Multimodality Risk Assessment of Patients with Ischemic Heart Disease Using Deep Learning Models Applied to Electrocardiograms and Chest X-rays.

Comprehensive management approaches for patients with ischemic heart disease (IHD) are important aids for prognostication and treatment planning. While single-modality deep neural networks (DNNs) have shown promising performance for detecting cardiac abnormalities, the potential benefits of using DNNs for multimodality risk assessment in patients with IHD have not been reported. The purpose of this study was to investigate the effectiveness of multimodality risk assessment in patients with IHD using a DNN that utilizes 12-lead electrocardiograms (ECGs) and chest X-rays (CXRs), with the prediction of major adverse cardiovascular events (MACEs) being of particular concern.DNN models were applied to detection of left ventricular systolic dysfunction (LVSD) on ECGs and identification of cardiomegaly findings on CXRs. A total of 2107 patients who underwent elective percutaneous coronary intervention were categorized into 4 groups according to the models' outputs: Dual-modality high-risk (n = 105), ECG high-risk (n = 181), CXR high-risk (n = 392), and No-risk (n = 1,429).A total of 342 MACEs were observed. The incidence of a MACE was the highest in the Dual-modality high-risk group (P < 0.001). Multivariate Cox hazards analysis for predicting MACE revealed that the Dual-modality high-risk group had a significantly higher risk of MACE than the No-risk group (hazard ratio (HR): 2.370, P < 0.001), the ECG high-risk group (HR: 1.906, P = 0.010), and the CXR high-risk group (HR: 1.624, P = 0.018), after controlling for confounding factors.The results suggest the usefulness of multimodality risk assessment using DNN models applied to 12-lead ECG and CXR data from patients with IHD.

Deep Learning Models for Predicting Left Heart Abnormalities From Single-Lead Electrocardiogram for the Development of Wearable Devices.

BACKGROUND: Left heart abnormalities are risk factors for heart failure. However, echocardiography is not always available. Electrocardiograms (ECGs), which are now available from wearable devices, have the potential to detect these abnormalities. Nevertheless, whether a model can detect left heart abnormalities from single Lead I ECG data remains unclear.Methods and Results: We developed Lead I ECG models to detect low ejection fraction (EF), wall motion abnormality, left ventricular hypertrophy (LVH), left ventricular dilatation, and left atrial dilatation. We used a dataset comprising 229,439 paired sets of ECG and echocardiography data from 8 facilities, and validated the model using external verification with data from 2 facilities. The area under the receiver operating characteristic curves of our model was 0.913 for low EF, 0.832 for wall motion abnormality, 0.797 for LVH, 0.838 for left ventricular dilatation, and 0.802 for left atrial dilatation. In interpretation tests with 12 cardiologists, the accuracy of the model was 78.3% for low EF and 68.3% for LVH. Compared with cardiologists who read the 12-lead ECGs, the model's performance was superior for LVH and similar for low EF. CONCLUSIONS: From a multicenter study dataset, we developed models to predict left heart abnormalities using Lead I on the ECG. The Lead I ECG models show superior or equivalent performance to cardiologists using 12-lead ECGs.

An explainable artificial intelligence-enabled electrocardiogram analysis model for the classification of reduced left ventricular function.

Aims: The black box nature of artificial intelligence (AI) hinders the development of interpretable AI models that are applicable in clinical practice. We aimed to develop an AI model for classifying patients of reduced left ventricular ejection fraction (LVEF) from 12-lead electrocardiograms (ECG) with the decision-interpretability. Methods and results: We acquired paired ECG and echocardiography datasets from the central and co-operative institutions. For the central institution dataset, a random forest model was trained to identify patients with reduced LVEF among 29 907 ECGs. Shapley additive explanations were applied to 7196 ECGs. To extract the model's decision criteria, the calculated Shapley additive explanations values were clustered for 192 non-paced rhythm patients in which reduced LVEF was predicted. Although the extracted criteria were different for each cluster, these criteria generally comprised a combination of six ECG findings: negative T-wave inversion in I/V5-6 leads, low voltage in I/II/V4-6 leads, Q wave in V3-6 leads, ventricular activation time prolongation in I/V5-6 leads, S-wave prolongation in V2-3 leads, and corrected QT interval prolongation. Similarly, for the co-operative institution dataset, the extracted criteria comprised a combination of the same six ECG findings. Furthermore, the accuracy of seven cardiologists' ECG readings improved significantly after watching a video explaining the interpretation of these criteria (before, 62.9% ± 3.9% vs. after, 73.9% ± 2.4%; P = 0.02). Conclusion: We visually interpreted the model's decision criteria to evaluate its validity, thereby developing a model that provided the decision-interpretability required for clinical application.

