2nd International ICST Conference on Pervasive Computing Technologies for Healthcare

Research Article

ECG monitoring of cardiac patients at home: experiences with scenarios and signal processing methods

Download447 downloads
  • @INPROCEEDINGS{10.4108/ICST.PERVASIVEHEALTH2008.2575,
        author={Juho Merilahti and Mark van Gils and Tuula Pet\aa{}koski-Hult and Outi Kentt\aa{} and Esko Hyv\aa{}rinen and Jari Hyttinen and Harri Kailanto},
        title={ECG monitoring of cardiac patients at home: experiences with scenarios and signal processing methods},
        proceedings={2nd International ICST Conference on Pervasive Computing Technologies for Healthcare},
        publisher={IEEE},
        proceedings_a={PERVASIVEHEALTH},
        year={2008},
        month={7},
        keywords={ECG signal processing scenario evaluation cardiac patient},
        doi={10.4108/ICST.PERVASIVEHEALTH2008.2575}
    }
    
  • Juho Merilahti
    Mark van Gils
    Tuula Petäkoski-Hult
    Outi Kenttä
    Esko Hyvärinen
    Jari Hyttinen
    Harri Kailanto
    Year: 2008
    ECG monitoring of cardiac patients at home: experiences with scenarios and signal processing methods
    PERVASIVEHEALTH
    ICST
    DOI: 10.4108/ICST.PERVASIVEHEALTH2008.2575
Juho Merilahti1,*, Mark van Gils1,*, Tuula Petäkoski-Hult1,*, Outi Kenttä1,*, Esko Hyvärinen2,*, Jari Hyttinen2,*, Harri Kailanto2,*
  • 1: VTT Technical Research Centre of Finland, Tampere, Finland
  • 2: Tampere University of Technology, Ragnar Granit Institute, Tampere, Finland.
*Contact email: juho.merilahti@vtt.fi, mark.vangils@vtt.fi, tuula.Petakoski-hult@vtt.fi, o-kentta@ti.com, esko.hyvarinen@tut.fi, jari.hyttinen@tut.fi, harri.kailanto@tut.fi

Abstract

Controlled cardiac rehabilitation has shown to be an effective and cost-efficient form of treatment. However, it could be supported by technology. We used a scenario-based method to approach the issue and to consider it from the technical perspective. We also tried out signal processing methods in rejection of artifacts and quantization of ECG features from ECG data collected with two different portable recorders. For example correct R-peak detection rates ranged between 98.2% and 99.9% with 5 different tested R-peak detection routines. Conclusions: comments resulted from the scenario work was educative and simple ECG signal processing results promising.