Open AccessMethod to Determine the Applicability of Multi-Parameter Fusion Analysis
Author(s) -
Xianliang He,
Sanchao Liu,
Zehui Sun,
Wenyu Ye,
Jianwei Su,
Haoyu Jiang,
Ping Yang,
Cadathur Rajagopalan
Publication year - 2020
Publication title -
2019 computing in cardiology (cinc)
Language(s) - English
Resource type - Conference proceedings
SCImago Journal Rank - 0.257
H-Index - 55
ISSN - 2325-887X
ISBN - 978-1-7281-6936-1
DOI - 10.22489/cinc.2019.066
Subject(s) - bioengineering , computing and processing , signal processing and analysis
Alarm fatigue has long been a clinical problem. Multi-parameter Fusion Analysis (MPA) is a technique often used to address this. However, it cannot be used in some cases such as in severe peripheral circulation disorder, cardiopulmonary resuscitation (CPR) etc.. In these cases, critical alarms may be erroneously suppressed if MPA is applied increasing clinical risk for subjects.Electrocardiogram (ECG) and photoplethesmogram (PPG) signals are normally synchronous. Based on this, our method evaluates the applicability of MPA as follows: Multiple physiological signals are first analyzed by highly reliable single-parameter analysis methods to obtain all the parameters from these as well as signal quality indices. Then, signal features of different channels for each cardiac cycle are grouped. Finally, group features are derived and used with an integrated decision tree to determine the applicability of MPA. Three databases were used to test algorithm performance. DB1 containing 200 clinical waveform records; DB2 with 100 clinical waveform records; DB3 comprising 30 cases recorded in cardiopulmonary resuscitation, each case spanning 2 hours. The recognition rates for the three databases were all greater than 90% indicating that the proposed method can effectively decide whether the considered physiological signals can be used for MPA.