Physiomarker for early, equitable cerebral palsy prediction

/ABSTRACT Significance: Cerebral palsy (CP) is primarily recognized as a motor disability but is often accompanied by significant comorbidities that impact overall development, quality of life, and environmental engagement. Many infants at high risk for cerebral palsy start their journey in the neonatal intensive care unit (NICU) due to being born prematurely. CP is particularly prevalent in Black infants, underscoring critical health disparities that demand attention. Early identification is vital, as targeted therapies post-NICU can enhance developmental trajectories. However, current diagnostic methods for CP are not consistently accessible across all NICUs, highlighting the need for a universal screening tool similar to the hearing screen to facilitate timely referrals for advanced evaluation and CP-specific interventions. All NICU infants require continuous vital sign monitoring, presenting an opportunity for early risk assessment through analytics that identify central nervous system dysfunction. Our goal is to develop a novel physiologic biomarker ("physiomarker") for CP risk by analyzing continuous NICU waveform and vital sign data. Preliminary data indicate that infants later diagnosed with CP exhibit distinct heart rate patterns during their NICU stay. Our innovations include Highly Comparative Time Series Analysis and the FAIRSCAPE platform for global data sharing, enabling detection of abnormal cardiorespiratory patterns linked to CP and fostering collaboration among researchers. We will utilize both high- resolution (250 Hz) waveform data and universally accessible low-resolution metrics to create a scalable and equitable screening tool for all NICUs. Aim 1: Identify cardiorespiratory control signatures linked to CP using high-resolution waveform data, analyzing heart rate variability, apnea patterns, and cardiorespiratory coupling to detect central nervous system dysregulation. Aim 2: Develop a physiomarker for CP risk using low-resolution heart rate and pulse oximeter data (0.5 Hz), universally available in all NICUs, guiding early and equitable referrals for advanced testing and targeted therapies. Our experienced investigator team has NIH-funded research and access to extensive preterm infant data from eight NICUs across the U.S. Strong clinician-data science partnerships highlight the importance of interdisciplinary collaboration in developing predictive models. We have successfully predicted multiple adverse outcomes, including neurodevelopmental impairment, through our datasets and validated methodologies. Successful completion of this proposal will result in a translational physiomarker for CP risk, providing accessible early CP risk screening akin to the hearing screen and facilitating timely interventions that can alter the lifelong trajectory of high-risk infants and families. Project Number: 1R01HD119199-01 | Fiscal Year: 2025 | NIH Institute/Center: Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD) | Principal Investigator: Lisa Letzkus | Institution: UNIVERSITY OF VIRGINIA, CHARLOTTESVILLE, VA | Award Amount: $630,861 | Activity Code: R01 | Study Section: Special Emphasis Panel[ZRG1 HSS-R (90)] View on NIH RePORTER: https://reporter.nih.gov/project-details/1R01HD11919901