Heart rate patterns predicting cerebral palsy in preterm infants

Letzkus, Lisa; Picavia, Robin; Lyons, Genevieve; Brandberg, Jackson; Qiu, Jiaxing; Kausch, Sherry; Lake, Doug; Fairchild, Karen

doi:10.1038/s41390-023-02853-2

Cited by 5 publications

(3 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This suggests HCTSA is an especially good approach for analyzing time series in settings where there has not been an extensive amount of prior research. As an example of this, HCTSA was recently used to identify a new signature of high HR variability around PMA of 36 weeks as a predictor of cerebral palsy in preterm infants (Letzkus et al 2023).…”

Section: Discussionmentioning

confidence: 99%

Highly comparative time series analysis of oxygen saturation and heart rate to predict respiratory outcomes in extremely preterm infants

Qiu,

Di Fiore,

Krishnamurthi

et al. 2024

Physiol. Meas.

View full text Add to dashboard Cite

Objective: Highly comparative time series analysis (HCTSA) is a novel approach involving massive feature extraction using publicly available code from many disciplines. The Prematurity-Related Ventilatory Control (Pre-Vent) observational multicenter prospective study collected bedside monitor data from >700 extremely preterm infants to identify physiologic features that predict respiratory outcomes. We calculated a subset of 33 HCTSA features on >7M 10-minute windows of oxygen saturation (SPO2) and heart rate (HR) from the PreVent cohort to quantify predictive performance. This subset included representatives previously identified using unsupervised clustering on >3500 HCTSA algorithms. Performance of each feature was measured by individual area under the receiver operating curve (AUC) at various days of life and binary respiratory outcomes. We hypothesized that the best HCTSA algorithms would compare favorably to optimal PreVent physiologic predictor IH90_DPE (duration per event of intermittent hypoxemia events below 90%).Main Results: The top HCTSA features were from a cluster of algorithms associated with the autocorrelation of SPO2 time series and identified low frequency patterns of desaturation as high risk. These features had comparable performance to and were highly correlated with IH90_DPE but perhaps measure the physiologic status of an infant in a more robust way that warrants further investigation. The top HR HCTSA features were symbolic transformation measures that had previously been identified as strong predictors of neonatal mortality. HR metrics were only important predictors at early days of life which was likely due to the larger proportion of infants whose outcome was death by any cause. A simple HCTSA model using 3 top features outperformed IH90_DPE at day of life 7 (.778 versus .729) but was essentially equivalent at day of life 28 (.849 versus .850). These results validated the utility of a representative HCTSA approach but also provides additional evidence supporting IH90\_DPE as an optimal predictor of respiratory outcomes.

show abstract

Section: Discussionmentioning

confidence: 99%

Highly comparative time series analysis of oxygen saturation and heart rate to predict respiratory outcomes in extremely preterm infants

Qiu,

Di Fiore,

Krishnamurthi

et al. 2024

Physiol. Meas.

View full text Add to dashboard Cite

show abstract

“…Recently, HR and oxygen saturation (SpO2) features have been associated with other adverse neonatal outcomes. An algorithm that detects a lack of successive HR increases predicts all-cause mortality (7), HR skewness predicts autism spectrum disorder (8), HR-SPO2 cross-correlation predicts sepsis and necrotizing enterocolitis (NEC) (9), and complex lagged-difference HR metrics predict cerebral palsy (10).…”

Section: Introductionmentioning

confidence: 99%

Can machine learning or deep learning discover novel signatures of illness in continuous cardiorespiratory monitoring data?

