Automatic fetal movement recognition from multi-channel accelerometry data.

BACKGROUND AND OBJECTIVE: Significant health care resources are allocated to monitoring high risk pregnancies to minimize growth compromise, reduce morbidity and prevent stillbirth. Fetal movement has been recognized as an important indicator of fetal health. Studies have shown that 25% of pregnancies with decreased fetal movement in the third trimester led to poor outcomes at birth. The studies have also shown that maternal perception of fetal movement is highly subjective and varies from person to person. A non-invasive system for fetal movement detection that can be used outside hospital would represent an advance in at-home monitoring of at-risk pregnancies. This is a challenging task that requires the use of advanced signal processing techniques to differentiate genuine fetal movements from contaminating artefacts. METHODS: This manuscript proposes a novel algorithm for automatic fetal movement recognition using data collected from wearable tri-axial accelerometers strategically placed on the maternal abdomen. The novelty of the work resides in the efficient removal of artefacts and in distinctive feature extraction. The proposed algorithm used independent component analysis (ICA) for dimensionality reduction and artefact removal. A supplemental technique based on discrete wavelet transform (DWT) was also used to remove artefacts. RESULTS: To identify fetal movements, 31 features were extracted from the acceleration data. Based on these features, several classifiers were used to distinguish fetal from non-fetal movements. Robustness of the classifiers was tested for various concentrations of artefacts in the classification data. The best performance was achieved by Bagging classifier algorithm, with random forest as its basis classifier, yielding an accuracy ranging from 87.6% to 95.8% depending on the artefact concentration level. CONCLUSIONS: A high performance detection of fetal movements can be achieved using accelerometery-based systems suitable for long-term monitoring.

Sleep architectural dysfunction and undiagnosed obstructive sleep apnea after chronic ischemic stroke.

OBJECTIVE/BACKGROUND: Sleep-wake dysfunction is bidirectionally associated with the incidence and evolution of acute stroke. It remains unclear whether sleep disturbances are transient post-stroke or are potentially enduring sequelae in chronic stroke. Here, we characterize sleep architectural dysfunction, sleep-respiratory parameters, and hemispheric sleep in ischemic stroke patients in the chronic recovery phase compared to healthy controls. PATIENTS/METHODS: Radiologically confirmed ischemic stroke patients (n = 28) and matched control participants (n = 16) were tested with ambulatory polysomnography, bi-hemispheric sleep EEG, and demographic, stroke-severity, mood, and sleep-circadian questionnaires. RESULTS: Twenty-eight stroke patients (22 men; mean age = 69.61 ± 7.4 years) were cross-sectionally evaluated 4.1 ± 0.9 years after mild-moderate ischemic stroke (baseline NIHSS: 3.0 ± 2.0). Fifty-seven percent of stroke patients (n = 16) exhibited undiagnosed moderate-to-severe obstructive sleep apnea (apnea-hypopnea index >15). Despite no difference in total sleep or wake after sleep onset, stroke patients had reduced slow-wave sleep time (66.25 min vs 99.26 min, p = 0.02), increased time in non-rapid-eye-movement (NREM) stages 1-2 (NREM-1: 48.43 vs 28.95, p = 0.03; NREM-2: 142.61 vs 115.87, p = 0.02), and a higher arousal index (21.46 vs 14.43, p = 0.03) when compared to controls. Controlling for sleep apnea severity did not attenuate the magnitude of sleep architectural differences between groups (NREM 1-3=ηp2 >0.07). We observed no differences in ipsilesionally versus contralesionally scored sleep architecture. CONCLUSIONS: Fifty-seven percent of chronic stroke patients had undiagnosed moderate-severe obstructive sleep apnea and reduced slow-wave sleep with potentially compensatory increases in NREM 1-2 sleep relative to controls. Formal sleep studies are warranted after stroke, even in the absence of self-reported history of sleep-wake pathology.

Regional neurodegeneration correlates with sleep-wake dysfunction after stroke.

Sleep-wake disruption is a key modifiable risk factor and sequela of stroke. The pathogenesis of poststroke sleep dysfunction is unclear. It is not known whether poststroke sleep pathology is due to focal infarction to sleep-wake hubs or to accelerated poststroke neurodegeneration in subcortical structures after stroke. We characterize the first prospective poststroke regional brain volumetric and whole-brain, fiber-specific, white matter markers of objectively measured sleep-wake dysfunction. We hypothesized that excessively long sleep (>8 h) duration and poor sleep efficiency (<80%) measured using the SenseWear Armband 3-months poststroke (n = 112) would be associated with reduced regional brain volumes of a priori-selected sleep-wake regions of interest when compared to healthy controls with optimal sleep characteristics (n = 35). We utilized a novel technique known as a whole-brain fixel-based analysis to investigate the fiber-specific white matter differences in participants with long sleep duration. Stroke participants with long sleep (n = 24) duration exhibited reduced regional volumes of the ipsilesional thalamus and contralesional amygdala when compared with controls. Poor sleep efficiency after stroke (n = 29) was associated with reduced ipsilesional thalamus, contralesional hippocampus, and contralesional amygdala volumes. Whole-brain fixel-based analyses revealed widespread macrostructural degeneration to the corticopontocerebellar tract in stroke participants with long sleep duration, with fiber reductions of up to 40%. Neurodegeneration to subcortical structures, which appear to be vulnerable to accelerated brain volume loss after stroke, may drive sleep-wake deficiencies poststroke, independent of lesion characteristics and confounding comorbidities. We discuss these findings in the context of the clinicopathological implications of sleep-related neurodegeneration and attempt to corroborate previous mechanistic-neuroanatomical findings.