期刊名称：Lecture Notes in Engineering and Computer Science
印刷版ISSN：2078-0958
电子版ISSN：2078-0966
出版年度：2018
卷号：2233&2234
页码：35-39
出版社：Newswood and International Association of Engineers
摘要：Automatic detection of abnormal
electrocardiogram (ECG) waves is a key issue in the field of
medical engineering. Many sever heart diseases show periodic
abnormal waves in ECG. This provide informative suggestions
for identifying the staging or abnormal site of heart disease.
However, so far, few studies have tackled automatic detection of
periodic abnormal ECG wave. In this paper, we propose a new
method for detecting periodic abnormal waves in ECG. This
method is based on the deep neural network model that learns
wave’s shape and their temporal relevance by combing
AutoEncoder and Long Short-Term Memory (LSTM). In the
experiments, using ECG data of a myocardial infarction
patient, it is shown that our method can identify adequately
interval of abnormal wave, which the existing method was not
able to detect.
