@article{3068571,
    title = "Remote sensing natural time analysis of heartbeat data by means of a portable photoplethysmography device",
    author = "Baldoumas, G. and Peschos, D. and Tatsis, G. and Christofilakis, V. and Chronopoulos, S.K. and Kostarakis, P. and Varotsos, P.A. and Sarlis, N.V. and Skordas, E.S. and Bechlioulis, A. and Michalis, L.K. and Naka, K.K.",
    journal = "International Journal of Remote Sensing",
    year = "2021",
    volume = "42",
    number = "6",
    pages = "2292-2302",
    publisher = "Taylor and Francis Ltd.",
    issn = "0143-1161, 1366-5901",
    doi = "10.1080/2150704X.2020.1847351",
    keywords = "Electrocardiography;  Electron devices;  Heart;  Millimeter wave devices;  Millimeter waves;  Photoplethysmography;  Radar;  Smartphones;  Support vector machines, Bluetooth communications;  Congestive heart failures;  Direct communications;  Doppler frequency shift;  Millimetre-wave radar;  Photoplethysmography (PPG);  Sudden cardiac deaths;  Support vector machine classifiers, Remote sensing, cardiovascular disease;  complexity;  electronic equipment;  remote sensing;  statistical analysis;  support vector machine;  telecommunication, Varanidae",
    abstract = "Very recent work reported that patients can monitor their heartbeat at home and specify their heart conditions by means of a millimetre wave radar, but there exist serious limitations because the radar sensor is sensitive to significant body motions that cause Doppler frequency shifts. Such limitations do not exist when using a recently constructed portable photoplethysmography (PPG) electronic device, which gives results comparable with a standard electrocardiogram (ECG). Since all portable modern devices such as smart phones tablets etc support Bluetooth communication that allows easy and direct communication with our PPG device, it may give us remote sensing heart related information. Applying natural time analysis to data simultaneously collected with an ECG system and a PPG device and using two complexity measures quantifying the entropy change in natural time under time reversal, a distinction is achieved between healthy (H) individuals and congestive heart failure (CHF) patients. Employing a support vector machine classifier for CHF discrimination to a total of 99 individuals (including 67 CHF), we obtained 97.7% sensitivity. In a follow up study challenging results are obtained since during the subsequent period six individuals died, who remarkably obeyed additional complexity measures that may distinguish sudden cardiac death individuals from CHF. © 2020 Informa UK Limited, trading as Taylor & Francis Group."
}