Software defect prediction using hybrid model (CBIL) of convolutional neural network (CNN) and bidirectional long short-term memory (Bi-LSTM).

In recent years, the software industry has invested substantial effort to improve software quality in organizations. Applying proactive software defect prediction will help developers and white box testers to find the defects earlier, and this will reduce the time and effort. Traditional software defect prediction models concentrate on traditional features of source code including code complexity, lines of code, etc. However, these features fail to extract the semantics of source code. In this research, we propose a hybrid model that is called CBIL. CBIL can predict the defective areas of source code. It extracts Abstract Syntax Tree (AST) tokens as vectors from source code. Mapping and word embedding turn integer vectors into dense vectors. Then, Convolutional Neural Network (CNN) extracts the semantics of AST tokens. After that, Bidirectional Long Short-Term Memory (Bi-LSTM) keeps key features and ignores other features in order to enhance the accuracy of software defect prediction. The proposed model CBIL is evaluated on a sample of seven open-source Java projects of the PROMISE dataset. CBIL is evaluated by applying the following evaluation metrics: F-measure and area under the curve (AUC). The results display that CBIL model improves the average of F-measure by 25% compared to CNN, as CNN accomplishes the top performance among the selected baseline models. In average of AUC, CBIL model improves AUC by 18% compared to Recurrent Neural Network (RNN), as RNN accomplishes the top performance among the selected baseline models used in the experiments.

Prevalence and Clinical Significance of Latent Brugada Syndrome in Atrial Fibrillation Patients Below 45 Years of Age.

Aim: This study aims to describe prevalence and clinical significance of latent Brugada syndrome (BrS) in a young population with atrial fibrillation (AF). Methods: Between September 2015 and November 2017, among 111 AF patients below 45 years of age, those without pre-existing pathologies and/or known risk factors were selected for the study. Based on baseline 12-lead-24-h Holter electrocardiogram (ECG), previous class 1C antiarrhythmic drug therapy, or ajmaline testing, patients were stratified as latent type 1 BrS or not. Results: Within the 78 enrolled patients, 13 (16.7%; group 1) revealed a type 1 BrS ECG pattern, while 65 (83.3%; group 2) did not. Mean age was 37 ± 8 vs. 35 ± 7 (p = 0.42), and males were 7 (54%) vs. 54 (83%) (p = 0.02) in the two groups, respectively. Family history of BrS was significantly more common within group 1 patients (2, 15% vs. 0; p = 0.03), and 4 (31%) patients experienced syncope in group 1 vs. 5 (8%) in group 2 (p = 0.02). After a mean follow-up of 42 ± 18 months from the index AF event, more than 80% of the patients, in both study groups, were in sinus rhythm. Conclusion: In young patients with AF without pre-existing pathologies and/or known risk factors, latent BrS should be suspected. Syncope and a family history of BrS emerge as easily identifiable factors related to BrS. Long-term sinus rhythm maintenance appears satisfactory, either in the presence or not of BrS.