Deep learning-based age estimation from chest X-rays indicates cardiovascular prognosis.

BACKGROUND: In recent years, there has been considerable research on the use of artificial intelligence to estimate age and disease status from medical images. However, age estimation from chest X-ray (CXR) images has not been well studied and the clinical significance of estimated age has not been fully determined. METHODS: To address this, we trained a deep neural network (DNN) model using more than 100,000 CXRs to estimate the patients' age solely from CXRs. We applied our DNN to CXRs of 1562 consecutive hospitalized heart failure patients, and 3586 patients admitted to the intensive care unit with cardiovascular disease. RESULTS: The DNN's estimated age (X-ray age) showed a strong significant correlation with chronological age on the hold-out test data and independent test data. Elevated X-ray age is associated with worse clinical outcomes (heart failure readmission and all-cause death) for heart failure. Additionally, elevated X-ray age was associated with a worse prognosis in 3586 patients admitted to the intensive care unit with cardiovascular disease. CONCLUSIONS: Our results suggest that X-ray age can serve as a useful indicator of cardiovascular abnormalities, which will help clinicians to predict, prevent and manage cardiovascular diseases.

Chest X-ray is one of the most widely used medical imaging tests worldwide to diagnose and manage heart and lung diseases. In this study, we developed a computer-based tool to predict patients' age from chest X-rays. The tool precisely estimated patients' age from chest X-rays. Furthermore, in patients with heart failure and those admitted to the intensive care unit for cardiovascular disease, elevated X-ray age estimated by our tool was associated with poor clinical outcomes, including readmission for heart failure or death from any cause. With further testing, our tool may help clinicians to predict outcomes in patients with heart disease based on a simple chest X-ray.

Age prediction from coronary angiography using a deep neural network: Age as a potential label to extract prognosis-related imaging features.

Coronary angiography (CAG) is still considered the reference standard for coronary artery assessment, especially in the treatment of acute coronary syndrome (ACS). Although aging causes changes in coronary arteries, the age-related imaging features on CAG and their prognostic relevance have not been fully characterized. We hypothesized that a deep neural network (DNN) model could be trained to estimate vascular age only using CAG and that this age prediction from CAG could show significant associations with clinical outcomes of ACS. A DNN was trained to estimate vascular age using ten separate frames from each of 5,923 CAG videos from 572 patients. It was then tested on 1,437 CAG videos from 144 patients. Subsequently, 298 ACS patients who underwent percutaneous coronary intervention (PCI) were analysed to assess whether predicted age by DNN was associated with clinical outcomes. Age predicted as a continuous variable showed mean absolute error of 4 years with R squared of 0.72 (r = 0.856). Among the ACS patients stratified by predicted age from CAG images before PCI, major adverse cardiovascular events (MACE) were more frequently observed in the older vascular age group than in the younger vascular age group (p = 0.017). Furthermore, after controlling for actual age, gender, peak creatine kinase, and history of heart failure, the older vascular age group independently suffered from more MACE (hazard ratio 2.14, 95% CI 1.07 to 4.29, p = 0.032). The vascular age estimated based on CAG imaging by DNN showed high predictive value. The age predicted from CAG images by DNN could have significant associations with clinical outcomes in patients with ACS.

Automatic Detection of Left Ventricular Dilatation and Hypertrophy from Electrocardiograms Using Deep Learning.

