Gallbladder cancer is a highly aggressive malignancy of the biliary system, often diagnosed at the advanced stage due to its insidious early symptoms, leading to poor overall progno-sis. In recent years, the rapid advancement of artificial intelligence (AI) technologies and their inte-gration into medicine have opened new avenues for the early diagnosis and precision treatment of gallbladder cancer. Currently, AI incorporating deep learning algorithm has significantly improved diagnostic sensitivity and specificity in ultrasound, computed tomography, and pathological analysis. However, clinical translation of AI models remains limited by challenges such as insufficient annota-ted data and limited model interpretability. Future research should focus on establishing multi-center data-sharing mechanisms, developing interpretability tools, and optimizing multimodal data integration strategies, thereby promoting the transformation of AI technologies from an auxiliary diagnostic tool to a core component of clinical decision-making.

Objective·To develop a graph neural network(GNN)-based auxiliary diagnostic model for gallbladder cancer on CT images,and validate its accuracy and feasibility.Methods·From January 2010 to November 2023,1 774 contrast-enhanced CT arterial-phase images were acquired from 887 patients with normal gallbladder,benign gallbladder disease,or gallbladder cancer at Xinhua Hospital and Renji Hospital,Shanghai Jiao Tong University School of Medicine.These images were randomly divided into training and testing sets at a 4∶1 ratio to develop a hybrid GNN-convolutional neural network(CNN)model,named VJK-GIN.The model constructed a pixel-level graph in which each pixel served as a node,and spatial adjacency defined the edges,enabling extraction of local texture features.In the model architecture design,VJK-GIN integrated a three-layer graph isomorphism network,augmented with virtual nodes and jump-knowledge connections;global pooling compressed node features into a graph-level representation,which was classified by a multi-layer perceptron head.Five-fold cross-validation was used to compare VJK-GIN with GNN baselines(GCN,GraphSAGE,GAT,and GIN)and CNN baselines(ViT,EfficientNetV2,and ConvNeXt)in terms of accuracy,precision,recall,F1-score,and area under the receiver operating characteristic curve(AUC).Results·The results of five-fold cross-validation showed that VJK-GIN achieved an F1-score of 0.799(95%CI 0.775?0.823),recall of 0.795(95%CI 0.773?0.817),precision of 0.799(95%CI 0.775?0.823),AUC of 0.812(95%CI 0.792?0.832),and accuracy of 0.773(95%CI 0.748?0.798),surpassing all competing models across every metric.Conclusion·The VJK-GIN model exhibits high stability and accuracy in identifying contrast-enhanced CT images of normal,benign,and malignant gallbladder conditions.

Gallbladder cancer is a highly aggressive malignancy of the biliary system, often diagnosed at the advanced stage due to its insidious early symptoms, leading to poor overall progno-sis. In recent years, the rapid advancement of artificial intelligence (AI) technologies and their inte-gration into medicine have opened new avenues for the early diagnosis and precision treatment of gallbladder cancer. Currently, AI incorporating deep learning algorithm has significantly improved diagnostic sensitivity and specificity in ultrasound, computed tomography, and pathological analysis. However, clinical translation of AI models remains limited by challenges such as insufficient annota-ted data and limited model interpretability. Future research should focus on establishing multi-center data-sharing mechanisms, developing interpretability tools, and optimizing multimodal data integration strategies, thereby promoting the transformation of AI technologies from an auxiliary diagnostic tool to a core component of clinical decision-making.

