Reducing False Alarm Rates in Neonatal Intensive Care: A New Machine Learning Approach

Ostojić, Draženka; Guglielmini, Sabino; Moser, Virginie; Fauchère, Jean‐Claude; Bucher, Hans Ulrich; Bassler, Dirk; Wolf, Martin; Kleiser, Stefan; Scholkmann, Felix; Karen, Tanja

doi:10.1007/978-3-030-34461-0_36

Cited by 20 publications

(13 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…al. [17]. Using HR and SpO 2 with four machine learning algorithms, they classified alarms responded to by the medical team as true or false alarms.…”

Section: Related Workmentioning

confidence: 99%

“…Nonetheless, not all alarms in a (N)ICU need to be responded to immediately if they are not yet considered clinically urgent [9,13]. Other reasons behind the lack of response for some alarms include self-patient stabilization, a large number of false or irrelevant alarms (up to 70% of total number of alarms delivered to caregivers), or work overload in the medical team [7,13,17]. Alarm fatigue is the term used for the overload of monitor alarms, resulting in desensitization to alarms in caregivers (particularly in nurses) and could potentially lead to missing important ones [6,14,19].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Listen to the Real Experts: Detecting Need of Caregiver Response in a NICU using Multimodal Monitoring Signals

Quiros

Varisco

Zhan

et al. 2021

Companion Publication of the 2021 International Conference on Multimodal Interaction

View full text Add to dashboard Cite

“…al. [17]. Using HR and SpO 2 with four machine learning algorithms, they classified alarms responded to by the medical team as true or false alarms.…”

Section: Related Workmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Listen to the Real Experts: Detecting Need of Caregiver Response in a NICU using Multimodal Monitoring Signals

Quiros

Varisco

Zhan

et al. 2021

Companion Publication of the 2021 International Conference on Multimodal Interaction

View full text Add to dashboard Cite

“…In this difficult task, NICU monitors based on ML algorithms could be useful for clinicians: integrating the infant's past and present data, they could discriminate among the several pathologies that could affect the infant, thus pointing at the best preventive and therapeutic strategy. ML algorithms have already helped in predicting the risk of various conditions (e.g., cerebral hemorrhage and hyperbilirubinemia) and in reducing false alarms, thus improving neonatal care (Daunhawer et al, 2019;Malacova et al, 2020;Ostojic et al, 2020;Turova et al, 2020;Vassar et al, 2020).…”

Section: A Technological Analysis To Efficiently Monitor and Use Hrv mentioning

confidence: 99%

“…If all these factors can impact HRV, thereby biasing its measurement, a real-time monitor could overcome these biases by recording how HRV changes before, during, and after the occurrence of these factors. Again, ML algorithms could help cope with these confounding factors: by weighting and integrating their effects, such technology could help reduce false alarms, exactly as those algorithms that can detect artifacts due to baby's movements (Ostojic et al, 2020).…”

Section: Factors That Can Alter Hrv Assessmentmentioning

confidence: 99%

“…For this purpose, modern machine learning (ML) and artificial intelligence (AI) algorithms could play a crucial role: through their computational power, they could define models capable of managing the complex physiological interactions between HRV, ANS, and the whole organism, thus boosting our ability to predict the infant's prognosis. Indeed, we already have experiences about the clinical usefulness of ML in both neonatology (Semenova et al, 2018;Ostojic et al, 2020) and HRV analysis (Chiew et al, 2019;Lin et al, 2020).…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Heart Rate Variability in the Perinatal Period: A Critical and Conceptual Review

Chiera¹,

Cerritelli²,

Casini³

et al. 2020

Front. Neurosci.

