensive care records, using g-methods that correct for the fact that sicker patients receive more aggressive care.
Romain Pirracchio
University of California-San Francisco, CA·Medicine & Clinical Research
Train machine learning models on electronic health record data from three health systems to estimate each patient's individual treatment effect from antibiotics, fluid resuscitation and vasopressors in suspected community-onset lung sepsis.
Jocelyn Grunwell
Emory University, GA·Medicine & Clinical Research
Test whether markers of neutrophil cell death, including neutrophil extracellular traps (NETs), in blood and airway samples taken twice in the first week of intubation predict longer intubation or death in children with PARDS.
The school
Duke University: 61 labs on PI Pages, $1.6B a year on research, 417 research doctorates a year.
Reach out
A start for your email. Fill in the brackets.