562015-03-24 , 法政大学大学院理工学・工学研究科
This paper proposes a novel method of detecting the presence of T-wave alternans (TWA) based on 3-channel Holter ECG data. Namely a new index named αpercentile of alternans ratio (AR) values AR(α) is proposed as an efficient and reliable measure for the sudden cardiac death risk assessment. The proposed method intends to realize an accurate and efficient TWA detection on the data recorded under non-controlled common daily activity. The method examines the change in T-wave morphology based on singular value decomposition (SVD). The T-Loop divided into four sections is utilized for calculating the AR ratio defined as the relative frequency power at the Nyquist frequency. Thenαpercentile of ordered AR values AR(α) are compared between subject with high cardiac risk and control subjects. It was shown that the index AR(0.1) effectively classifies the high risk subject group. The method will be useful for the risk assessment of sudden cardiac death in the daily common activity.