Machine Learning-Based Critical Congenital Heart Disease Screening Using Dual-Site Pulse Oximetry Measurements
Heather Siefkes1, Luca Cerny Oliveira2, Robert Koppel3
1Department of Pediatrics University of California Davis CA.
Insights
Machine learning pulse oximetry enhances critical congenital heart disease (CCHD) detection. This advanced screening, incorporating perfusion data and pulse delay, significantly improves early identification of CCHD and coarctation of the aorta (CoA).
Area of Science:
- Neonatal cardiology
- Medical device technology
- Machine learning in healthcare
Background:
- Standard oxygen saturation (SpO2) screening has limitations in the early detection of critical congenital heart disease (CCHD).
- Coarctation of the aorta (CoA) is a specific CCHD that can be challenging to detect via traditional methods.
- Enhancing pulse oximetry with additional physiological data may improve diagnostic accuracy for CCHD.
Purpose of the Study:
- To develop and evaluate a machine learning (ML) algorithm utilizing pulse oximetry features to improve CCHD detection.
- To assess the efficacy of incorporating perfusion data and radiofemoral pulse delay into ML algorithms for enhanced CCHD screening.
- To compare the diagnostic performance of the ML algorithm against standard SpO2 screening for CCHD and CoA.
Main Methods:
- Prospective enrollment of 523 newborns across six sites, including those with and without CCHD.
- Collection of simultaneous pre- and postductal pulse oximetry data, including perfusion metrics and pulse delay.
- Development and comparison of ML algorithms using one versus two time points and with or without pulse delay data against SpO2-alone algorithms.
Main Results:
- The SpO2-alone algorithm would miss 26.2% of CCHD cases.
- ML models incorporating two time points and pulse delay demonstrated significantly improved sensitivity for CCHD (92.86%) and CoA (66.67%) detection compared to SpO2 alone.
- All ML models achieved 100% specificity, and the area under the ROC curve improved substantially for both CCHD and CoA detection with the advanced ML model.
Conclusions:
- Machine learning pulse oximetry integrating oxygenation, perfusion data, and pulse delay at two time points offers a promising approach for improved CCHD and CoA detection.
- This advanced screening method has the potential to enhance early diagnosis of critical congenital heart defects within the first 48 hours of life.
Background:
Oxygen saturation (Spo2) screening has not led to earlier detection of critical congenital heart disease (CCHD). Adding pulse oximetry features (ie, perfusion data and radiofemoral pulse delay) may improve CCHD detection, especially coarctation of the aorta (CoA). We developed and tested a machine learning (ML) pulse oximetry algorithm to enhance CCHD detection.
Methods And Results:
Six sites prospectively enrolled newborns with and without CCHD and recorded simultaneous pre- and postductal pulse oximetry. We focused on models at 1 versus 2 time points and with/without pulse delay for our ML algorithms. The sensitivity, specificity, and area under the receiver operating characteristic curve were compared between the Spo2-alone and ML algorithms. A total of 523 newborns were enrolled (no CHD, 317; CHD, 74; CCHD, 132, of whom 21 had isolated CoA). When applying the Spo2-alone algorithm to all patients, 26.2% of CCHD would be missed. We narrowed the sample to patients with both 2 time point measurements and pulse-delay data (no CHD, 65; CCHD, 14) to compare ML performance. Among these patients, sensitivity for CCHD detection increased with both the addition of pulse delay and a second time point. All ML models had 100% specificity. With a 2-time-points+pulse-delay model, CCHD sensitivity increased to 92.86% (P=0.25) compared with Spo2 alone (71.43%), and CoA increased to 66.67% (P=0.5) from 0. The area under the receiver operating characteristic curve for CCHD and CoA detection significantly improved (0.96 versus 0.83 for CCHD, 0.83 versus 0.48 for CoA; both P=0.03) using the 2-time-points+pulse-delay model compared with Spo2 alone.
Conclusions:
ML pulse oximetry that combines oxygenation, perfusion data, and pulse delay at 2 time points may improve detection of CCHD and CoA within 48 hours after birth.
Registration:
URL: https://www.clinicaltrials.gov/study/NCT04056104?term=NCT04056104&rank=1; Unique identifier: NCT04056104.
Related Concept Videos
01:24Pulse Oximetry
Purpose
Average SpO2 values are greater than 95%. If the readings fall below 90%, it indicates that...
01:30Pulse rhythm
Conversely, an irregular pulse pattern is termed dysrhythmia, stemming from disruptions in cardiac...
01:19Guidelines For Measuring Vital Signs
Before taking a patient's vital signs, a nurse would consider and assess the patient's comfort level and ensure appropriate equipment is available.
01:13Special considerations while measuring pulse
01:19Special considerations while measuring oxygen saturation
Ensuring accuracy in vital sign recordings while prioritizing patient comfort and minimizing anxiety is...