Subtype classification of age-related macular degeneration (AMD) based on optical coherence tomography (OCT) images serves as an effective auxiliary tool for clinicians in diagnosing disease progression and formulating treatment plans. To improve the classification accuracy of AMD subtypes, this study proposes a keypoint-based, attention-enhanced residual network (KPA-ResNet). The proposed architecture adopts a 50-layer residual network (ResNet-50) as the backbone, preceded by a keypoint localization module based on heatmap regression to outline critical lesion regions. A two-dimensional relative self-attention mechanism is incorporated into convolutional layers to enhance the representation of key lesion areas. Furthermore, the network depth is appropriately increased and an improved residual module, ConvNeXt, is introduced to enable comprehensive extraction of high-dimensional features and enrich the detail of lesion boundary contours, ultimately achieving higher classification accuracy of AMD subtypes. Experimental results demonstrate that KPA-ResNet achieves significant improvements in overall classification accuracy compared with conventional convolutional neural networks. Specifically, for the wet AMD subtypes, the classification accuracies for inactive choroidal neovascularization (CNV) and active CNV reach 92.8% and 95.2%, respectively, representing substantial improvement over ResNet-50. These findings validate the superior performance of KPA-ResNet in AMD subtype classification tasks. This work provides a high-accuracy, generalizable network architecture for OCT-based AMD subtype classification and offers new insights into integrating attention mechanisms with convolutional neural networks in ophthalmic image analysis.
Tomography, Optical Coherence/methods* ; Humans ; Macular Degeneration/diagnostic imaging* ; Neural Networks, Computer

Objective:To investigate the relationship between the changes in extracellular vesicles (EVs) size distribution before and after cardiopulmonary bypass (CPB) cardiac surgery and postoperative coagulation function.Methods:A total of 103 patients undergoing cardiac surgery with CPB were enrolled. Venous blood samples were collected at preoperation, postoperative 12 h and 3 days. Additionally, 50 age- and gender-matched healthy volunteers served as a control group. EVs were isolated using gradient centrifugation, and their size distribution was assessed by dynamic light scattering (DLS). The relationship between EV size characteristics, including peak diameter, peak height, and interquartile range( IQR), and postoperative coagulation function was analyzed. Results:Compared to patients with normal postoperative coagulation function, those with postoperative coagulation dysfunction had lower size at peak and IQR, and significantly higher peak intensity. Logistic regression analysis indicated that elevated peak intensity and lower size at peak and IQR were risk factors for coagulation dysfunction. The area under the curve ( AUC) for diagnosing coagulation dysfunction with 12 h postoperative EVs peak intensity was 0.76, with a positive predictive value of 85% at the optimal cutoff of 8.2; the AUC for IQR was 0.84, with a sensitivity of 83%, specificity of 82%, and negative predictive value of 86% at the optimal cutoff of 125.05 nm. Conclusion:The size distribution of circulating EVs show a correlation with coagulation function after cardiac surgery with CPB and may serve as a novel biomarker to predict postoperative coagulation dysfunction.

Objective:To analyze the underlying issues and root causes of primary healthcare resource allocation in China,to identify the key elements of the current mismatch in primary healthcare resource allocation,and to propose targeted optimization strategies.Methods:Utilizing the fuzzy set Qualitative Comparative Analysis(fsQCA)method to conduct a comprehensive investigation on the fundamental drivers of misallocation in the provision of primary healthcare resources in China,along with the associated grouping patterns.Results:From 2011 to 2021,the overall misallocation of primary medical and health resources in 31 provinces and cities of China showed a downward trend,with the average value decreasing from 0.801 to 0.661.There were 8 configuration paths to improve the optimal allocation of primary medical resources,including 2 biased factors-driven,4"system,society,and population"three-way synergistic types,and 2 multi-factor comprehensive synergistic types.Conclusion:It is recommended that all provinces need to adhere to the principle of government leadership and promote comprehensive reform of the medical and healthcare sector,with particular consideration given to the specific medical needs of the elderly.Furthermore,it is essential to implement the strategic guiding principle of diversified and synergistic development,with the objective of promoting a balanced distribution of high-quality medical resources.

