Machine learning of microvolt-level 12-lead electrocardiogram can help distinguish takotsubo syndrome and acute anterior myocardial infarction

Figures

• 

  Figure 1. Explanation of measurement on 1 beat of the electrocardiogram (ECG). All the parameters were measured automatically. Left figure shows schema of the measurement, and right figure a real ECG wave and real results. The ST level was measured at 3 points: (1) ST level at the J point (STJ), which was recorded at the end of the QRS complex as measured in μV with respect to the baseline; (2) the middle of the ST level (STmid), which is the ST level at the point of 1/16th of the preceding R-R interval after the J point; and (3) the end of the ST level (STend), which is the ST level at the point of 2/16th of the preceding R-R interval after the J point. The T-wave amplitude was defined as the absolute distance from the apex of the T wave to the baseline. TTS = takotsubo syndrome.

• 

  Figure 2. Comparison of ST levels of takotsubo syndrome (TTS) and acute anterior myocardial infarction (Ant-AMI) in prediction data (90 cases). Left figure shows ST level at the J point (STJ), middle figure shows the middle of the ST level (STmid), and right figure shows the end of the ST level (STend). After Mann–Whitney U test, a box-and-whisker plot was drawn in each lead. The box indicates interquartile range and median value, and the whisker corresponds to maximum and minimum value. The red and blue triangles show significant upper and lower values, respectively.

• 

  Figure 3. Representative 12-lead electrocardiograms. Left figure demonstrates a takotsubo syndrome (TTS) case, in which ST repression is observed in aVR and V1. Right figure displays an ST-elevation acute anterior myocardial infarction (Ant-AMI) case, in which ST elevation is found in aVR and V1–V3. The patient’s coronary arteriography showed occlusion of the left anterior descending branch (segment 6).

• 

  Figure 4. Interpretation of feature importance by SHapley Additive exPlanation (SHAP) method on 2 representative machine learning models. Left side: Plot of light gradient boosting machine; right side: plot of extra trees classifier. Each point on the summary plot corresponds to a SHAP value for a feature and an instance. Each red and blue point shows a case with takotsubo syndrome and acute anterior myocardial infarction, respectively. On the y-axis, features are sorted based on their importance; color shows the feature value from low (blue) to high (red). The SHAP value is displayed on the x-axis, wherein left side (minus value) shows negative impact and right side (plus value) shows positive impact. Abbreviations as in Figure 1 and Table 2. Brief explanation of SHAP method is demonstrated in Supplemental File 2.