An experimental BallistoCardioGram (BCG) measurement and analysis system is presented, featuring accelerometers placed between the mattress and bed slats. The system can sample BCG waveforms at 500 Hz and can perform unsupervised BCG heartbeat detection and heartbeat interval measurements. The approach is validated on an experimental dataset consisting in 14 subjects recorded while lying in three different positions: supine, left side and right side. The overall performance is good, comparing to other works in literature. In fact, the average sensitivity and precision across subjects and positions is 98.2% and 98.0%, respectively; similarly, an R2 score of 98.2% was achieved between BCG and reference ECG measurements, while Mean Absolute Error and Root Mean Squared Error are as low as 3.9 ms and 5.6 ms. The presented methodology is shown to be resilient to different sleeping positions, as confirmed by Kruskal-Wallis statistical tests (p≈0.91 for sensitivity, p≈0.73 for precision, p≈0.81 for R2, p≈0.26 for MAE, p≈0.20 for RMSE). Moreover, results are in line and comparable to those already achieved on a different measurement scenario featuring a different bed structure. This further proves the stability of the presented BCG measurement and analysis system.
An accurate and stable bed-based ballistocardiogram measurement and analysis system / Mora, Niccolo; Cocconcelli, Federico; Matrella, Guido; Chiorboli, Giovanni; Ciampolini, Paolo. - (2021), pp. 482-487. (Intervento presentato al convegno 2021 IEEE International Workshop on Metrology for Industry 4.0 & IoT (MetroInd4.0&IoT) tenutosi a Roma, Italia nel 7-9/6/2021) [10.1109/MetroInd4.0IoT51437.2021.9488475].
An accurate and stable bed-based ballistocardiogram measurement and analysis system
Mora, Niccolo
;Cocconcelli, Federico
;Matrella, Guido;Chiorboli, Giovanni;Ciampolini, Paolo
2021-01-01
Abstract
An experimental BallistoCardioGram (BCG) measurement and analysis system is presented, featuring accelerometers placed between the mattress and bed slats. The system can sample BCG waveforms at 500 Hz and can perform unsupervised BCG heartbeat detection and heartbeat interval measurements. The approach is validated on an experimental dataset consisting in 14 subjects recorded while lying in three different positions: supine, left side and right side. The overall performance is good, comparing to other works in literature. In fact, the average sensitivity and precision across subjects and positions is 98.2% and 98.0%, respectively; similarly, an R2 score of 98.2% was achieved between BCG and reference ECG measurements, while Mean Absolute Error and Root Mean Squared Error are as low as 3.9 ms and 5.6 ms. The presented methodology is shown to be resilient to different sleeping positions, as confirmed by Kruskal-Wallis statistical tests (p≈0.91 for sensitivity, p≈0.73 for precision, p≈0.81 for R2, p≈0.26 for MAE, p≈0.20 for RMSE). Moreover, results are in line and comparable to those already achieved on a different measurement scenario featuring a different bed structure. This further proves the stability of the presented BCG measurement and analysis system.File | Dimensione | Formato | |
---|---|---|---|
An_accurate_and_stable_bed-based_ballistocardiogram_measurement_and_analysis_system.pdf
non disponibili
Descrizione: Articolo principale
Tipologia:
Versione (PDF) editoriale
Licenza:
NON PUBBLICO - Accesso privato/ristretto
Dimensione
277.71 kB
Formato
Adobe PDF
|
277.71 kB | Adobe PDF | Visualizza/Apri Richiedi una copia |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.