Sullivan,

Mesner,

Niestroy

et al. 2024

Preprint

View full text Add to dashboard Cite

Background: Cardiorespiratory deterioration due to sepsis is a leading cause of morbidity and mortality for extremely premature infants with very low birth weight (VLBW, birthweight <1500g). Abnormal heart rate (HR) patterns precede the clinical diagnosis of late-onset sepsis in this population. Decades ago, clinicians recognized a pattern of reduced HR variability and increased HR decelerations in electrocardiogram tracings of septic preterm infants. A predictive logistic regression model was developed from this finding using mathematical algorithms developed to detect this signature of illness. Display of this model as the fold increase in risk of imminent sepsis reduced mortality in a large randomized trial. Here, we sought to determine if machine learning or deep learning approaches would identify this rare but distinctive signature of sepsis in VLBW infants. Methods: We studied VLBW infants admitted from 2012 to 2021 to a regional Level IV NICU. We collected one-hour HR time series data from bedside monitoring sampled at 0.5 Hz (n=300 HR values per series) throughout the NICU admission. First, we applied the principles of highly comparative time series analysis (HCTSA) to generate many mathematical time series features and combined them in a machine learning model. Next, we used deep learning in the form of a convolutional neural network on the raw data to learn the HR features. The output was a set of HR records determined by HCTSA or deep learning to be at high risk for imminent sepsis. Results: We analyzed data from 566 infants with 61 episodes of sepsis. HCTSA and deep learning models predicted sepsis with high out-of-sample validation metrics. The riskiest records determined by both approaches demonstrated the previously identified HR signatures- reduced variability and increased decelerations. Conclusions: We tested the ability of unguided machine learning approaches to detect the novel HR signature of sepsis in VLBW infants previously identified by human experts. Our main finding is that the computerized approach returned the same result - it identified heart rate characteristics of reduced variability and transient decelerations. We conclude that unguided machine learning can be as effective as human experts in identifying even a very rare phenotype in clinical data.

show abstract

“…Recently, Letzkus et al have created a prediction model for cerebral palsy risk in premature infants, using HRV parameters from the early neonatal period. Their model, which exhibited an 85.3% predictability for the development of cerebral palsy, combined clinical features with HRV parameters (obtained at 1 week postpartum and 37 gestational weeks) [30]. HRV has also been intensely studied in adult patients with psychiatric disorders.…”

Section: Introductionmentioning

confidence: 99%

Anthracycline’s Effects on Heart Rate Variability in Children with Acute Lymphoblastic Leukemia: Early Toxicity Signs—Pilot Study

Lazar,

Cainap,

Maniu

et al. 2023

JCM

View full text Add to dashboard Cite

Anthracycline treatments are known to cause cardiotoxic long-term side effects in cancer survivors. Recently, a decrease in heart rate variability (HRV) has been identified in these patients, signaling autonomic dysfunction and altered cardiac fitness. This study aimed at evaluating changes in HRV in children treated with anthracyclines. A total of 35 pediatric patients with acute lymphoblastic leukemia were evaluated by means of a 24 h Holter ECG, at baseline and after reaching half the total cumulative dose of doxorubicin equivalent (120 mg/m2). Parameters of HRV were assessed, as well as any arrhythmic episodes, bradycardia and tachycardia percentages. The results showed a significant decrease in both time-domain and frequency-domain HRV parameters, following anthracycline treatment. The low-frequency (LF) to high-frequency (HF) parameters’ ratio also displayed a significant difference (p = 0.035), suggestive of early cardiac autonomic dysfunction. Of note, none of the patients presented symptoms of heart disease or elevated troponins, and only two patients presented echocardiographic signs of diastolic dysfunction. The present study showed that cardiac autonomic nervous system regulation is compromised in children treated with anthracyclines even before reaching the total cumulative dose. Therefore, HRV parameters could be the first indicators of subclinical cardiac toxicity, making Holter ECG monitoring of the oncological patient a necessity.

show abstract

Heart rate patterns predicting cerebral palsy in preterm infants

Cited by 5 publications

References 32 publications

Highly comparative time series analysis of oxygen saturation and heart rate to predict respiratory outcomes in extremely preterm infants

Highly comparative time series analysis of oxygen saturation and heart rate to predict respiratory outcomes in extremely preterm infants

Can machine learning or deep learning discover novel signatures of illness in continuous cardiorespiratory monitoring data?

Anthracycline’s Effects on Heart Rate Variability in Children with Acute Lymphoblastic Leukemia: Early Toxicity Signs—Pilot Study

Contact Info

Product

Resources

About