Target Recognition in SAR Images by Deep Learning with Training Data Augmentation.

Mass production of high-quality synthetic SAR training imagery is essential for boosting the performance of deep-learning (DL)-based SAR automatic target recognition (ATR) algorithms in an open-world environment. To address this problem, we exploit both the widely used Moving and Stationary Target Acquisition and Recognition (MSTAR) SAR dataset and the Synthetic and Measured Paired Labeled Experiment (SAMPLE) dataset, which consists of selected samples from the MSTAR dataset and their computer-generated synthetic counterparts. A series of data augmentation experiments are carried out. First, the sparsity of the scattering centers of the targets is exploited for new target pose synthesis. Additionally, training data with various clutter backgrounds are synthesized via clutter transfer, so that the neural networks are better prepared to cope with background changes in the test samples. To effectively augment the synthetic SAR imagery in the SAMPLE dataset, a novel contrast-based data augmentation technique is proposed. To improve the robustness of neural networks against out-of-distribution (OOD) samples, the SAR images of ground military vehicles collected by the self-developed MiniSAR system are used as the training data for the adversarial outlier exposure procedure. Simulation results show that the proposed data augmentation methods are effective in improving both the target classification accuracy and the OOD detection performance. The purpose of this work is to establish the foundation for large-scale, open-field implementation of DL-based SAR-ATR systems, which is not only of great value in the sense of theoretical research, but is also potentially meaningful in the aspect of military application.

Thirty-day outcomes of in-hospital multi-vessel versus culprit-only revascularization strategy for ST-segment elevation myocardial infarction with multivessel coronary disease.

BACKGROUND: Many studies have demonstrated the benefit of complete multivessel revascularization versus culprit-only intervention in patients of ST-segment elevation myocardial infarction (STEMI) and multivessel coronary artery disease. However, only a few single-center retrospective studies were performed on small Chinese cohorts. Our study aims to demonstrate the advantage of multivessel percutaneous intervention (PCI) strategy on 30-day in-hospital outcomes to patients with STEMI and multivessel disease in larger Chinese population. METHODS: From the Improving Care for Cardiovascular Disease in China-Acute Coronary Syndrome (CCC-ACS) project, 5935 patients with STEMI and multivessel disease undergoing PCI and hospitalized for fewer than 30 days were analyzed. After 5: 1 propensity score matching, 3577 patients with culprit-only PCI and 877 with in-hospital multivessel PCI were included. The primary outcome was major adverse cardiovascular and cerebrovascular event (MACCE), defined as a composite of myocardial infarction, all-cause death, stent thrombosis, heart failure, and stroke. RESULTS: Multivariable logistic regression analysis revealed that in-hospital multivessel PCI was associated with lower risk of 30-day MACCE (adjusted OR = 0.75, 95% CI: 0.57-0.98, P = 0.032) than culprit-only PCI and conferred no increased risk of all-cause death, myocardial infarction, stent thrombosis, stroke, or bleeding. Subgroup analysis showed that MACCE reduction was observed more often from patients with trans-femoral access (OR = 0.34, 95% CI: 0.15-0.74) than with trans-radial access (OR = 0.87, 95% CI: 0.66-1.16, P for interaction = 0.017). CONCLUSIONS: The in-hospital multivessel PCI strategy was associated with a lower risk of 30-day MACCE than culprit-only PCI in patients with STEMI and multivessel coronary artery disease.

[An excel VBA-based electronic marrow testing report].

This paper introduces an Excel VBA-based method developed for electronic marrow testing reports.