Logistic regression (LR) and artificial intelligence algorithms were used to analyze the risk factors for the early rupture of acute type A aortic dissection (ATAAD). Data from electronic medical records of 200 patients diagnosed with ATAAD from the Department of Emergency of Guangdong Provincial People’s Hospital from April 2012 to March 2017 were collected. Logistic regression and artificial intelligence algorithms were used to establish prediction models, and the prediction effects of four models were analyzed. According to the LR models, we elucidated independent risk factors for ATAAD rupture, which included age > 63 years (odds ratio (OR) = 1.69), female sex (OR = 1.77), ventilator assisted ventilation (OR = 3.05), AST > 80 U/L (OR = 1.59), no distortion of the inner membrane (OR = 1.57), the diameter of the aortic sinus > 41 mm (OR = 0.92), maximum aortic diameter > 48 mm (OR = 1.32), the ratio of false lumen area to true lumen area > 2.12 (OR = 1.94), lactates > 1.9 mmol/L (OR = 2.28), and white blood cell > 14.2 × 109 /L (OR = 1.23). The highest sensitivity and accuracy were found with the convolutional neural network (CNN) model. Its sensitivity was 0.93, specificity was 0.90, and accuracy was 0.90. In this present study, we found that age, sex, select biomarkers, and select morphological parameters of the aorta are independent predictors for the rupture of ATAAD. In terms of predicting the risk of ATAAD, the performance of random forests and CNN is significantly better than LR, but the performance of the support vector machine (SVM) is worse than LR.
19 May 2017 in Concurrency and Computation: Practice and Experience
Wafer defect detection is critical to semiconductor manufacturing, where precise identification of defect types is a key to optimizing process and enhancing efficiency. To address challenges such as low resolution, complex defect distributions, and high-density in wafer images, this paper proposes a novel density-guided dual-stream vision transformer model (DG-ViT). DG-ViT innovatively integrates a Semantic-guided Global transformer (SGGT) and a density-aware graph convolutional network (DAGCN). SGGT captures long-range semantic correlations of defect through global semantic attention, while DAGCN deeply characterizes local structural features via density-driven topological modeling. An adaptive density-aware feature masking leverages adaptive density heatmaps to filter noise, significantly enhancing the clarity of defect representations. Together, these components empower DG-ViT with profound analytical capabilities for complex defect patterns. Experiments on two large datasets demonstrate that DG-ViT outperforms mainstream models, markedly improving defect classification accuracy and robustness in complex and low-resolution scenarios, achieving accuracy of 94.81% on WM-811 K and 98.54% on Mixed-WM38.
Evolutionary algorithms have been widely applied for solving multi-objective optimization problems, while the feature selection in classification can also be treated as a discrete bi-objective optimization problem if attempting to minimize both the classification error and the ratio of selected features. However, traditional multi-objective evolutionary algorithms (MOEAs) may have drawbacks for tackling large-scale feature selection, due to the curse of dimensionality in the decision space. Therefore, in this paper, we concentrated on designing an multi-task decomposition-based evolutionary algorithm (abbreviated as MTDEA), especially for handling high-dimensional bi-objective feature selection in classification. To be more specific, multiple subpopulations related to different evolutionary tasks are separately initialized and then adaptively merged into a single integrated population during the evolution. Moreover, the ideal points for these multi-task subpopulations are dynamically adjusted every generation, in order to achieve different search preferences and evolutionary directions. In the experiments, the proposed MTDEA was compared with seven state-of-the-art MOEAs on 20 high-dimensional classification datasets in terms of three performance indicators, along with using comprehensive Wilcoxon and Friedman tests. It was found that the MTDEA performed the best on most datasets, with a significantly better search ability and promising efficiency.