Left ventricular dilatation (LVD) and left ventricular hypertrophy (LVH) are risk factors for heart failure, and their detection improves heart failure screening. This study aimed to investigate the ability of deep learning to detect LVD and LVH from a 12-lead electrocardiogram (ECG). Using ECG and echocardiographic data, we developed deep learning and machine learning models to detect LVD and LVH. We also examined conventional ECG criteria for the diagnosis of LVH. We calculated the area under the receiver operating characteristic (AUROC) curve, sensitivity, specificity, and accuracy of each model and compared the performance of the models. We analyzed data for 18,954 patients (mean age (standard deviation): 64.2 (16.5) years, men: 56.7%). For the detection of LVD, the value (95% confidence interval) of the AUROC was 0.810 (0.801-0.819) for the deep learning model, and this was significantly higher than that of the logistic regression and random forest methods (P < 0.001). The AUROCs for the logistic regression and random forest methods (machine learning models) were 0.770 (0.761-0.779) and 0.757 (0.747-0.767), respectively. For the detection of LVH, the AUROC was 0.784 (0.777-0.791) for the deep learning model, and this was significantly higher than that of the logistic regression and random forest methods and conventional ECG criteria (P < 0.001). The AUROCs for the logistic regression and random forest methods were 0.758 (0.751-0.765) and 0.716 (0.708-0.724), respectively. This study suggests that deep learning is a useful method to detect LVD and LVH from 12-lead ECGs.

Achalasia with esophageal intramural hematoma treated by per-oral endoscopic myotomy (POEM).

Esophageal intramural hematoma (EIH) is a condition which occurs as a result of hemorrhage within the esophageal wall including the submucosal layer. However, reports of EIH on achalasia patients are quite limited and per-oral endoscopic myotomy (POEM) for achalasia with EIH has not been reported. This is the first case report that demonstrated a successful treatment of achalasia with EIH by POEM. In achalasia, since there is absence of lower esophageal sphincter relaxation, as barotraumatic pathogenesis, an increase in the intraesophageal pressure may cause EIH. As direct traumatic pathogenesis, the stasis of food may directly injure the esophageal wall resulting in EIH. After confirming the hematoma healed until it became an ulcer, POEM was performed on the posterior axis since the intramural hematoma was located anteriorly. The procedure was completed successfully without any occurrence of adverse events. On 2-months follow-up, improvement in dysphagia was noted, and complete epithelialization of the intramural hematoma region was seen on endoscopic examination. On 1-year follow-up, patient did not have recurrence of dysphagia and intramural hematoma. In summary, we reported a case of achalasia with EIH, which was treated by POEM. POEM procedure may be effective not only for the improvement of dysphagia but also for a better ulcer healing and prevention of intramural hematoma recurrence.

Deep learning model to detect significant aortic regurgitation using electrocardiography.

BACKGROUND: Aortic regurgitation (AR) is a common heart disease, with a relatively high prevalence of 4.9% in the Framingham Heart Study. Because the prevalence increases with advancing age, an upward shift in the age distribution may increase the burden of AR. To provide an effective screening method for AR, we developed a deep learning-based artificial intelligence algorithm for the diagnosis of significant AR using electrocardiography (ECG). METHODS: Our dataset comprised 29,859 paired data of ECG and echocardiography, including 412 AR cases, from January 2015 to December 2019. This dataset was divided into training, validation, and test datasets. We developed a multi-input neural network model, which comprised a two-dimensional convolutional neural network (2D-CNN) using raw ECG data and a fully connected deep neural network (FC-DNN) using ECG features, and compared its performance with the performances of a 2D-CNN model and other machine learning models. In addition, we used gradient-weighted class activation mapping (Grad-CAM) to identify which parts of ECG waveforms had the most effect on algorithm decision making. RESULTS: The area under the receiver operating characteristic curve of the multi-input model (0.802; 95% CI, 0.762-0.837) was significantly greater than that of the 2D-CNN model alone (0.734; 95% CI, 0.679-0.783; p<0.001) and those of other machine learning models. Grad-CAM demonstrated that the multi-input model tended to focus on the QRS complex in leads I and aVL when detecting AR. CONCLUSIONS: The multi-input deep learning model using 12-lead ECG data could detect significant AR with modest predictive value.

Potential value of saline-induced Pd/Pa ratio in patients with coronary artery stenosis.