Objective:To explore the temporal characteristics,spatial effects,spatial correlation patterns and dynamic evolution laws of the coordinated development of medical resource allocation and economic society in China from 2010 to 2021,and to provide scientific references for promoting the coordinated and orderly development of medical and health services and economic society.Methods:The coupling coordination theory is introduced to construct a"time+space"dual-dimensional dynamic coupling coordination degree model,and the spatial correlation analysis method is used to analyze the spatial and temporal evolution characteristics of China's medical resource allocation and economic and social coupling coordination level from 2010 to 2021.On this basis,the grey prediction model is used to dynamically predict the future development and changes of the coupling coordination level of 31 provinces from 2024 to 2028.Results:From 2010 to 2021,the level of coupling and coordination between medical resource allocation and economic and social development has been steadily increased,but the overall coordination level is low,the regional gap is large,and the benign resonance coupling and coordinated and orderly development have not been fully realized;there is a significant spatial positive correlation and spatial agglomeration effect between the allocation of medical resources and the level of economic and social coupling and coordination,mainly based on the spatial correlation mode of high-high agglomeration and low-low agglomeration.From 2024 to 2028,the gap of interregional coupling coordination level will be narrowed,and the western region will still be in the stage of imbalance.Conclusion:We should focus on regional geospatial differences,improve the cross-regional multilateral coordinated development mechanism,promote mutual assistance and sharing of medical resources,technology,capital and management experience,establish and improve the investment mechanism suitable for the level of economic and social development,and promote the coordinated and orderly development of medical and health undertakings and economic society.

Objective:To analyze the underlying issues and root causes of primary healthcare resource allocation in China,to identify the key elements of the current mismatch in primary healthcare resource allocation,and to propose targeted optimization strategies.Methods:Utilizing the fuzzy set Qualitative Comparative Analysis(fsQCA)method to conduct a comprehensive investigation on the fundamental drivers of misallocation in the provision of primary healthcare resources in China,along with the associated grouping patterns.Results:From 2011 to 2021,the overall misallocation of primary medical and health resources in 31 provinces and cities of China showed a downward trend,with the average value decreasing from 0.801 to 0.661.There were 8 configuration paths to improve the optimal allocation of primary medical resources,including 2 biased factors-driven,4"system,society,and population"three-way synergistic types,and 2 multi-factor comprehensive synergistic types.Conclusion:It is recommended that all provinces need to adhere to the principle of government leadership and promote comprehensive reform of the medical and healthcare sector,with particular consideration given to the specific medical needs of the elderly.Furthermore,it is essential to implement the strategic guiding principle of diversified and synergistic development,with the objective of promoting a balanced distribution of high-quality medical resources.

Objective:To explore the temporal characteristics,spatial effects,spatial correlation patterns and dynamic evolution laws of the coordinated development of medical resource allocation and economic society in China from 2010 to 2021,and to provide scientific references for promoting the coordinated and orderly development of medical and health services and economic society.Methods:The coupling coordination theory is introduced to construct a"time+space"dual-dimensional dynamic coupling coordination degree model,and the spatial correlation analysis method is used to analyze the spatial and temporal evolution characteristics of China's medical resource allocation and economic and social coupling coordination level from 2010 to 2021.On this basis,the grey prediction model is used to dynamically predict the future development and changes of the coupling coordination level of 31 provinces from 2024 to 2028.Results:From 2010 to 2021,the level of coupling and coordination between medical resource allocation and economic and social development has been steadily increased,but the overall coordination level is low,the regional gap is large,and the benign resonance coupling and coordinated and orderly development have not been fully realized;there is a significant spatial positive correlation and spatial agglomeration effect between the allocation of medical resources and the level of economic and social coupling and coordination,mainly based on the spatial correlation mode of high-high agglomeration and low-low agglomeration.From 2024 to 2028,the gap of interregional coupling coordination level will be narrowed,and the western region will still be in the stage of imbalance.Conclusion:We should focus on regional geospatial differences,improve the cross-regional multilateral coordinated development mechanism,promote mutual assistance and sharing of medical resources,technology,capital and management experience,establish and improve the investment mechanism suitable for the level of economic and social development,and promote the coordinated and orderly development of medical and health undertakings and economic society.

Objective:To investigate the relationship between the changes in extracellular vesicles (EVs) size distribution before and after cardiopulmonary bypass (CPB) cardiac surgery and postoperative coagulation function.Methods:A total of 103 patients undergoing cardiac surgery with CPB were enrolled. Venous blood samples were collected at preoperation, postoperative 12 h and 3 days. Additionally, 50 age- and gender-matched healthy volunteers served as a control group. EVs were isolated using gradient centrifugation, and their size distribution was assessed by dynamic light scattering (DLS). The relationship between EV size characteristics, including peak diameter, peak height, and interquartile range( IQR), and postoperative coagulation function was analyzed. Results:Compared to patients with normal postoperative coagulation function, those with postoperative coagulation dysfunction had lower size at peak and IQR, and significantly higher peak intensity. Logistic regression analysis indicated that elevated peak intensity and lower size at peak and IQR were risk factors for coagulation dysfunction. The area under the curve ( AUC) for diagnosing coagulation dysfunction with 12 h postoperative EVs peak intensity was 0.76, with a positive predictive value of 85% at the optimal cutoff of 8.2; the AUC for IQR was 0.84, with a sensitivity of 83%, specificity of 82%, and negative predictive value of 86% at the optimal cutoff of 125.05 nm. Conclusion:The size distribution of circulating EVs show a correlation with coagulation function after cardiac surgery with CPB and may serve as a novel biomarker to predict postoperative coagulation dysfunction.

