Application of phonocardiography on preterm infants with patent ductus arteriosus

A Framework for AI-Assisted Detection of Patent Ductus Arteriosus from Neonatal Phonocardiogram

The current diagnosis of Congenital Heart Disease (CHD) in neonates relies on echocardiography. Its limited availability requires alternative screening procedures to prioritise newborns awaiting ultrasound. The routine screening for CHD is performed using a multidimensional clinical examination including (but not limited to) auscultation and pulse oximetry. While auscultation might be subjective with some heart abnormalities not always audible it increases the ability to detect heart defects. This work aims at developing an objective clinical decision support tool based on machine learning (ML) to facilitate differentiation of sounds with signatures of Patent Ductus Arteriosus (PDA)/CHDs, in clinical settings. The heart sounds are pre-processed and segmented, followed by feature extraction. The features are fed into a boosted decision tree classifier to estimate the probability of PDA or CHDs. Several mechanisms to combine information from different auscultation points, as well as consecutive sound cycles, are presented. The system is evaluated on a large clinical dataset of heart sounds from 265 term and late-preterm newborns recorded within the first six days of life. The developed system reaches an area under the curve (AUC) of 78% at detecting CHD and 77% at detecting PDA. The obtained results for PDA detection compare favourably with the level of accuracy achieved by an experienced neonatologist when assessed on the same cohort.

Heart Rate Assessment Immediately after Birth

BACKGROUND

Heart rate assessment immediately after birth in newborn infants is critical to the correct guidance of resuscitation efforts. There are disagreements as to the best method to measure heart rate.

OBJECTIVE

The aim of this study was to assess different methods of heart rate assessment in newborn infants at birth to determine the fastest and most accurate method.

METHODS

PubMed, EMBASE and Google Scholar were systematically searched using the following terms: 'infant', 'heart rate', 'monitoring', 'delivery room', 'resuscitation', 'stethoscope', 'auscultation', 'palpation', 'pulse oximetry', 'electrocardiogram', 'Doppler ultrasound', 'photoplethysmography' and 'wearable sensors'.

RESULTS

Eighteen studies were identified that described various methods of heart rate assessment in newborn infants immediately after birth. Studies examining auscultation, palpation, pulse oximetry, electrocardiography and Doppler ultrasound as ways to measure heart rate were included. Heart rate measurements by pulse oximetry are superior to auscultation and palpation, but there is contradictory evidence about its accuracy depending on whether the sensor is connected to the infant or the oximeter first. Several studies indicate that electrocardiogram provides a reliable heart rate faster than pulse oximetry. Doppler ultrasound shows potential for clinical use, however future evidence is needed to support this conclusion.

CONCLUSION

Heart rate assessment is important and there are many measurement methods. The accuracy of routinely applied methods varies, with palpation and auscultation being the least accurate and electrocardiogram being the most accurate. More research is needed on Doppler ultrasound before its clinical use.