Background: Fractional flow reserve (FFR) is the current gold standard for identifying myocardial ischemia in individuals with coronary artery stenosis. However, FFR is not penetrated as much worldwide due to time consumption, costs associated with adenosine, FFR-related discomfort, and complications. Resting physiological indexes may be widely accepted alternatives to FFR, while the discrepancies with FFR were found in up to 20% of lesions. The saline-induced Pd/Pa ratio (SPR) is a new simplified option for evaluating coronary stenosis. However, the clinical implication of SPR remains unclear. Objectives: In the present study, we aimed to compare the accuracies of SPR and resting full-cycle ratio (RFR) and to investigate the incremental value of SPR in clinical practice. Methods: In this multicenter prospective study, 112 coronary lesions (105 patients) were evaluated by SPR, RFR, and FFR. Results: The overall median age was 71 years, and 84.8% were men. SPR was correlated more strongly with FFR than with RFR (r = 0.874 vs. 0.713, respectively; p < 0.001). Using FFR < 0.80 as the reference standard variable, the area under the receiver-operating characteristic (ROC) curve for SPR was superior to that of RFR (0.932 vs. 0.840, respectively; p = 0.009). Conclusion: Saline-induced Pd/Pa ratio predicted FFR more accurately than RFR. SPR could be an alternative method for evaluating coronary artery stenosis and further investigation including elucidation of the mechanism of SPR is needed (225 words).

The Effectiveness of a Deep Learning Model to Detect Left Ventricular Systolic Dysfunction from Electrocardiograms.

Deep learning models can be applied to electrocardiograms (ECGs) to detect left ventricular (LV) dysfunction. We hypothesized that applying a deep learning model may improve the diagnostic accuracy of cardiologists in predicting LV dysfunction from ECGs. We acquired 37,103 paired ECG and echocardiography data records of patients who underwent echocardiography between January 2015 and December 2019. We trained a convolutional neural network to identify the data records of patients with LV dysfunction (ejection fraction < 40%) using a dataset of 23,801 ECGs. When tested on an independent set of 7,196 ECGs, we found the area under the receiver operating characteristic curve was 0.945 (95% confidence interval: 0.936-0.954). When 7 cardiologists interpreted 50 randomly selected ECGs from the test dataset of 7,196 ECGs, their accuracy for predicting LV dysfunction was 78.0% ± 6.0%. By referring to the model's output, the cardiologist accuracy improved to 88.0% ± 3.7%, which indicates that model support significantly improved the cardiologist diagnostic accuracy (P = 0.02). A sensitivity map demonstrated that the model focused on the QRS complex when detecting LV dysfunction on ECGs. We developed a deep learning model that can detect LV dysfunction on ECGs with high accuracy. Furthermore, we demonstrated that support from a deep learning model can help cardiologists to identify LV dysfunction on ECGs.

Automatic detection of vessel structure by deep learning using intravascular ultrasound images of the coronary arteries.

Intravascular ultrasound (IVUS) is a diagnostic modality used during percutaneous coronary intervention. However, specialist skills are required to interpret IVUS images. To address this issue, we developed a new artificial intelligence (AI) program that categorizes vessel components, including calcification and stents, seen in IVUS images of complex lesions. When developing our AI using U-Net, IVUS images were taken from patients with angina pectoris and were manually segmented into the following categories: lumen area, medial plus plaque area, calcification, and stent. To evaluate our AI's performance, we calculated the classification accuracy of vessel components in IVUS images of vessels with clinically significantly narrowed lumina (< 4 mm2) and those with severe calcification. Additionally, we assessed the correlation between lumen areas in manually-labeled ground truth images and those in AI-predicted images, the mean intersection over union (IoU) of a test set, and the recall score for detecting stent struts in each IVUS image in which a stent was present in the test set. Among 3738 labeled images, 323 were randomly selected for use as a test set. The remaining 3415 images were used for training. The classification accuracies for vessels with significantly narrowed lumina and those with severe calcification were 0.97 and 0.98, respectively. Additionally, there was a significant correlation in the lumen area between the ground truth images and the predicted images (ρ = 0.97, R2 = 0.97, p < 0.001). However, the mean IoU of the test set was 0.66 and the recall score for detecting stent struts was 0.64. Our AI program accurately classified vessels requiring treatment and vessel components, except for stents in IVUS images of complex lesions. AI may be a powerful tool for assisting in the interpretation of IVUS imaging and could promote the popularization of IVUS-guided percutaneous coronary intervention in a clinical setting.