Objective:To investigate the effect of regional leptomeningeal collateral circulation (rLMC) score based on CT angiography (CTA) and onset-to-reperfusion time (OTR) on the outcome after endovascular treatment (EVT) in patients with anterior circulation acute large vessel occlusive stroke (ACA-LVOS).Methods:Patients with ACA-LVOS underwent EVT in the Department of Neurology, the Affiliated Hospital of Yangzhou University from July 2017 to July 2023 were included retrospectively. The rLMC score 0-10 was defined as poor collateral circulation, and 11-20 were defined as good collateral circulation. At 90 days after EVT, the modified Rankin Scale (mRS) was used to evaluate the outcome. A score of 0-2 was defined as a good outcome and 3-6 were defined as a poor outcome. Multivariate logistic regression analysis was used to determine the independent influencing factors of the outcome after EVT. Results:A total of 144 patients with ACA-LVOS underwent EVT were enrolled, including 78 males (54.2%), median aged 73 years. The median baseline National Institutes of Health Stroke Scale (NIHSS) score was 16, the median baseline Alberta Stroke Program Early CT Score (ASPECTS) was 9, and the median OTR was 330.5 minutes. Eighty patients (55.6%) had good collateral circulation, 63 (43.8%) had poor outcome, including 13 deaths. Univariate analysis showed that there were significant differences in hypertension, previous stroke history, smoking, triglycerides, baseline NIHSS score, baseline ASPECTS, OTR, and collateral circulation status between the good outcome group and the poor outcome group (all P<0.05). Multivariate logistic regression analysis showed that good collateral circulation (odds ratio [ OR] 0.223, 95% confidence interval [ CI] 0.077-0.643; P=0.005) was an independent predictor of good outcome. In the poor collateral circulation group, longer OTR was an independent predictor of poor outcome ( OR 1.020, 95% CI 1.008-1.032; P=0.001). In the good collateral circulation group, longer OTR was not an independent risk factor for poor outcome ( OR 1.005, 95% CI 1.000-1.010; P=0.062). Conclusion:rLMC score based on CTA and OTR are the independent predictors of the outcome after EVT in patients with ACA-LVOS.

Objective:To investigate the correlations between CO-029 expression and cholangiocarcinoma invasion and metastasis, and the further explore the potential mechanism involved.Methods:The constructed lentiviral vector of vshRNA-CO-029 (LV/GFP/CO-029) was used to transfect and screen the stable transfected cholangiocarcinoma cell line HCCC-9810-vshRNA-CO-029 as the silence group, HCCC-9810 cells transfected with the mock plasmid were used as the mock group, and the untransfected cells were used as the control group. Cell scratch assay, Transwell assay and in vivo implantation assay were used to detect the migration, invasion and metastasis of the three groups of cells. Immunoprecipitation and tumor necrosis factor (TNF)-α stimulation assay were used to detect the effect of CO-029 on the expression of EMT-related genes.Results:The scratch healing rate of the silence group was (27.11±4.58)%, which was lower than that in mock group (92.84±6.24)%, the number of cells passing through Matrigel in silence group was (57.15±6.10), which was significantly lower than that in mock group (108.20±9.21) and control group (112.00±10.45), the differences were statistically significant ( all P<0.01). The volume of liver tumors in the silence group of orthotopic xenograft mouse model was (2.17±0.54) cm 3, while the volume of liver tumors in the transplanted simulation group was (0.74±0.15) cm 3, the differences were statistically significant ( P<0.05). The incidence of lung metastasis and the number of lung metastases in the simulated group was 100%(6/6) and (214.17±35.64), respectively, while that in the silence group was 16.7% (1/6) and (41.56±14.15), respectively, the differences were statistically significant (all P<0.05). Co-immunoprecipitation showed that CO-029 can form a complex with TNF-αR1. TNF-α induced the down-regulation of E-cadherin and up-regulation of vimentin and N-cadherin in the mock group, but no significant changes were observed in the silence group. Conclusion:CO-029 expression is positively correlated with tumor invasion and metastasis of cholangiocarcinoma, and could couple with TNF-α to induce EMT, which is a novel well-established potential prognostic and therapeutic target for cholangiocarcinoma metastasis and prognosis intervention.