Vehicle re-identification based on dimensional decoupling strategy and non-local relations.

Vehicle re-identification (Re-ID) is a challenging task that aims to recognize the same vehicle across different non-overlapping cameras. Existing attention mechanism-based methods for vehicle Re-ID often suffer from significant intra-class variation and inter-class variation due to various factors such as illumination, occlusion, viewpoint, etc. In this paper, we propose a novel network architecture for vehicle Re-ID, named Dimensional Decoupling Strategy and Non-local Relationship Network (DMNR-Net), which uses three modules to extract complementary features: global feature extraction module, non-local relationship capture module(NRCM), and dimensional decoupling module (DDS). The global feature extraction module captures complete and coarse-grained features from the whole image; the NRCM module extracts saliency information from feature maps in both spatial and channel dimensions; and the DDS decouples spatial and channel features into two branches to extract fine-grained features and focus on specific subspaces. We conduct extensive experiments on two popular publicly datasets, VeRi-776 and VehicleID, to evaluate the effectiveness of our method. The experimental results show that our DMNR-Net outperforms state-of-the-art methods by a large margin on both datasets.

Assessment of the invisible blood contamination on nurses' gloved hands during vascular access procedures in a hemodialysis unit.

A prospective study was conducted to assess potential invisible blood contamination on nurses' gloved hands during vascular access procedures using the occult blood detection method in a hemodialysis unit. 60.13% (273/454) of samples tested positive for hemoglobin. These results highlighted the importance of hand hygiene and glove change during hemodialysis access care.

[Establishment of a nomogram prediction model for coronary artery disease risk in elderly patients with acute myocardial infarction].

OBJECTIVE: To establish a nomogram model for predicting the risk of coronary artery disease in elderly patients with acute myocardial infarction (AMI). METHODS: The clinical data of elderly patients with AMI who underwent coronary angiography in the department of cardiology of Cangzhou Central Hospital from July 2015 to March 2020 were analyzed, including age, gender, smoking history, underlying diseases, family history, blood pressure, left ventricular ejection fraction (LVEF), and several biochemical indicators at admission, such as total cholesterol (TC), triglyceride (TG), high-density lipoprotein cholesterol (HDL-C), low-density lipoprotein cholesterol (LDL-C), lipoprotein [Lp(a)], apolipoproteins (ApoA, ApoB), ApoA/B ratio, total bilirubin (TBil), direct bilirubin (DBil), indirect bilirubin (IBil), fasting blood glucose (FBG) and uric acid (UA). Patients were divided into model group (2 484 cases) and validation group (683 cases) according to the ratio of 8:2. According to Gensini score, the model group and validation group were divided into mild lesion group (0-20 points) and severe lesion group (≥ 81 points). The differences of each index between different coronary lesion degree groups were compared. Lasso regression and Logistic regression were used to analyze the risk factors of aggravating coronary lesion risk in elderly patients with AMI, and then the nomogram prediction model was established for evaluation and external validation. RESULTS: (1) In the model group, there were significant differences in the family history of coronary heart disease, FBG and HDL-C between the mild lesion group (411 cases) and the severe lesion group (417 cases) [family history of coronary heart disease: 3.6% vs. 7.7%, FBG (mmol/L): 5.88±1.74 vs. 6.43±2.06, HDL-C (mmol/L): 1.48±0.69 vs. 1.28±0.28, all P < 0.05]. In the validation group, there were significant differences between the mild lesion group (153 cases) and the severe lesion group [132 cases; FBG (mmol/L): 5.58±0.88 vs. 6.85±0.79, HDL-C (mmol/L): 1.59±0.32 vs. 1.16±0.21, both P < 0.05]. (2) Lasso regression analysis showed that family history of coronary heart disease, FBG, and HDL-C were risk factors of coronary artery disease in elderly patients with AMI, with coefficients 0.118, 0.767, and -0.558, respectively. Logistic regression analysis showed that FBG [odds ratio (OR) = 1.479, 95% confidence interval (95%CI) was 1.051-2.082, P = 0.025] and HDL-C (OR = 0.386, 95%CI was 0.270-0.553, P < 0.001] were independent risk factors of coronary artery disease in elderly patients with AMI. (3) According to the rank score of FBG and HDL-C, the nomogram prediction risk model of aggravating coronary artery disease degree was established for each patient. It was concluded that the risk of coronary artery disease in elderly people with higher FBG level and (or) lower HDL-C level was significantly increased. (4) The nomogram model constructed with the model group data predicted the risk concordance index (C-index) was 0.689, and the C-index of the external validation group was 0.709. CONCLUSIONS: FBG and HDL-C are independent risk factors for aggravating coronary artery disease in elderly patients with AMI. The nomogram model of aggravating coronary artery disease in elderly patients with AMI has good predictive ability, which can provide more intuitive research methods and clinical value for preventing the aggravation of coronary artery disease in elderly patients.