Heart Rate Variability Analysis of Ischemic and Heart Rate Related ST-segment Deviation Episodes Based on Time-frequency Method

ST-segment deviation is the routine method for the diagnosis of coronary heart diseases. However, other phenomena, such as heart rate changes and a posture change can cause similar manifestations in the ST segment, lowing the sensitivity and specificity of the detection. In this study, a different method, based on time-frequency analysis of heart rate variability, was proposed to evaluate the possibility of detecting the existence of ST-segment deviation episodes and the classification of them. From 43 free download ECG in Long-term ST-T (LTST) database, 87 ST-segment deviation episodes (61 ischemic and 26 heart rate related) were selected with the duration of 36 s to 3 min 28 s. RR intervals were got and preprocessed for heart rate variability (HRV) analysis based on SPVWD. The time-frequency analysis parameters were calculated in different periods of time (before ST-segment deviation, during ST-segment deviation and after ST-segment deviation) and different frequency bands (VLF, LF, HF and AF). Fisher linear discriminant analysis were used to find out sensitive parameters for the detection of ST-segment deviation episodes and the classification of ischemic and heart rate related episodes among the detected episodes. Jackknife method was employed for the accuracy test. Fisher linear discriminant analysis revealed that 8 parameters were sensitive for the detection of the ST-segment deviation episodes with the accuracy of 89.7%, and 6 parameters were sensitive for the classification of ischemic and heart rate related episodes with the accuracy of 86.2%. The proposed method was promising for both detection of ST-segment deviation episodes and the classification of them.