Advanced analyses of physiological signals in the neonatal intensive care unit

Enabling Timely Medical Intervention by Exploring Health-Related Multivariate Time Series with a Hybrid Attentive Model

Modern healthcare practice, especially in intensive care units, produces a vast amount of multivariate time series of health-related data, e.g., multi-lead electrocardiogram (ECG), pulse waveform, blood pressure waveform and so on. As a result, timely and accurate prediction of medical intervention (e.g., intravenous injection) becomes possible, by exploring such semantic-rich time series. Existing works mainly focused on onset prediction at the granularity of hours that was not suitable for medication intervention in emergency medicine. This research proposes a Multi-Variable Hybrid Attentive Model (MVHA) to predict the impending need of medical intervention, by jointly mining multiple time series. Specifically, a two-level attention mechanism is designed to capture the pattern of fluctuations and trends of different time series. This work applied MVHA to the prediction of the impending intravenous injection need of critical patients at the intensive care units. Experiments on the MIMIC Waveform Database demonstrated that the proposed model achieves a prediction accuracy of 0.8475 and an ROC-AUC of 0.8318, which significantly outperforms baseline models.

Novel Methods of Measuring Adherence Patterns Reveal Adherence Phenotypes with Distinct Asthma Outcomes

RATIONALE

Poor adherence with asthma controller medication contributes to worse symptom control and increased exacerbation risk. Adherence is often expressed as the mean proportion of prescribed doses taken over a period of 6-12 months. New metrics may capture individual day-to-day variability patterns linked with distinct clinical outcomes.

OBJECTIVES

To test the hypotheses that novel time- and dose-based adherence variability metrics offer independent value to mean adherence in identifying distinct adherence patterns, that are associated with symptom control (Asthma Control Test [ACT] score) and exacerbation risk, using electronically-recorded medication data from a 6-month cluster randomized trial examining the effect of inhaler reminders on adherence.

METHODS

Adherence metrics were calculated from the first two months (months 0-2) of the study period. In addition to mean adherence (%prescribed puffs/day taken, PTmean), we examined novel metrics including: time adherence area-under-curve (T-AUC), reflecting cumulative gaps in adherence over time, entropy (H), reflecting disorder in the ways in which a patient changed their medication dose adherence from day to day, and standard deviation of the %prescribed puffs/day taken (PTSD). Dominant metrics identified from factor analysis were included in hierarchical clustering analysis. We compared the resultant clusters in terms of outcomes over months 2-6, and exacerbation risk over the entire study period.

RESULTS

Two factors explained >65% of the total variance in adherence, primarily driven by T-AUC and H. Two main patient clusters based on their adherence metrics were identified: Cluster 1 ("high time adherence", n=75) had better T-AUC, i.e. fewer gaps between medication-taking days, than Cluster 2 ("low time adherence", n=23). Though both clusters had similar symptom control at 2 months, Cluster 1 showed less subsequent decline in ACT over months 2-6 (median(IQR) change in ACT score: 1(-1, 4) vs -2(-3.75,0.75); p=0.012), and had better symptom control at 6 months (ACT score: 20(17, 23) vs 17 (15, 20); p=0.034). There were no significant differences between the clusters in terms of proportion of exacerbators or time to exacerbation.

CONCLUSIONS

Novel metrics showed that low time adherence was associated with greater risk of decline in asthma symptom control. Adherence patterns may exhibit 'memory' relevant to future clinical status, warranting validation in a larger dataset.

Video and audio processing in paediatrics: a review

OBJECTIVE

Video and sound acquisition and processing technologies have seen great improvements in recent decades, with many applications in the biomedical area. The aim of this paper is to review the overall state of the art of advances within these topics in paediatrics and to evaluate their potential application for monitoring in the neonatal intensive care unit (NICU).

APPROACH

For this purpose, more than 150 papers dealing with video and audio processing were reviewed. For both topics, clinical applications are described according to the considered cohorts-full-term newborns, infants and toddlers or preterm newborns. Then, processing methods are presented, in terms of data acquisition, feature extraction and characterization.

MAIN RESULTS

The paper first focuses on the exploitation of video recordings; these began to be automatically processed in the 2000s and we show that they have mainly been used to characterize infant motion. Other applications, including respiration and heart rate estimation and facial analysis, are also presented. Audio processing is then reviewed, with a focus on the analysis of crying. The first studies in this field focused on induced-pain cries and the newest ones deal with spontaneous cries; the analyses are mainly based on frequency features. Then, some papers dealing with non-cry signals are also discussed.

SIGNIFICANCE

Finally, we show that even if recent improvements in digital video and signal processing allow for increased automation of processing, the context of the NICU makes a fully automated analysis of long recordings problematic. A few proposals for overcoming some of the limitations are given.