Assessing Death Risk of Patients with Cardiovascular Disease from Long-Term Electrocardiogram Streams Summarization

摘要

Cardiovascular disease (CVD) is the leading cause of death around the world. Researches on assessing patients death risk from Elec-trocardiographic (ECG) data has attracted increasing attention recently. In this paper, we summarize long-term overwhelming ECG data using morphological concern of overall evolution. And then assessing patients death risk from high value density ECG summarization instead of raw data. Our method is totally unsupervised without the help of expert knowledge. Moreover, it can assist in clinical practice without any additional burden like buy new devices or add more caregivers. Comprehensive results show effectiveness of our method.