Objective To evaluate the performance of the artificial intelligence (AI)-enabled snail identification system for recognition of Oncomelania hupensis robertsoni and Tricula in schistosomiasis-endemic areas of Yunnan Province. Methods Fifty O. hupensis robertsoni and 50 Tricula samples were collected from Yongbei Township, Yongsheng County, Lijiang City, a schistosomiasis-endemic area in Yunnan Province in May 2024. A total of 100 snail sample images were captured with smartphones, including front-view images of 25 O. hupensis robertsoni and 25 Tricula samples (upward shell opening) and back-view images of 25 O. hupensis robertsoni and 25 Tricula samples (downward shell opening). Snail samples were identified as O. hupensis robertsoni or Tricula by schistosomiasis control experts with a deputy senior professional title and above according to image quality and morphological characteristics. A standard dataset for snail image classification was created, and served as a gold standard for recognition of snail samples. A total of 100 snail sample images were recognized with the AI-enabled intelligent snail identification system based on a WeChat mini program in smartphones. Schistosomiasis control professionals were randomly sampled from stations of schistosomisis prevention and control and centers for disease control and prevention in 18 schistosomiasis-endemic counties (districts, cities) of Yunnan Province, for artificial identification of 100 snail sample images. All professionals are assigned to two groups according the median years of snail survey experiences, and the effect of years of snail survey experiences on O. hupensis robertsoni sample image recognition was evaluated. A receiver operating characteristic (ROC) curve was plotted, and the sensitivity, specificity, accuracy, Youden’s index and the area under the curve (AUC) of the AI-enabled intelligent snail identification system and artificial identification were calculated for recognition of snail sample images. The snail sample image recognition results of AI-enabled intelligent snail identification system and artificial identification were compared with the gold standard, and the internal consistency of artificial identification results was evaluated with the Cronbach’s coefficient alpha. Results A total of 54 schistosomiasis control professionals were sampled for artificial identification of snail sample image recognition, with a response rate of 100% (54/54), and the accuracy, sensitivity, specificity, Youden’s index, and AUC of artificial identification were 90%, 86%, 94%, 0.80 and 0.90 for recognition of snail sample images, respectively. The overall Cronbach’s coefficient alpha of artificial identification was 0.768 for recognition of snail sample images, and the Cronbach’s coefficient alpha was 0.916 for recognition of O. hupensis robertsoni snail sample images and 0.925 for recognition of Tricula snail sample images. The overall accuracy of artificial identification was 90% for recognition of snail sample images, and there was no significant difference in the accuracy of artificial identification for recognition of O. hupensis robertsoni (86%) and Tricula snail sample images (94%) (χ² = 1.778, P > 0.05). There was no significant difference in the accuracy of artificial identification for recognition of snail sample images with upward (88%) and downward shell openings (92%) (χ² = 0.444, P > 0.05), and there was a significant difference in the accuracy of artificial identification for recognition of snail sample images between schistosomiasis control professionals with snail survey experiences of 6 years and less (75%) and more than 6 years (90%) (χ² = 7.792, P < 0.05). The accuracy, sensitivity, specificity and AUC of the AI-enabled intelligent snail identification system were 88%, 100%, 76% and 0.88 for recognition of O. hupensis robertsoni snail sample images, and there was no significant difference in the accuracy of recognition of O. hupensis robertsoni snail sample images between the AI-enabled intelligent snail identification system and artificial identification (χ² = 0.204, P > 0.05). In addition, there was no significant difference in the accuracy of artificial identification for recognition of snail sample images with upward (90%) and downward shell openings (86%) (χ² = 0.379, P > 0.05), and there was a significant difference in the accuracy of artificial identification for recognition of snail sample images between schistosomiasis control professionals with snail survey experiences of 6 years and less and more than 6 years (χ² = 5.604, P_adjusted < 0.025). Conclusions The accuracy of recognition of snail sample images is comparable between the AI-enabled intelligent snail identification system and artificial identification by schistosomiasis control professionals, and the AI-enabled intelligent snail identification system is feasible for recognition of O. hupensis robertsoni and Tricula in Yunnan Province.

