# Integrated physiomarker, biomarker and clinical predictive analytics for early warning of sepsis and necrotizing enterocolitis in very low birth weight infants.

> **NIH NIH K23** · UNIVERSITY OF VIRGINIA · 2020 · $165,137

## Abstract

PROJECT ABSTRACT
Candidate: Dr. Brynne Sullivan is an Assistant Professor at the University of Virginia (UVA) with experience in
research involving clinical applications of predictive analytics in NICU patients. She is currently supported by
an internal mentored career develop award and enrolled in courses to earn a MSc degree in clinical research.
Career development plan and goals: The proposed training plan will establish Dr. Sullivan as a clinician-
investigator with expertise in predictive analytics to improve diagnosis and outcomes of life-threatening
inflammatory illness in very low birth weight infants (VLBW). Training activities during the award period include
graduate-level coursework and tutorial in time series analysis, clinical trial methodology and immunology.
Research Plan: Late-onset sepsis (LOS) and necrotizing enterocolitis (NEC) occur in 20% of VLBW infants and
cause significant morbidity and mortality. Early diagnosis and treatment can start the course to recovery before
serious injury occurs; therefore, tools to detect the first signs of illness could save lives and improve outcomes
for survivors. Physiomarkers, biomarkers and clinical risk markers of LOS and NEC exist, currently as separate
sources of data. The critical need for making a substantial difference in morbidity and mortality from LOS and
NEC is to integrate data to develop tools for earlier diagnosis and treatment. The objective is to develop
predictive analytics using clinical, physiomarker and biomarker data and test the hypothesis that integrating
these data improves algorithm performance. SPECIFIC AIMS: 1) Develop and validate a Pulse Oximetry
Warning Score for early detection of LOS and NEC using patterns in heart rate (HR) and oxygen saturation
(SpO2) from a large, multi-center data set and compare performance to predictive models using clinical risk
factors; 2) Determine whether biomarkers of immune activation inform risk of imminent LOS and NEC in the
context of physiomarker and clinical monitoring; 3) Test feasibility of integrated physiomarker, biomarker and
clinical monitoring in a pilot randomized trial to inform next steps of a multi-center trial that will be the focus of
an R01 proposal. The proposed research is consistent with the NICHD mission to ensure that all children have
the chance to fulfill their potential to live healthy and productive lives free from disease or disability.
Mentors: The primary mentor, Randall Moorman, MD, and secondary mentors, Karen Fairchild, MD and
William Petri, MD, PhD, have broad and diverse expertise and strong track records in successful completion of
NIH-supported studies and in mentoring trainees and junior faculty to become independent investigators.
Environment: The UVA Center for Advanced Medical Analytics and Dr. Petri's laboratory will provide the
intellectual environment and resources necessary to accomplish the goals of this proposal. The exceptional
resources and institutional support at UVA and outstandi...

## Key facts

- **NIH application ID:** 9882292
- **Project number:** 5K23HD097254-02
- **Recipient organization:** UNIVERSITY OF VIRGINIA
- **Principal Investigator:** Brynne Archer Sullivan
- **Activity code:** K23 (R01, R21, SBIR, etc.)
- **Funding institute:** NIH
- **Fiscal year:** 2020
- **Award amount:** $165,137
- **Award type:** 5
- **Project period:** 2019-07-01 → 2024-06-30

## Primary source

NIH RePORTER: https://reporter.nih.gov/project-details/9882292

## Citation

> US National Institutes of Health, RePORTER application 9882292, Integrated physiomarker, biomarker and clinical predictive analytics for early warning of sepsis and necrotizing enterocolitis in very low birth weight infants. (5K23HD097254-02). Retrieved via AI Analytics 2026-05-22 from https://api.ai-analytics.org/grant/nih/9882292. Licensed CC0.

---

*[NIH grants dataset](/datasets/nih-grants) · CC0 1.0*