紀要論文

長時間心電図に対する効率的 T 波解析手法の開発と 心臓突然死リスク評価への応用

jpn

21879923

https://doi.org/10.15002/00014203

法政大学大学院理工学・工学研究科

58

1

5

2017-03-31

Version of Record

ヘッダー部分に誤記 ; (誤) 法政大学大学院理工学・工学研究科紀要 (正) 法政大学大学院紀要. 理工学・工学研究科編

This paper proposes a method of efficient assessment of the cardiac risk based on the long term Holter ECG recordings. In order to process a vast amount of clinical data a simplified yet effective method to characterize T-wave morphology changes has been proposed. To eliminate artifact, the beat to beat adaptive artifact detection and rectification methods have been introduced. The method is highly efficient and accurate to be able to substitute the manual artifact detection. Twave alternans, T-wave variability and related indices are suggested to be utilized for the integrated cardiac risk assessment. Automatically derived percentile ordered TWA and TAV values TWAP and TAVP are used for the risk assessment. The method has been applied to the Holter ECG data taken from the 351 outpatients visited Fujita Health University Hospital. It has been shown that TAVP could differentiate patients with ejection fraction < 40. The result agreed with data analysis made by manual artifact rectification by medical doctor. The method will be utilized for the future ubiquitous health care to analyze ECG data on the cloud realized by the life-long ECG data logging by IoT technology. Keywords: T-wave alternans, T-wave amplitude variability, Automatic artifact rectification, Holter ECG, Cardiac risk assessment

Article

AA12677220