Deep Learning Algorithm to Detect Cardiac Sarcoidosis From Echocardiographic Movies.

BACKGROUND: Because the early diagnosis of subclinical cardiac sarcoidosis (CS) remains difficult, we developed a deep learning algorithm to distinguish CS patients from healthy subjects using echocardiographic movies.Methods and Results:Among the patients who underwent echocardiography from January 2015 to December 2019, we chose 151 echocardiographic movies from 50 CS patients and 151 from 149 healthy subjects. We trained two 3D convolutional neural networks (3D-CNN) to identify CS patients using a dataset of 212 echocardiographic movies with and without a transfer learning method (Pretrained algorithm and Non-pretrained algorithm). On an independent set of 41 echocardiographic movies, the area under the receiver-operating characteristic curve (AUC) of the Pretrained algorithm was greater than that of Non-pretrained algorithm (0.842, 95% confidence interval (CI): 0.722-0.962 vs. 0.724, 95% CI: 0.566-0.882, P=0.253). The AUC from the interpretation of the same set of 41 echocardiographic movies by 5 cardiologists was not significantly different from that of the Pretrained algorithm (0.855, 95% CI: 0.735-0.975 vs. 0.842, 95% CI: 0.722-0.962, P=0.885). A sensitivity map demonstrated that the Pretrained algorithm focused on the area of the mitral valve. CONCLUSIONS: A 3D-CNN with a transfer learning method may be a promising tool for detecting CS using an echocardiographic movie.

Factors associated with left ventricular reverse remodelling after percutaneous coronary intervention in patients with left ventricular systolic dysfunction.

Percutaneous coronary intervention (PCI) is sometimes considered as an alternative therapeutic strategy to surgical revascularization in patients with coronary artery disease (CAD) and reduced left ventricular ejection fraction (LVEF). However, the types or conditions of patients that receive the clinical benefit of left ventricular reverse remodelling (LVRR) remain unknown. The purpose of this study was to investigate the determinants of LVRR following PCI in CAD patients with reduced LVEF. From 4394 consecutive patients who underwent PCI, a total of 286 patients with reduced LV systolic function (LVEF < 50% at initial left ventriculography) were included in the analysis. LVRR was defined as LV end-systolic volume reduction ≥ 15% and improvement of LVEF ≥ 10% at 6 months follow-up left ventriculography. Patients were divided into LVRR (n = 63) and non-LVRR (n = 223) groups. Multivariate logistic regression analysis revealed that unprotected left main coronary artery (LMCA) intervention was significantly associated with LVRR (P = 0.007, odds ratios [OR] 4.70, 95% confidence interval [CI] 1.54-14.38), while prior PCI (P = 0.001, OR 0.35, 95% CI 0.19-0.66), presence of in-stent restenosis (P = 0.016, OR 0.32, 95% CI 0.12-0.81), and presence of de-novo stenosis (P = 0.038, OR 0.36, 95% CI 0.14-0.95) were negatively associated with LVRR. These data suggest the potential prognostic benefit of unprotected LMCA intervention for LVRR and importance of angiographic follow-up in patients with CAD and LV systolic dysfunction.

High spatial endothelial shear stress gradient independently predicts site of acute coronary plaque rupture and erosion.