Objective To construct a visual intelligent recognition model for Oncomelania hupensis robertsoni in Yunnan Province based on the EfficientNet-B4 model, and to evaluate the impact of data augmentation methods and model hyperparameters on the recognition of O. hupensis robertsoni. Methods A total of 400 O. hupensis robertsoni and 400 Tricula snails were collected from Yongsheng County, Yunnan Province in June 2024, and snail images were captured following identification and classification of 300 O. hupensis robertsoni and 300 Tricula snails. A total of 925 O. hupensis robertsoni images and 1 062 Tricula snail images were collected as a dataset and divided into a training set and a validation set at a ratio of 8:2, while 352 images captured from the remaining 100 O. hupensis robertsoni and 354 images from the remaining 100 Tricula snails served as an external test set. All acquired images were subjected to preprocessing, including cropping and resizing. Three data augmentation approaches were employed, including baseline, Mixup and Gaussian blurring, and model hyperparameters included two optimization algorithms of adaptive moment estimation (Adam) and stochastic gradient descent (SGD), two loss functions of focal loss and cross entropy loss, and two learning rate decay strategies of cosine annealing and multi-step. The intelligent recognition models of O. hupensis robertsoni and Tricula snails were constructed based on the EfficientNet-B4 model, and 7 training strategy groups were generated by combinations of different data augmentation approaches and hyperparameters. The performance of intelligent recognition models was tested with external test sets, and evaluated with accuracy, precision, recall, F1 score, loss, Youden’s index, and the area under the receiver operating characteristic curve (AUC) under different training strategies. Results The variation of loss values was comparable among intelligent recognition models with different data augmentation approaches. The Group 4 model constructed with Mixup and Gaussian blurring data augmentation approaches showed the optimal performance, with an accuracy of 90.38%, precision of 90.07%, F1 score of 89.44%, Youden’s index of 0.81 and AUC of 0.961 in the external test set. The accuracy of models using the SGD optimizer reduced by 29.16% as compared to those using the Adam optimizer (χ² = 81.325, P < 0.001), and the accuracy of models using the cross entropy loss function reduced by 0.80% as compared to the Group 4 model (χ² = 3.147, P > 0.05), while the accuracy of models using the multi-step learning rate decay strategy increased by 0.65% as compared to the Group 4 model (χ² = 0.208, P > 0.05). In addition, the model with the baseline + Mixup + Gaussianblurring data augmentation approach and hyperparameters of Adam optimizer, focal loss function and multi-step learning rate decay strategy showed the highest performance, with an accuracy of 91.03%, precision of 91.97%, recall of 88.11%, F1 score of 90.00%, Youden’s index of 0.82 and AUC values of 0.969 in external test set, respectively. Conclusions The intelligent recognition model of O. hupensis robertsoni based on EfficientNet-B4 model is accurate for identification of O. hupensis robertsoni and Tricula snails in Yunnan Province.

