Long-Term Continuous Monitoring of Signal Quality and LVET: Analysis of 24-Hour Period Variations
Authors
Balta, Emma ; Petersen, Frederikke Thaarup
Term
4. term
Education
Publication year
2025
Pages
10
Abstract
Background: Heart failure is a major public health challenge and a leading cause of hospitalization, especially in older adults. Telemedicine and wearable devices can help by enabling home monitoring. However, little is known about the feasibility and signal quality of long-term measurements outside controlled laboratory settings, particularly for seismocardiography (SCG), which captures tiny chest vibrations caused by each heartbeat. This study evaluates SCG signal quality and variations in left ventricular ejection time (LVET) recorded with the eMech wearable sensor in real-world conditions. Methods: Five healthy adults wore the eMech sensor during daily activities across a full 24-hour period, aiming for about 150 signals per participant. We manually rated signal quality and annotated LVET (the time the heart’s left ventricle spends pushing blood out with each beat). We compared signal quality and LVET across day, evening, and night using statistical tests. Results: SCG signal quality varied by time of day. The largest share of good-quality signals occurred at night, while daytime and evening produced more medium- and poor-quality signals. LVET was significantly longer at night (306.8 ms) than in the evening (269.5 ms). No significant differences were found between day and evening or between day and night. Conclusion: SCG can be reliably captured by a wearable sensor in real-world, non-controlled environments. Nighttime recordings tended to be of the highest quality, and LVET showed circadian differences. Analyses of heart measurements from wearables should account for time of day.
Baggrund: Hjertesvigt er en stor folkesundhedsudfordring og en hyppig årsag til indlæggelse, især hos ældre. Telemedicin og bærbare enheder kan aflaste sundhedsvæsenet ved at muliggøre hjemmemonitorering. Der mangler dog viden om, hvor godt langvarige målinger uden for kontrollerede laboratorier fungerer, især for seismokardiografi (SCG), som registrerer bittesmå brystvibrationer forårsaget af hvert hjerteslag. Dette studie vurderer SCG-signalernes kvalitet og variationer i venstre ventrikels uddrivningstid (LVET) registreret med eMech-sensoren i hverdagsmiljøer. Metoder: Fem raske voksne bar eMech-sensoren under daglige aktiviteter over et helt døgn, med mål om ca. 150 signaler per deltager. Vi vurderede manuelt signalkvaliteten og markerede LVET (tiden hjertets venstre pumpekammer bruger på at sende blod ud ved hvert slag). Vi sammenlignede signalkvalitet og LVET mellem dag, aften og nat ved hjælp af statistiske tests. Resultater: Signalkvaliteten varierede med tidspunktet på døgnet. Den største andel af signaler af høj kvalitet blev optaget om natten, mens dag og aften oftere gav signaler af middel eller lav kvalitet. LVET var signifikant længere om natten (306,8 ms) end om aftenen (269,5 ms). Der blev ikke fundet signifikante forskelle mellem dag og aften eller mellem dag og nat. Konklusion: SCG kan registreres pålideligt med en bærbar sensor i virkelige, ukontrollerede omgivelser. Nattens optagelser havde oftest den bedste kvalitet, og LVET viste døgnmæssige forskelle. Analyser af hjerteparametre fra bærbare enheder bør tage højde for tidspunktet på døgnet.
[This apstract has been rewritten with the help of AI based on the project's original abstract]
Keywords