AIMS: To investigate local haemodynamics in the setting of acute coronary plaque rupture and erosion. METHODS AND RESULTS: Intracoronary optical coherence tomography performed in 37 patients with acute coronary syndromes caused by plaque rupture (n = 19) or plaque erosion (n = 18) was used for three-dimensional reconstruction and computational fluid dynamics simulation. Endothelial shear stress (ESS), spatial ESS gradient (ESSG), and oscillatory shear index (OSI) were compared between plaque rupture and erosion through mixed-effects logistic regression. Lipid, calcium, macrophages, layered plaque, and cholesterol crystals were also analysed. By multivariable analysis, only high ESSG [odds ratio (OR) 5.29, 95% confidence interval (CI) 2.57-10.89, P < 0.001], lipid (OR 12.98, 95% CI 6.57-25.67, P < 0.001), and layered plaque (OR 3.17, 95% CI 1.82-5.50, P < 0.001) were independently associated with plaque rupture. High ESSG (OR 13.28, 95% CI 6.88-25.64, P < 0.001), ESS (OR 2.70, 95% CI 1.34-5.42, P = 0.005), and OSI (OR 2.18, 95% CI 1.33-3.54, P = 0.002) independently associated with plaque erosion. ESSG was higher at rupture sites than erosion sites [median (interquartile range): 5.78 (2.47-21.15) vs. 2.62 (1.44-6.18) Pa/mm, P = 0.009], OSI was higher at erosion sites than rupture sites [1.04 × 10-2 (2.3 × 10-3-4.74 × 10-2) vs. 1.29 × 10-3 (9.39 × 10-5-3.0 × 10-2), P < 0.001], but ESS was similar (P = 0.29). CONCLUSIONS: High ESSG is independently associated with plaque rupture while high ESSG, ESS, and OSI associate with plaque erosion. While ESSG is higher at rupture sites than erosion sites, OSI is higher at erosion sites and ESS was similar. These results suggest that ESSG and OSI may play critical roles in acute plaque rupture and erosion, respectively.

Comparison of post-stent optical coherence tomography findings: Layered versus non-layered culprit lesions.

OBJECTIVES: This study aimed to investigate the vascular response of lesions with a layered phenotype. BACKGROUND: Recent studies have shown that layered plaques at culprit lesions detected by optical coherence tomography (OCT) have greater plaque burden and more inflammatory features than non-layered plaques. METHODS: This is a retrospective observational study. A total of 193 target lesions from 193 patients [100 patients with acute coronary syndromes (ACS) and 93 with stable angina pectoris (SAP)] who had undergone OCT imaging of the culprit lesion both before and after stenting were included. Layered plaques were identified by OCT as plaques with layers of different optical density. Patients were divided into two groups based on the presence or absence of a layered phenotype at the culprit lesion, and pre- and post-procedure OCT findings were compared. RESULTS: Among 193 patients, 36 (36.0%) lesions in ACS patients and 56 (60.2%) lesions in SAP patients were found to have a layered phenotype at the culprit lesion. At baseline, percent area stenosis was greater in layered plaque than in non-layered plaque (p = .019). Following stent implantation, the stent expansion ratio and mean stent eccentricity index were significantly lower in layered plaques than in non-layered plaques (p = .041, p = .017, respectively), mainly derived from ACS patients. CONCLUSION: Following stent implantation, plaques with a layered phenotype had less stent expansion and more eccentric lumens. Aggressive balloon dilation may be required to obtain optimal stent outcomes in patients with a layered plaque phenotype at the culprit lesion.

Efficacy of Fractional Flow Reserve-Guided Percutaneous Cornary Intervention for Patients with Angina Pectoris.

Evaluation of hemodynamic parameters, such as fractional flow reserve (FFR), is recommended before percutaneous coronary intervention (PCI) for patients with angina pectoris (AP). However, the advantage of FFR-guided PCI has not been fully established. This study was performed to confirm whether FFR-guided PCI improves the prognosis compared with other treatments. Multiple databases were searched for studies published from 2000 to 2018, and a network meta-analysis (NMA) was performed to compare outcomes of FFR-guided PCI, non-FFR-guided PCI, coronary artery bypass grafting (CABG), and medical treatment (MT) for AP based on estimated odds ratios (ORs). The study included 18,093 patients from 15 randomized controlled trials (RCTs). No evidence of inconsistency was observed among the studies. The NMA showed that the all-cause mortality of FFR-guided PCI was not significantly different from that of the other treatment groups (CABG: OR, 1.1; 95% confidence interval [CI], 0.67-1.7; non-FFR-guided PCI: OR, 0.85; 95% CI, 0.53-1.4; and MT: OR, 0.83; 95% CI, 0.52-1.3). The NMA for the composite of all-cause mortality and myocardial infarction, which included 15,454 patients from 12 RCTs, showed that FFR-guided PCI significantly reduced the composite outcome compared with non-FFR-guided PCI and MT (non-FFR-guided PCI: OR, 0.66; 95% CI, 0.46-0.95 and MT: OR, 0.66; 95% CI, 0.46-0.95). Although FFR-guided PCI for AP did not show significant prognostic improvement compared with non-FFR-guided PCI, CABG, and MT, FFR-guided PCI may significantly reduce the composite of all-cause mortality and myocardial infarction compared with non-FFR-guided PCI and MT.

