Background: Radiofrequency catheter ablation (RFCA) is an effective therapy for atrial fibrillation (AF). However, it the problem of AF recurrence remains. This study investigates whether a deep convolutional neural network (CNN) can accurately predict AF recurrence in patients with AF who underwent RFCA, and compares CNN with conventional statistical analysis.Methods and Results: Three-hundred and ten patients with AF after RFCA treatment, including 94 patients with AF recurrence, were enrolled. Nine variables are identified as candidate predictors by univariate Cox proportional hazards regression (CPH). A CNNSurv model for AF recurrence prediction was proposed. The model's discrimination ability is validated by a 10-fold cross validation method and measured by C-index. After back elimination, 4 predictors are used for model development, they are N-terminal pro-BNP (NT-proBNP), paroxysmal AF (PAF), left atrial appendage volume (LAAV) and left atrial volume (LAV). The average testing C-index is 0.76 (0.72-0.79). The corresponding calibration plot appears to fit well to a diagonal, and the P value of the Hosmer-Lemeshow test also indicates the proposed model has good calibration ability. The proposed model has superior performance compared with the DeepSurv and multivariate CPH. The result of risk stratification indicates that patients with non-PAF, higher NT-proBNP, larger LAAV and LAV would have higher risks of AF recurrence. Conclusions:The proposed CNNSurv model has better performance than conventional statistical analysis, which may provide valuable guidance for clinical practice.
Identifying patient prognostic phenotypes facilitates precision medicine. This study aimed to explore phenotypes of patients with heart failure (HF) corresponding to prognostic condition (risk of mortality) and identify the phenotype of new patients by machine learning (ML). A unsupervised ML was applied to explore phenotypes of patients in a derivation dataset (n = 562) based on their medical records. Thereafter, supervised ML models were trained on the derivation dataset to classify these identified phenotypes. Then, the trained classifiers were further validated on an independent validation dataset (n = 168). Finally, Shapley additive explanations were used to interpret decision making of phenotype classification. Three patient phenotypes corresponding to stratified mortality risk (high, low, and intermediate) were identified. Kaplan–Meier survival curves among the three phenotypes had significant difference (pairwise comparison p < 0.05). Hazard ratio of all-cause mortality between patients in phenotype 1 (n = 91; high risk) and phenotype 3 (n = 329; intermediate risk) was 2.08 (95%CI 1.29–3.37, p = 0.003), and 0.26 (95%CI 0.11–0.61, p = 0.002) between phenotype 2 (n = 142; low risk) and phenotype 3. For phenotypes classification by random forest, AUCs of phenotypes 1, 2, and 3 were 0.736 ± 0.038, 0.815 ± 0.035, and 0.721 ± 0.03, respectively, slightly better than the decision tree. Then, the classifier effectively identified the phenotypes for new patients in the validation dataset with significant difference on survival curves and hazard ratios. Finally, age and creatinine clearance rate were identified as the top two most important predictors. ML could effectively identify patient prognostic phenotypes, facilitating reasonable management and treatment considering prognostic condition.
A 70-years-old male with a history of hypertension and drug resistant paroxysmal atrial fibrillation (AF) presented to our hospital for catheter ablation to his symptomatic AF. He had no prior surgical or percutaneous procedure to close or exclude the left atrial appendage (LAA). A transesophageal echocardiography (TEE) was performed to rule out intra-cardiac thrombus prior to the ablation procedure. Although the TEE imaging at multiple acquisition angles was obtained, the LAA could not be visualized and an absence of the LAA was suspected. An absence of the LAA was confirmed using cardiac computed tomography (CT), which included 3D reconstruction. Additionally, the LAA was not visualized with left atrium (LA) angiography. During the ablation procedure, 3D voltage mapping in LA was created and no low voltage area or abnormal potential was recorded around the usual root location of the LAA. Successful electrical pulmonary vein isolation was achieved with no major complications. After six months of follow-up, the patient remained in sinus rhythm without any antiarrhythmic drugs and showed no related clinical symptoms. He stopped his anticoagulation therapy due to lack of evidence of AF recurrence and an absence of LAA. Multimodality imaging allowed us to identify the congenital absence of LAA.
Heart failure (HF) is challenging public medical and healthcare systems. This study aimed to develop and validate a novel deep learning-based prognostic model to predict the risk of all-cause mortality for patients with HF. We also compared the performance of the proposed model with those of classical deep learning- and traditional statistical-based models. The present study enrolled 730 patients with HF hospitalized at Toho University Ohashi Medical Center between April 2016 and March 2020. A recurrent neural network-based model (RNNSurv) involving time-varying covariates was developed and validated. The proposed RNNSurv showed better prediction performance than those of a deep feed-forward neural network-based model (referred as “DeepSurv”) and a multivariate Cox proportional hazard model in view of discrimination (C-index: 0.839 vs. 0.755 vs. 0.762, respectively), calibration (better fit with a 45-degree line), and ability of risk stratification, especially identifying patients with high risk of mortality. The proposed RNNSurv demonstrated an improved prediction performance in consideration of temporal information from time-varying covariates that could assist clinical decision-making. Additionally, this study found that significant risk and protective factors of mortality were specific to risk levels, highlighting the demand for an individual-specific clinical strategy instead of a uniform one for all patients.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.