Characterizing acute ischemia via ECG: A study of severity, extent, and spatial variability in myocardial infarction and non-infarction groups

Figures

• 

  Fig 1. Schematic illustration of the computation of ι from an ECG marker. This computation in shown here for Tam, in the CR (left) and in the PCIR (right).

• 

  Fig 2. Illustrative example of the marker Tam during CR and PCIR and corresponding relative changes. Tam values in the CR (left) and in the PCIR (middle), along with the corresponding relative changes quantified by |Rι| (right), across all 12 ECG leads for a representative patient with RCA occlusion.

• 

  Fig 3. Temporal evolution of the ECG markers during the first 4 minutes of coronary occlusion. Top: Percentage of patients remaining under coronary occlusion over a 4-minute period. Bottom: Median values of |Rι| across all patients under occlusion, where each patient’s value is derived from the 90th percentile of the 12 analyzed ECG leads. Values are reported every 30 seconds from the onset of occlusion.

• 

  Fig 4. Temporal evolution of the ECG markers during the first 4 minutes of coronary occlusion, stratified by sex. A) Top: Percentage of patients under occlusion over 4 minutes. Bottom: Median |Rι| across patients under occlusion, with each patient’s value computed from the 90th percentile of the 12 analyzed leads. Values are calculated every 30 seconds from the onset of occlusion, shown separately for female (left) and male (right) patients. B) Box plots showing extent and severity of ischemia for female and male patients.

• 

  Fig 5. Temporal evolution of the ECG markers during the first 4 minutes of coronary occlusion, stratified by age. A) Top: Percentage of patients under occlusion over 4 minutes. Bottom: Median |Rι| across patients under occlusion, with each patient’s value computed from the 90th percentile of the 12 analyzed leads. Values are calculated every 30 seconds from the onset of occlusion, shown separately for young (left) and old (right) patients. B) Box plots showing extent and severity of ischemia for young and old patients.

• 

  Fig 6. Temporal evolution of the ECG markers during the first 4 minutes of coronary occlusion, stratified by prior MI. A) Top: Percentage of patients under occlusion over 4 minutes. Bottom: Median |Rι| across patients under occlusion, with each patient’s value computed from the 90th percentile of the 12 analyzed leads. Values are calculated every 30 seconds from the onset of occlusion, shown separately for patients with (left) or without (right) prior MI. B) Box plots showing extent and severity of ischemia for patients with and without prior MI.

• 

  Fig 7. Scatter plots showing the relationship between the extent of ischemia and |Rι| for STdev, Tam, and QRSw. The |Rι| values are calculated as the 90th percentile across the 12 leads for each patient at 30 seconds (left panels) and 150 seconds after the onset of occlusion (right panels). Results are shown for patients with prior MI (red squares) and without prior MI (gray circles). The solid and dotted lines depict the linear regression and its corresponding 95% confidence interval, respectively. Correlation coefficients (ρ) are reported; * denotes statistically significant correlations.

• 

  Fig 8. ECG heartbeats during ischemia progression for a patient without prior MI (left) and a patient with prior MI (right), both with LAD occlusion, shown in leads V3 (top) and I (bottom). Black indicates the beginning of the occlusion, and progressively lighter gray shades indicate ischemia progression, with the lightest gray corresponding to approximately 3.5 minutes from the start of the occlusion.

• 

  Fig 9. Box plots of |Rι| stratified by LAD, RCA, and LCx occlusions. Left: Box plots of |Rι| across all patients, with patient-specific values computed from the 90th percentile of the 12 analyzed leads at 150 seconds after the onset of occlusion, for STdev, Tam and QRSw, stratified by LAD, RCA, and LCx occlusions. * denotes statistically significant differences. Effect size |r| and ηH2 are reported. Right: Lead-wise |Rι| values across all patients at 150 seconds from the onset of occlusion for all 12 analyzed leads, shown for STdev, Tam and QRSw for each occluded artery.

• 

  Fig 10. Box plots illustrating extent and severity of ischemia for patients with occlusions in the LAD, RCA, and LCx coronary arteries. Statistical significance of the comparisons between artery groups is shown for ischemia extent and severity. * denotes statistically significant differences. Effect size |r| and ηH2 are reported.