Objective:To investigate the surgical treatment effect and prognostic factors of hilar cholangiocarcinoma.Methods:This is an ambispective cohort study. From August 2005 to December 2022,data of 510 patients who diagnosed with hilar cholangiocarcinoma and underwent surgical resection at the Hepatobiliary Center of the First Affiliated Hospital of Nanjing Medical University were retrospectively collected. In the cohort,there were 324 males and 186 females,with an age of ( M (IQR)) 63(13)years (range:25 to 85 years). The liver function at admission was Child-Pugh A (343 cases,67.3%) and Child-Pugh B (167 cases,32.7%). Three hundred and seventy-two(72.9%) patients had jaundice symptoms and the median total bilirubin was 126.3(197.6) μmol/L(range: 5.4 to 722.8 μmol/L) at admission. Two hundred and fourty-seven cases (48.4%) were treated with percutaneous transhepatic cholangial drainage or endoscopic nasobiliary drainage before operation. The median bilirubin level in the drainage group decreased from 186.4 μmol/L to 85.5 μmol/L before operation. Multivariate Logistic regression was used to identify the influencing factors for R0 resection,and Cox regression was used to construct multivariate prediction models for overall survival(OS) and disease-free survival(DFS). Results:Among 510 patients who underwent surgical resection,Bismuth-Corlett type Ⅲ-Ⅳ patients accounted for 71.8%,among which 86.1% (315/366) underwent hemi-hepatectomy,while 81.9% (118/144) underwent extrahepatic biliary duct resection alone in Bismuch-Corlett type Ⅰ-Ⅱ patients. The median OS time was 22.8 months, and the OS rates at 1-,3-,5-and 10-year were 72.2%,35.6%,24.8% and 11.0%,respectively. The median DFS time was 15.2 months,and the DFS rates was 66.0%,32.4%,20.9% and 11.0%,respectively. The R0 resection rate was 64.5% (329/510), and the OS rates of patients with R0 resection at 1-,3-,5-and 10-year were 82.5%, 48.6%, 34.4%, 15.2%,respectively. The morbidity of Clavien-Dindo grade Ⅲ-Ⅴ complications was 26.1%(133/510) and the 30-day mortality was 4.3% (22/510). Multivariate Logistic regression indicated that Bismuth-Corlett type Ⅰ-Ⅲ ( P=0.009), hemi-hepatectomy and extended resection ( P=0.001),T1 and T2 patients without vascular invasion (T2 vs. T1: OR=1.43 (0.61-3.35), P=0.413;T3 vs. T1: OR=2.57 (1.03-6.41), P=0.010;T4 vs. T1, OR=3.77 (1.37-10.38), P<0.01) were more likely to obtain R0 resection. Preoperative bilirubin,Child-Pugh grade,tumor size,surgical margin,T stage,N stage,nerve infiltration and Edmondson grade were independent prognostic factors for OS and DFS of hilar cholangiocarcinoma patients without distant metastasis. Conclusions:Radical surgical resection is necessary to prolong the long-term survival of hilar cholangiocarcinoma patients. Hemi-hepatectomy and extended resection,regional lymph node dissection and combined vascular resection if necessary,can improve R0 resection rate.

Objective:To investigate the surgical treatment effect and prognostic factors of hilar cholangiocarcinoma.Methods:This is an ambispective cohort study. From August 2005 to December 2022,data of 510 patients who diagnosed with hilar cholangiocarcinoma and underwent surgical resection at the Hepatobiliary Center of the First Affiliated Hospital of Nanjing Medical University were retrospectively collected. In the cohort,there were 324 males and 186 females,with an age of ( M (IQR)) 63(13)years (range:25 to 85 years). The liver function at admission was Child-Pugh A (343 cases,67.3%) and Child-Pugh B (167 cases,32.7%). Three hundred and seventy-two(72.9%) patients had jaundice symptoms and the median total bilirubin was 126.3(197.6) μmol/L(range: 5.4 to 722.8 μmol/L) at admission. Two hundred and fourty-seven cases (48.4%) were treated with percutaneous transhepatic cholangial drainage or endoscopic nasobiliary drainage before operation. The median bilirubin level in the drainage group decreased from 186.4 μmol/L to 85.5 μmol/L before operation. Multivariate Logistic regression was used to identify the influencing factors for R0 resection,and Cox regression was used to construct multivariate prediction models for overall survival(OS) and disease-free survival(DFS). Results:Among 510 patients who underwent surgical resection,Bismuth-Corlett type Ⅲ-Ⅳ patients accounted for 71.8%,among which 86.1% (315/366) underwent hemi-hepatectomy,while 81.9% (118/144) underwent extrahepatic biliary duct resection alone in Bismuch-Corlett type Ⅰ-Ⅱ patients. The median OS time was 22.8 months, and the OS rates at 1-,3-,5-and 10-year were 72.2%,35.6%,24.8% and 11.0%,respectively. The median DFS time was 15.2 months,and the DFS rates was 66.0%,32.4%,20.9% and 11.0%,respectively. The R0 resection rate was 64.5% (329/510), and the OS rates of patients with R0 resection at 1-,3-,5-and 10-year were 82.5%, 48.6%, 34.4%, 15.2%,respectively. The morbidity of Clavien-Dindo grade Ⅲ-Ⅴ complications was 26.1%(133/510) and the 30-day mortality was 4.3% (22/510). Multivariate Logistic regression indicated that Bismuth-Corlett type Ⅰ-Ⅲ ( P=0.009), hemi-hepatectomy and extended resection ( P=0.001),T1 and T2 patients without vascular invasion (T2 vs. T1: OR=1.43 (0.61-3.35), P=0.413;T3 vs. T1: OR=2.57 (1.03-6.41), P=0.010;T4 vs. T1, OR=3.77 (1.37-10.38), P<0.01) were more likely to obtain R0 resection. Preoperative bilirubin,Child-Pugh grade,tumor size,surgical margin,T stage,N stage,nerve infiltration and Edmondson grade were independent prognostic factors for OS and DFS of hilar cholangiocarcinoma patients without distant metastasis. Conclusions:Radical surgical resection is necessary to prolong the long-term survival of hilar cholangiocarcinoma patients. Hemi-hepatectomy and extended resection,regional lymph node dissection and combined vascular resection if necessary,can improve R0 resection rate.

Diabetic neuropathic pain (DNP) is the most common disabling complication of diabetes. Emerging evidence has linked the pathogenesis of DNP to the aberrant sprouting of sensory axons into the epidermal area; however, the underlying molecular events remain poorly understood. Here we found that an axon guidance molecule, Netrin-3 (Ntn-3), was expressed in the sensory neurons of mouse dorsal root ganglia (DRGs), and downregulation of Ntn-3 expression was highly correlated with the severity of DNP in a diabetic mouse model. Genetic ablation of Ntn-3 increased the intra-epidermal sprouting of sensory axons and worsened the DNP in diabetic mice. In contrast, the elevation of Ntn-3 levels in DRGs significantly inhibited the intra-epidermal axon sprouting and alleviated DNP in diabetic mice. In conclusion, our studies identified Ntn-3 as an important regulator of DNP pathogenesis by gating the aberrant sprouting of sensory axons, indicating that Ntn-3 is a potential druggable target for DNP treatment.
Mice ; Animals ; Diabetes Mellitus, Experimental/metabolism* ; Axons/physiology* ; Diabetic Neuropathies ; Sensory Receptor Cells/metabolism* ; Neuralgia/metabolism*

Objective:To explore the potential mechanism of Fuzheng Jiedu Decoction created by professor Yu Huiping in the treatment of primary immune thrombocytopenia (ITP) in children based on network pharmacology.Methods:The targets of Fuzheng Jiedu Decoction and ITP were retrieved within SymMap database and TCMID database, and all the common genes in the potential targets of the decoction and ITP were retained. The interaction relationship among the targets was obtained in the String database, and cluster analysis was conducted to obtain the core target group of Fuzheng Jiedu Decoction for ITP. In the David database, the potential KEGG Pathway was obtained through enrichment analysis, the Pathway of non-specific diseases was classified and selected, and a network of "Traditional Chinese Medicine - Target - Pathway" was constructed.Results:There are 500 potential targets for Fuzheng Jiedu Decoction to treat ITP. After Cluster analysis of PPI network, a total of 16 gene clusters were obtained, among which Cluster 1 score was 65.663, making it a potential core target group for Fuzheng Jiedu Decoction to treat ITP. The core enriched target group amounts to 114 pathways, and there were four first-level catalogs which includes Human Diseases (50%), Organismal Systems (25%), Environmental Information Processing (17%), and Cellular Processes (8%). Among them, TNF signaling pathway and HIF-1 signaling pathway were highly enriched for non-specific diseases. In the nodes of the network, The Chinese herbs with the highest Degree of aggregation in the network nodes were Agrimoniae herba (Degree=66), Glycyrrhizae radix et rhizoma praeparata cum melle (Degree=64), the target proteins were MAPK3 (Degree=51),MAPK1 (Degree=50),and the pathway was PI3K-Akt signaling pathway (Degree=29). Conclusion:Fuzheng Jiedu Decoction is mainly used to treat children's ITP with Agrimoniae herba and Glycyrrhizae radix et rhizoma praeparata cum melle,and it is related to the regulation of platelet number, adhesion and focusing.

Objective:To study ApoE gene polymorphism in patients with intracerebral hemorrhage at different ages and its relationship with blood lipid level.Methods:From December 2017 to December 2019, 360 inpatients with cerebral hemorrhage received by neurosurgery department of Weihai Central Hospital were selected as the cerebral hemorrhage group.According to the age of patients, patients with cerebral hemorrhage were divided into low age group (age<30 years), middle age group (30 years≤age≤60 years) and high age group (age>60 years), with 120 cases in each group.Another 120 healthy persons who received health examination during the same period were selected as control group.The levels of high density lipoprotein (HDL), low density lipoprotein (LDL), triglyceride (TG) and total cholesterol (TC) in peripheral blood were measured.ApoE gene polymorphism was determined by polymerase chain reaction-restriction fragment length polymorphism (PCR-RFLP).Results:There were no statistically significant differences in sex, age, body mass index between the cerebral hemorrhage group and the control group (χ 2=0.045, t=0.459, t=0.655, P=0.832, 0.323, 0.256). The systolic blood pressure, diastolic blood pressure, history of hypertension, history of diabetes, smoking and drinking in the cerebral hemorrhage group were significantly higher than those in the control group ( t=12.598, t=13.003, χ 2=49.308, χ 2=10.246, χ 2=45.562, χ 2=8.602, P=0.000, 0.000, 0.000, 0.000, 0.000, 0.003). There were statistically significant differences in TC, TG, LDL and HDL among the control group, the younger group, the middle-aged group and the older group ( F=36.893, 131.263, 808.908, 143.188, all P=0.000). The highest genotype frequency was E3/3 in the control group, the younger group, the middle-aged group and the older group(70.00%, 68.33%, 68.33% and 66.67%). There was no statistically significant difference in genotype frequencies of ApoE among the groups (χ 2=2.025, 1.524, 0.693, 0.308, 0.815, 2.025, P=0.567, 0.677, 0.875, 0.959, 0.856, 0.567). The highest allele frequencies of ApoE in the control group, the younger group, the middle-aged group and the older group were epsilon 3 (81.66%, 81.66%, 79.99%, 80.83%). There was no statistically significant difference in the allele frequencies of ApoE among the groups (χ 2=0.061, 0.125, 0.241, P=0.996, 0.989, 0.971). For patients with E3/3 genotype cerebral hemorrhage, the levels of TC, TG, LDL and HDL in the younger age group were (4.43±0.33)mmol/L, (1.50±0.37)mmol/L, (2.58±0.32)mmol/L and (1.53±0.32)mmol/L, respectively.The TC, TG, LDL and HDL levels in middle-aged group were (4.22±0.34)mmol/L, (1.87±0.32)mmol/L, (3.30±0.31)mmol/L and (1.85±0.35)mmol/L, respectively.The TC, TG, LDL and HDL levels in the elderly group were (3.85±0.35)mmol/L, (2.09±0.35)mmol/L, (3.60±0.33)mmol/L and (2.13±0.34)mmol/L, respectively.There were statistically significant differences in TC, TG, LDL and HDL among all groups ( F=60.289, 59.844, 218.120, 64.364, all P=0.000). Conclusion:There is no significant correlation between ApoE gene polymorphism and the occurrence of widespread intracerebral hemorrhage in different age groups.There is no correlation between E3/3 genotype and blood lipid level in patients with intracerebral hemorrhage in different age groups.

Informatization plays an important role in the management of clinical diagnosis and treatment. However, due to the specialty of the discipline, the informatization construction of critical care medicine(CCM)is faced with such problems as the inefficient application of data, the low compliance of diagnosis and treatment operation, and the lack of intelligent quality control tools. The authors discussed the new mode of CCM information management based on data driven. By upgrading clinical information system, establishing single disease control system, introducing comprehensive intelligent analysis platform and building open remote platform, the bottleneck of CCM informatization was broken. The information collection and interaction in ICU was realized, the automatic monitoring and early warning of diagnosis and treatment process was realized, the operation of medical staff according to the guidelines was effectively improved, and the ability of diagnosis and treatment and management efficiency was improved.Furthermore, the homogenization of regional critical medical information could be promoted.

Objective: To study the contamination, serotype, pulsed field gel electrophoresis (PFGE) and drug resistance of listeria monocytogenes (L.monocytogenes) in the process of restaurant kitchens in Heilongjiang Province. Methods: Seventeen typical restaurants were selected from three cities in Heilongjiang Province in 2016, and 590 kitchen samples were collected and tested according to the national standard method. The serotype, pulsed field gel electrophoresis (PFGE) and drug resistance of isolated strains were analyzed. Results: L. monocytogenes was found in 104 of 590 of the samples analysed (17.63%). The isolates belong to six serotypes (1/2 a, 1/2 b, 1/2c, 3a, 3 b, 4 b) and self-condensing bacteria, and 57.38% (70 strains) of the strains belong to serotype 1/2b. Two highly pathogenic serotype 4b was detected for human listeria disease. The results of PFGE analysis show that the bacteria have cross-contamination in the environment, tools, equipment, food and personnel. The drug resistance results showed that 2 strains were resistant to tetracycline, 1 strain was resistant to erythromycin, 13 strains were intermediate to tetracycline, and 2 strains were resistant to tetracycline and erythromycin. Conclusion There is a certain degree of L. monocytogenes cross-contamination in the catering kitchen in Heilongjiang Province. And an important serotype 4b that can cause human Listeria disease was detected.

Objective: To evaluate the left ventricular systolic synchrony and investigate the early diagnostic value of left ventricular systolic dyssynchrony on cardiotoxicity caused by anthracyclines in patients with diffuse large B-cell lymphoma (DLBCL). Methods: Thirty-two patients (22 males, 10 females, age: 22-73(54.4±14.2) years) from June 2016 to January 2019 with confirmed DLBCL and normal gated myocardial perfusion imaging (GMPI) before anthracyclines chemotherapy were enrolled prospectively. GMPI was performed after 6 cycles or more of chemotherapy. Changes of myocardial markers, electrocardiogram (ECG) indicators, left ventricular function indicators including left ventricular ejection fraction (LVEF), left ventricular end-diastolic volume (LVEDV), left ventricular end-systolic volume (LVESV), peak filling rate (PFR), summed motion score (SMS) and summed thickening score (STS) as well as left ventricular systolic synchrony indicators including phase bandwidth (BW), phase standard deviation (SD) and entropy before and after anthracyclines chemotherapy were analyzed. Paired t test, Wilcoxon signed rank test and χ² test were used for data analysis. Results: Compared with pre-chemotherapy, the left ventricular systolic synchrony indicators were significantly higher than those before chemotherapy (BW: (42.81±11.37)° vs (29.28±8.68)°; SD: (11.65±4.64)° vs (8.79±3.14)°; entropy: (39.84±5.51)% vs (36.19±5.94)%; t values: -9.132 to -3.173, all P<0.05). There were no significant differences in other indicators (t values: -1.161 to 1.750, z values: -1.633 to -0.096, all P>0.05). Of 32 patients, 13 patients (40.62%) had left ventricular systolic dyssynchrony, and the rate of chemotherapy-induced left ventricular systolic dyssynchrony was significantly higher than that of left ventricular dysfunction (15.62%, 5/32; χ²=4.947, P=0.025). All 5 patients with left ventricular dysfunction caused by chemotherapy had left ventricular systolic dyssynchrony. The LVEF of the chemotherapy-induced left ventricular systolic dyssynchrony group was significantly lower than that of the left ventricular systolic synchronization group ((54.54±9.25)% vs (66.79±7.65)%; t=4.087, P<0.01). Conclusion Left ventricular systolic dyssynchrony can be appeared in DLBCL patients after chemotherapy and is significantly earlier than left ventricular dysfunction, which can be an early indicator of cardiotoxicity caused by anthracycline chemotherapy.