学位論文

特異値分解による特徴ベクトルを利用した心電図T波オルタナンス検出手法の提案

jpn

2010-03-24

Not Applicable (or Unknown)

2010-03-24

修士(工学)

工学研究科情報電子工学専攻; 指導教授: 八名和夫

This paper proposes a noble method for detecting T-wave alternans (TWA) based on 3-channel Holter ECG data. The method is intended to be the core of realizing an accurate and efficient TWA detection system based on the data acquired under non-controlled common daily activity. The method utilizes singular value decomposition (SVD) for highly sensitive differentiation of T-wave morphology in noisy recording conditions． SVD has been adopted for detecting the precise T-wave morphology change in the noisy Holter recording condition. We call T-wave morphology change found by the SVD, T-wave Vector lternans (TWVA), which is the key notion in the proposed method. The method has been applied to ten normal subjects and ten subjects with TWA to confirm its validity．The method will be a useful tool to extract an extra important diagnostic index from the popular Holter ECG recordings.

Thesis