Objective·To develop a graph neural network(GNN)-based auxiliary diagnostic model for gallbladder cancer on CT images,and validate its accuracy and feasibility.Methods·From January 2010 to November 2023,1 774 contrast-enhanced CT arterial-phase images were acquired from 887 patients with normal gallbladder,benign gallbladder disease,or gallbladder cancer at Xinhua Hospital and Renji Hospital,Shanghai Jiao Tong University School of Medicine.These images were randomly divided into training and testing sets at a 4∶1 ratio to develop a hybrid GNN-convolutional neural network(CNN)model,named VJK-GIN.The model constructed a pixel-level graph in which each pixel served as a node,and spatial adjacency defined the edges,enabling extraction of local texture features.In the model architecture design,VJK-GIN integrated a three-layer graph isomorphism network,augmented with virtual nodes and jump-knowledge connections;global pooling compressed node features into a graph-level representation,which was classified by a multi-layer perceptron head.Five-fold cross-validation was used to compare VJK-GIN with GNN baselines(GCN,GraphSAGE,GAT,and GIN)and CNN baselines(ViT,EfficientNetV2,and ConvNeXt)in terms of accuracy,precision,recall,F1-score,and area under the receiver operating characteristic curve(AUC).Results·The results of five-fold cross-validation showed that VJK-GIN achieved an F1-score of 0.799(95%CI 0.775?0.823),recall of 0.795(95%CI 0.773?0.817),precision of 0.799(95%CI 0.775?0.823),AUC of 0.812(95%CI 0.792?0.832),and accuracy of 0.773(95%CI 0.748?0.798),surpassing all competing models across every metric.Conclusion·The VJK-GIN model exhibits high stability and accuracy in identifying contrast-enhanced CT images of normal,benign,and malignant gallbladder conditions.

Research on the mechanisms of depression is currently a focus of the field of neuroscience.The degeneration of brain plasticity(e.g.,decrease in volume,structural degradation,and functional disorder of the hippocampus,PFC,and CG)leads to depression.Exercise is an important means of improving the symptoms of depression.Current research confirms the importance of improving the volume,structure,and function of the hippocampus,PFC,and CG,in this process,but related research has focused solely on changes to the volume of certain brain regions or connectivity functions.Thus,we lack a comprehensive understanding of the antidepressant mechanisms that improve brain plasticity with exercise.Therefore,this study aimed to explore the roles of brain plasticity in the occurrence of depression and improvements to depression through exercise,promoting a more comprehensive understanding the role of brain plasticity in depression.We also provide new ideas for exercise intervention in depression.

Objective:To construct a traditional Chinese medicine syndrome differentiation model for tinnitus using automatic machine learning technology, and to explore the key factors that affect the results of tinnitus syndrome differentiation.Methods:The clinical characteristics of 594 patients with subjective tinnitus in seven medical units in Shanghai from January 2021 to January 2022 were retrospectively analyzed. The Auto-sklearn automatic machine learning method was used to compare 15 algorithms, and the model with the best classification effect was selected to analyze the key factors affecting tinnitus.Results:The results showed that the optimal algorithm for classification results was the random forest, its accuracy, precision, sensitivity, specificity, F1-score, AUC and kappa coefficient were 87.37%, 88.34%, 89.06%, 96.63%, 88.38%, 97.50%, and 83.37%, respectively. It is concluded that the key factors affecting the classification of the pattern of kidney yin deficiency and fire effulgence, the pattern of liver fire disturbing upward, the pattern of stagnation and binding of phlegm and fire, the pattern of spleen and stomach deficiency, the pattern of wind and heat attacking the external are smooth pulse, string pulse, smooth pulse, weak tongue, and floating pulse respectively.Conclusions:Random forest can provide a good classification prediction function for structured clinical data, suggesting that machine learning technology has clinical application value in assisting the diagnosis of subjective tinnitus.

