Key result
A machine learning algorithm combining multimodal physiological signals predicted late-onset sepsis in preterm infants with an AUC of 0.88 during the 6 hours preceding onset.
Why the study?
Does a machine learning algorithm using multimodal physiological signals accurately predict late-onset sepsis in preterm infants?
Case-Control (n=127)
Does a machine learning algorithm using multimodal physiological signals accurately predict late-onset sepsis in preterm infants?
Effect estimate: AUC 0.88
A machine learning algorithm combining cardiorespiratory and motion signals from standard patient monitors can accurately predict late-onset sepsis in preterm infants up to 6 hours before onset.
May aid early late-onset sepsis prediction in preterm infants; hypothesis-generating and requires prospective validation.
This study describes a machine learning algorithm based on multimodal signals obtained from a regular clinical patient monitor to predict late-onset sepsis (LOS) in preterm infants in a neonatal intensive care unit (NICU). The algorithm uses features that contain information on heart rate variability (HRV), respiration, and motion, based on continuously measured physiological waveforms including electrocardiogram (ECG) and chest impedance (CI). In this study, 127 preterm infants were included, of whom 59 were blood-culture-proven LOS patients and 68 were control patients. Features in 24 hours before the onset of sepsis (for the LOS group), and an age-matched onset time point (for the control group) were extracted and fed into machine learning classifiers together with gestational age (GA) and birth weight. We compared the prediction performance of several well-known classifiers using features extracted from different signal modalities (HRV, respiration, and motion) individually as well as their combinations. The prediction performance was evaluated using the area under the receiver-operating-characteristics curve (AUC). The best performance for LOS prediction was achieved by an extreme gradient boosting classifier combining features from all signal modalities, with an AUC of 0.88, a positive predictive value of 0.80, and a negative predictive value of 0.83 during 6 hours preceding LOS onset. Our study demonstrates the complementary predictive value of motion information in addition to cardiorespiratory information for LOS prediction. Furthermore, visualizations of how each feature impacts the algorithm strengthen its clinical decision support to predict LOS for individual patients in NICUs.
No takes yet. Share an insight, caveat, or question.
Varisco et al. (2022) conducted a case-control in Late-onset sepsis in preterm infants (n=127). Machine learning algorithm using multimodal physiological signals vs. Individual signal modalities was evaluated on Prediction of late-onset sepsis (AUC 0.88). A machine learning algorithm combining multimodal physiological signals predicted late-onset sepsis in preterm infants with an AUC of 0.88 during the 6 hours preceding onset.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: