Detection of Cardiovascular Abnormalities through 5-lead System Algorithm
Conference Paper
Overview
Research
Identity
Additional Document Info
Other
View All
Overview
abstract
2016 IEEE. Almost 30% of worldwide death cases are caused by cardiovascular diseases and this number is expected to increase. This paper aims to reduce this high-rate by developing a detection methodology to be used in finding cardiovascular abnormalities. Although researches have been focused on 12-lead systems, this paper narrows the detection down to 5-leads as a faster but still reliable procedure. The system utilizes two algorithms developed by the team. The first algorithm detects the critical points on the Electrocardiogram (ECG) waveforms such as P waves, QRS complex, T-waves and ST elevations. Although detection of typical QRS waveforms has been widely studied, detection of atypical waveforms with complex morphologies remains challenging. The second algorithm detects possible Myocardial Infractions (MI) based on the analysis of the aforementioned critical points. It is developed by mapping clinical definitions of different types of MI and their differential diagnosis into corresponding algorithmic rule sets. Essential pre-processing steps such as baseline correction, removal of ectopic beats, and median filtering are carried out on recorded ECG prior to its classification by the system. Techniques such as multi-stage polynomial correction and QRS subtraction are exploited to achieve reliable pre-processing. These two algorithms can be applied for both 12-lead and 5-lead ECG systems. Our current research targets 5-lead ECG system to make the process easier for the user and faster to implement. Results have shown 77.5% accuracy in the detection of abnormal ECG signals. Future work includes compiling the 5-lead algorithm with a phone application and webserver to make usable by all patients.
name of conference
2016 IEEE-EMBS International Conference on Biomedical and Health Informatics (BHI)