View full text Add to dashboard Cite

Neonatal intensive care units (NICUs) greatly expand the use of technology. There is a need to accurately diagnose discomfort, pain, and complications, such as sepsis, mainly before they occur. While specific treatments are possible, they are often timeconsuming, invasive, or painful, with detrimental effects for the development of the infant. In the last 40 years, heart rate variability (HRV) has emerged as a noninvasive measurement to monitor newborns and infants, but it still is underused. Hence, the present paper aims to review the utility of HRV in neonatology and the instruments available to assess it, showing how HRV could be an innovative tool in the years to come. When continuously monitored, HRV could help assess the baby's overall wellbeing and neurological development to detect stress-/painrelated behaviors or pathological conditions, such as respiratory distress syndrome and hyperbilirubinemia, to address when to perform procedures to reduce the baby's stress/pain and interventions, such as therapeutic hypothermia, and to avoid severe complications, such as sepsis and necrotizing enterocolitis, thus reducing mortality. Based on literature and previous experiences, the first step to efficiently introduce HRV in the NICUs could consist in a monitoring system that uses photoplethysmography, which is low-cost and non-invasive, and displays one or a few metrics with good clinical utility. However, to fully harness HRV clinical potential and to greatly improve neonatal care, the monitoring systems will have to rely on modern bioinformatics (machine learning and artificial intelligence algorithms), which could easily integrate infant's HRV metrics, vital signs, and especially past history, thus elaborating models capable to efficiently monitor and predict the infant's clinical conditions. For this reason, hospitals and institutions will have to establish tight collaborations between the obstetric, neonatal, and pediatric departments: this way, healthcare would truly improve in every stage of the perinatal period (from conception to the first years of life), since information about patients' health would flow freely among different professionals, and high-quality research could be performed integrating the data recorded in those departments.

show abstract

Machine Learning Distinguishes Familiar from Unfamiliar Pairs of Subjects Performing an Eye Contact Task: A Systemic Physiology Augmented Functional Near-Infrared Spectroscopy Hyperscanning Study

Guglielmini

Bopp

Marcar

et al. 2022

Advances in Experimental Medicine and Biology

Self Cite

View full text Add to dashboard Cite

Background: Eye contact is an important aspect of human communication and social interactions. Changes in brain and systemic physiological activity associated with interactions between humans can be measured with systemic physiology augmented functional near-infrared spectroscopy (SPA-fNIRS) hyperscanning, enabling to determine inter-brain and inter-body synchronization. In a previous study, we found that pairs of subjects that are socially connected show higher brain and body synchrony. Aim: To enable a deeper understanding, our aim was to build and automatically detect the best set of features to distinguish between two different groups (familiar and unfamiliar pairs). Material and methods: We defined several features based on the Spearman correlation and wavelet transform coherence (WTC) of biosignals measured on 23 pairs of subjects (13 familiar and 10 unfamiliar pairs) during eye contact for 10 min. Additional custom features that identify the maximum brain-to-body coupling instants between pairs were generated. Results: After testing on combinations of different feature extraction methods, four subsets of features with the strongest discrimination power have been taken into account to train a decision tree (DT) and a random forest (RF) machine learning (ML) algorithm. We have obtained 95.65% classification accuracy using a leave-one-out cross-validation. The coupling features which represent the three maximum mean values resulting from the sum of 7 time-dependent WTC signals (oxyhemoglobin concentration of the right prefrontal region, total hemoglobin concentration of the left and right prefrontal region, heart rate, electrodermal activity on the left and right wrist, and skin temperature on the right wrist) played an essential role in the classification accuracy. Conclusion: Training the DT and RF-ML algorithm with combined brain and systemic physiology data provided higher accuracy than training it only with brain or systemic data alone. The results demonstrate the power of the SPA-fNIRS hyperscanning approach and the potential in applying ML to investigate the strength of social bonds in a wide range of social interaction contexts.

show abstract

Reducing False Alarm Rates in Neonatal Intensive Care: A New Machine Learning Approach

Cited by 20 publications

References 14 publications

Listen to the Real Experts: Detecting Need of Caregiver Response in a NICU using Multimodal Monitoring Signals

Listen to the Real Experts: Detecting Need of Caregiver Response in a NICU using Multimodal Monitoring Signals

Heart Rate Variability in the Perinatal Period: A Critical and Conceptual Review

Machine Learning Distinguishes Familiar from Unfamiliar Pairs of Subjects Performing an Eye Contact Task: A Systemic Physiology Augmented Functional Near-Infrared Spectroscopy Hyperscanning Study

Contact Info

Product

Resources

About