Erratum: Diagnosing Heart Failure from Chest X-Ray Images Using Deep Learning.

An error appeared in the article entitled "Diagnosing Heart Failure from Chest X-Ray Images Using Deep Learning" by Takuya Matsumoto, Satoshi Kodera, Hiroki Shinohara, Hirotaka Ieki, Toshihiro Yamaguchi, Yasutomi Higashikuni, Arihiro Kiyosue, Kaoru Ito, Jiro Ando, Eiki Takimoto, Hiroshi Akazawa, Hiroyuki Morita, Issei Komuro (Vol. 61, No. 4, 781-786, 2020). The Figure 5on page 784 should be replaced by the following figure.

Spatial Distribution of Vulnerable Plaques: Comprehensive In Vivo Coronary Plaque Mapping.

OBJECTIVES: The authors performed a comprehensive analysis on the distribution of coronary plaques with different phenotypes from our 3-vessel optical coherence tomography (OCT) database. BACKGROUND: Previous pathology studies demonstrated that thin-cap fibroatheroma (TCFA) is localized in specific segments of the epicardial coronary arteries. A detailed description of in vivo coronary plaques of various phenotypes has not been reported. METHODS: OCT images of all 3 coronary arteries in 131 patients were analyzed every 1 mm to assess plaque phenotype and features of vulnerability. In addition, plaques were divided into tertiles according to percent area stenosis (%AS). RESULTS: Among 534 plaques identified in 393 coronary arteries, 27.0% were fibrous plaques, 13.3% were fibrocalcific plaques, 40.8% were thick-cap fibroatheromas, and 18.9% were TCFAs. TCFAs showed clustering in the proximal segment, particularly in the left anterior descending artery. On the other hand, fibrous plaques were relatively evenly distributed throughout the entire length of the coronary arteries. In patients with acute coronary syndromes (ACS), TCFAs showed stronger proximal clustering in the left anterior descending, 2 clustering peaks in the right coronary artery, and 1 clustering peak in the circumflex artery. The pattern of TCFA distribution was less obvious in patients without ACS. The prevalence of TCFA was higher in the highest %AS tertile, compared with the lowest %AS tertile (30% vs. 9%; p < 0.001). CONCLUSIONS: The present 3-vessel OCT study showed that TCFAs cluster at specific locations in the epicardial coronary arteries, especially in patients with ACS. TCFA was more prevalent in segments with tight stenosis. (The Massachusetts General Hospital Optical Coherence Tomography Registry; NCT01110538).

Diagnosing Heart Failure from Chest X-Ray Images Using Deep Learning.

The development of deep learning technology has enabled machines to achieve high-level accuracy in interpreting medical images. While many previous studies have examined the detection of pulmonary nodules in chest X-rays using deep learning, the application of this technology to heart failure remains rare. In this paper, we investigated the performance of a deep learning algorithm in terms of diagnosing heart failure using images obtained from chest X-rays. We used 952 chest X-ray images from a labeled database published by the National Institutes of Health. Two cardiologists verified and relabeled a total of 260 "normal" and 378 "heart failure" images, with the remainder being discarded because they had been incorrectly labeled. Data augmentation and transfer learning were used to obtain an accuracy of 82% in diagnosing heart failure using the chest X-ray images. Furthermore, heatmap imaging allowed us to visualize decisions made by the machine. Deep learning can thus help support the diagnosis of heart failure using chest X-ray images.