Predicting Intracardiac Thrombus Formation in the Left Atrial Appendage Using Machine Learning and CT Images.

This paper proposes a comprehensive method for estimating thrombus formation factors in the left atrial appendage (LAA). First, using 3D CT (Computer Tomography) image data as input, classification of thrombus presence/absence is learned using 3D ResNet. Besides, 3D Grad-CAM is applied to the prediction results to visualize regions of interest in thrombus formation. Second, features are extracted based on the visualization of regions of interest. Using the extracted features and numerical data obtained from the hospital as input, a regression analysis is performed to predict the presence/absence of thrombus using LightGBM. Visualization of regions of interest using 3D ResNet and 3D Grad-CAM shows that the right inferior pulmonary vein and the LAA were particularly correlated with thrombus formation. Estimation of important factors for thrombus formation using LightGBM shows that the LAA ostium area has the greatest influence on thrombus formation.Clinical Relevance-This paper shows the factors that contribute to thrombus formation in the LAA from the viewpoint of three-dimensional structure. In addition, the features considered important in thrombus formation were identified by comparing a variety of features.

The stability of flow velocity and intracoronary resistance in the intracoronary electrocardiogram-triggered pressure ratio.

Assessment of coronary artery lesions using the fractional flow reserve and instantaneous flow reserve (iFR) measurements has been found to reduce the incidence of further cardiovascular events. Here, we investigated differences in terms of coronary flow velocity and resistance within the analysis interval between the iFR and the intracoronary electrocardiogram (IC-ECG)-triggered distal/aortic pressure (Pd/Pa) ratio (ICE-T). We enrolled 23 consecutive patients (n = 33 stenoses) who required coronary flow measurements. ICE-T was defined as the average Pd/Pa ratio in the period corresponding to the isoelectric line of the IC-ECG. We compared the index value, flow velocity, and intracoronary resistance during the analysis intervals of the iFR and the ICE-T, both at rest and under hyperemia. ICE-T values and ICE-T intracoronary resistance were both found to be significantly lower, whereas flow velocity was significantly higher than those of the iFR at both rest and under hyperemia (P < 0.001), and all fluctuations in ICE-T values were also significantly smaller than those in the iFR. In conclusion, the ICE-T appears theoretically superior to pressure-dependent indices for analyzing phases with low and stable resistance, without an increase in invasiveness.

CORRIGENDUM: Diagnostic Performance and Pressure Stability of a Novel Myocardial Ischemic Diagnostic Index　- The Intracoronary-Electrocardiogram-Triggered Distal Pressure/Aortic Pressure Ratio.

[This corrects the article DOI: 10.1253/circrep.CR-20-0099.].

Classification of Aortic Stenosis Using ECG by Deep Learning and its Analysis Using Grad-CAM.

This paper proposes an automatic method for classifying Aortic valvular stenosis (AS) using ECG (Electrocardiogram) images by the deep learning whose training ECG images are annotated by the diagnoses given by the medical doctor who observes the echocardiograms. Besides, it explores the relationship between the trained deep learning network and its determinations, using the Grad-CAM.In this study, one-beat ECG images for 12-leads and 4-leads are generated from ECG's and train CNN's (Convolutional neural network). By applying the Grad-CAM to the trained CNN's, feature areas are detected in the early time range of the one-beat ECG image. Also, by limiting the time range of the ECG image to that of the feature area, the CNN for the 4-lead achieves the best classification performance, which is close to expert medical doctors' diagnoses.Clinical Relevance-This paper achieves as high AS classification performance as medical doctors' diagnoses based on echocardiograms by proposing an automatic method for detecting AS only using ECG.

Diagnostic Performance and Pressure Stability of a Novel Myocardial Ischemic Diagnostic Index　- The Intracoronary-Electrocardiogram-Triggered Distal Pressure/Aortic Pressure Ratio.

Background: We hypothesized that the intracoronary-electrocardiogram (IC-ECG)-based pressure index would be more stable and precise than the instantaneous flow reserve (iFR). We investigated the usefulness of the IC-ECG-based pressure index for diagnosing myocardial ischemia. Methods and Results: Thirty-seven consecutive patients with coronary stenosis requiring physiological assessment were enrolled in the study. iFR was measured at rest and under hyperemia in 51 and 40 lesions, respectively. The IC-ECG-triggered distal pressure (Pd)/aortic pressure (Pa) ratio (ICE-T) was defined as the mean Pd/Pa ratio in the period corresponding to the isoelectric line. The ICE-T was significantly lower than the iFR both at rest and during hyperemia (P<0.00001 for both). Fluctuations in the ICE-T pressure parameters (Pd/Pa, Pa, and Pd) were significantly smaller than those of iFR both at rest and during hyperemia. The diagnostic accuracy of predicting a fractional flow reserve (FFR) ≤0.80 of the ICE-T at rest was significantly higher than that of iFR (P=0.008). Receiver operating characteristic curve analyses showed that the ICE-T predicts FFR ≤0.80 more accurately than the iFR (area under curve 0.897 vs. 0.810 for ICE-T and iFR, respectively). Conclusions: We identified the period in the IC-ECG in which resting Pd/Pa was low and constant. The IC-ECG-based algorithm may improve the accuracy of diagnosing myocardial ischemia, without increasing invasiveness, compared with pressure-dependent indices.