Direct mathematical method for real-time ischemic episodes detection from electrocardiograms using the discrete...

Area of Science:

• Biomedical Engineering
• Signal Processing
• Cardiology

Background:

• Ischemic episodes in long-term electrocardiographic (ECG) signals pose diagnostic challenges.
• Automated detection methods are crucial for timely intervention and patient monitoring.

Purpose of the Study:

• To develop and evaluate a real-time automated technique for detecting ischemic episodes using the Discrete Dilated Hermite Transform (DHmT).
• To assess the efficacy of DHmT in identifying characteristic ECG waveform changes associated with ischemia.

Main Methods:

• Utilized mathematical expansions with the Discrete Dilated Hermite Transform for ECG signal analysis.
• Computed DHmT values via dot product between ECG complexes and discrete Hermite functions.
• Employed the European Society of Cardiology (ESC) ST-T database for method validation.

Main Results:

• The DHmT values effectively captured ECG shape information, highlighting ST segment and T wave alterations indicative of ischemia.
• Achieved high performance metrics: 87% sensitivity, 86% specificity, and 81% positive predictive accuracy.
• Demonstrated rapid analysis time of 0.031 seconds per heartbeat on a standard PC.

Conclusions:

• The developed real-time automated technique based on DHmT is effective for identifying ischemic episodes from ECG signals.
• The method offers a computationally efficient and accurate approach for cardiac ischemia detection.
• This technique holds promise for improving long-term ECG monitoring and diagnosis of ischemic events.