Forecasting Seizure Likelihood With Wearable Technology

Frontiers in Neurology, Journal Year: 2021, Volume and Issue: 12

Published: July 15, 2021

The unpredictability of epileptic seizures exposes people with epilepsy to potential physical harm, restricts day-to-day activities, and impacts mental well-being. Accurate seizure forecasters would reduce the uncertainty associated but need be feasible accessible in long-term. Wearable devices are perfect candidates develop non-invasive, forecasts yet investigated long-term studies. We hypothesized that machine learning models could utilize heart rate as a biomarker for well-established cycles activity, addition other wearable signals, forecast high low risk periods. This feasibility study tracked participants' ( n = 11) rates, sleep, step counts using smartwatches occurrence smartphone diaries at least 6 months (mean 14.6 months, SD 3.8 months). Eligible participants had diagnosis refractory reported 20 135, 123) during recording period. An ensembled neural network model estimated either daily or hourly, retraining occurring on weekly basis additional data was collected. Performance evaluated retrospectively against rate-matched random area under receiver operating curve. A pseudo-prospective evaluation also conducted held-out dataset. Of 11 participants, were predicted above chance all (100%) an hourly ten (91%) forecast. average time spent (prediction time) before occurred 37 min 3 days Cyclic features added most predictive value forecasts, particularly circadian multiday cycles. can used produce patient-specific when biomarkers activity utilized.

Language: Английский

---

Wearable Devices for Remote Monitoring of Heart Rate and Heart Rate Variability—What We Know and What Is Coming

Sensors, Journal Year: 2022, Volume and Issue: 22(22), P. 8903 - 8903

Published: Nov. 17, 2022

Heart rate at rest and exercise may predict cardiovascular risk. variability is a measure of variation in time between each heartbeat, representing the balance parasympathetic sympathetic nervous system adverse events. With advances technology increasing commercial interest, scope remote monitoring health systems has expanded. In this review, we discuss concepts behind cardiac signal generation recording, wearable devices, pros cons focusing on accuracy, ease application medical grade diagnostic which showed promising results terms reliability value. Incorporation artificial intelligence cloud based have been evolving to facilitate timely data processing, improve patient convenience ensure security.

Language: Английский

---

Evaluation and update of the expert consensus guidelines for the assessment of the cortisol awakening response (CAR)

Psychoneuroendocrinology, Journal Year: 2022, Volume and Issue: 146, P. 105946 - 105946

Published: Sept. 27, 2022

Language: Английский

---

Association Between Slow-Wave Sleep Loss and Incident Dementia

JAMA Neurology, Journal Year: 2023, Volume and Issue: 80(12), P. 1326 - 1326

Published: Oct. 30, 2023

Slow-wave sleep (SWS) supports the aging brain in many ways, including facilitating glymphatic clearance of proteins that aggregate Alzheimer disease. However, role SWS development dementia remains equivocal.

Language: Английский

---

Accuracy of 11 Wearable, Nearable, and Airable Consumer Sleep Trackers: Prospective Multicenter Validation Study

JMIR mhealth and uhealth, Journal Year: 2023, Volume and Issue: 11, P. e50983 - e50983

Published: Sept. 20, 2023

Background Consumer sleep trackers (CSTs) have gained significant popularity because they enable individuals to conveniently monitor and analyze their sleep. However, limited studies comprehensively validated the performance of widely used CSTs. Our study therefore investigated popular CSTs based on various biosignals algorithms by assessing agreement with polysomnography. Objective This aimed validate accuracy types through a comparison in-lab Additionally, including conducting multicenter large sample size, this seeks provide comprehensive insights into applicability these for monitoring in hospital environment. Methods The analyzed 11 commercially available CSTs, 5 wearables (Google Pixel Watch, Galaxy Watch 5, Fitbit Sense 2, Apple 8, Oura Ring 3), 3 nearables (Withings Sleep Tracking Mat, Google Nest Hub Amazon Halo Rise), airables (SleepRoutine, SleepScore, Pillow). were divided 2 groups, ensuring maximum inclusion while avoiding interference between within each group. Each group (comprising 8 CSTs) was also compared via Results enrolled 75 participants from tertiary primary sleep-specialized clinic Korea. Across centers, we collected total 3890 hours sessions along 543 polysomnography recordings. CST recording covered an average 353 hours. We 349,114 epochs polysomnography, where epoch-by-epoch stage classification showed substantial variation. More specifically, highest macro F1 score 0.69, lowest 0.26. Various exhibited diverse performances across stages, SleepRoutine excelling wake rapid eye movement like showing superiority deep stage. There distinct trend measure estimation according type device. Wearables high proportional bias efficiency, latency. Subgroup analyses revealed variations scores factors, such as BMI, apnea-hypopnea index, differences male female subgroups minimal. Conclusions that among examined, specific indicating potential application monitoring, other partially consistent offers strengths different classes interested wellness who wish understand proactively manage own

Language: Английский

---