Objective:To evaluate the clinical value of non-invasive prenatal testing (NIPT) in detecting copy number variations (CNVs).Methods:There were 37 845 pregnant women undergoing NIPT in Guangdong Women and Children Hospital from January 1, 2015 to September 1, 2018, of which 205 with CNVs were detected in addition to chromosome numerical abnormality and retrospectively analyzed. Among the 205 cases, 137 received invasive prenatal diagnosis. Pregnant outcomes were followed up and the efficiency of NIPT in detecting CNVs was analyzed by descriptive statistical analysis.Results:The detection rate of NIPT for CNVs was 0.54% (205/37 845). Among the 137 cases undergoing invasive prenatal diagnosis, 110 showed normal karyotype, 27 with abnormal including two having CNVs that were inconsistent with NIPT findings and 25 with consistent results. The positive predictive value, sensitivity and specificity of NIPT for CNVs were 18.2%(25/137), 100.0%(25/25) and 99.7%(37 625/37 737), respectively. Among the 27 pregnant women with positive findings in prenatal diagnosis, five were lost to follow-up; eight terminated their pregnancies; 14 gave birth to alive baby with normal phenotype. While among the 110 pregnant women with negative results in prenatal diagnosis, 87 delivered full-term neonates including two having patent foramen ovale and 85 with normal phenotype; three gave birth prematurely; one terminated pregnancy at 28 +2 gestational weeks due to preeclampsia; two had inevitable abortion; two requested termination and 15 were lost to follow-up. Conclusions:Routine NIPT has high performance in screening CNVs but those pregnant women with positive NIPT results should be counseled after referring to their invasive prenatal diagnosis results, ultrasound scan and clinical information.

Objective: To compare the outcome of radiofrequency catheter ablation under local anesthesia/sedation (S) or general anesthesia(GA) in atrial fibrillation patients. Methods: Data of 498 patients with atrial fibrillation undergoing radiofrequency catheter ablation in our departmentfrom January 2014 to December 2015 were retrospectively analyzed. Two hundred and twenty patients assigned to the GA group, the other 278 patients to the S group. Patients were followed clinically every 3 months within one year after procedure. Immediate electrocardiogram was performed in patients with palpitation or choking sensation in chest. The end point of the study was recurrence of any atrial tachyarrhythmia lasting >30 seconds in device interrogation, 24-hour Holter monitoring or 12-lead electrocardiogram after a single procedure. After the ablation procedure, a blanking period of 3 months was allowed according to the guidelines. Procedure time, radiofrequency time, fluoroscopy time, the detection of paroxysmal supraventricular tachycardia, the success rate and the complications were compared between the two groups. Results: There was no difference in the baseline characteristics between the two groups, such as age, gender, BMI, complications, LVEF, LAD (all P>0.05). The duration of procedure ((117.8±51.7)minutes vs.(115.4±36.9)minutes, P=0.79), duration of fluoroscopy((12.5±11.2)minutes vs. (10.4±10.2)minutes, P=0.35), duration of radiofrequency((40.1±12.9)minutes vs. (48.6±44.3)minutes, P=0.48) were similar between the two groups (P>0.05). Compared with S group, discovery of the frequency of atrioventricular node reentrant tachycardia (AVNRT) was significantly lower in GA group (0 vs. 3.6%(10/278), P<0.01), but the difference disappeared with repeat electrophysiological examination when patients become conscious from GA(3.2%(7/220) vs. 3.6%(10/278), P=0.311). The difference of atrioventricular reentrant tachycardia (AVRT) was similar between the two groups(0.9%(2/220) vs. 0.7%(2/278), P=0.841). Compared with S group, reflection of vagus nerve was less in GA group (1.4%(3/220) vs. 8.6%(24/278), P=0.026). After following up of (356±92) days, freedom from atrial fibrillation/atrial flutter/atrial tachyarrhythmia was similar between the two groups(77.9%(162/208) vs. 79.9%(215/269), P=0.818). Conclusion General anesthesia is a promising method to atrial fibrillation ablation, in view of stable patient status and safety for the procedure. There is no difference in complications, recurrence of arrhythmia between the two groups, but detection rate of AVNRT is lower in GA group.

Objective:To study the effect of quality control circle(QCC)on the reasonable ratio of clinical prescriptions. Methods:The dispensed prescriptions in orthopedic emergency department were reviewed in our hospital,and the reasons of unreasonable prescriptions were analyzed. According to the QCC technique,the activities were implemented,the standardized work process was made out and the results were studied. Results:After the six-month QCC activities,the unreasonable ratio of emergency orthopedics prescriptions was reduced from 70% to 21% ,and the target yield rate was 140% and the improvement rate was 70% . Conclusion:The QCC has obvious effect on the improvement of reasonable ratio of emergency orthopedics prescriptions.