ObjectiveTo investigate the effects of Huanglian Jiedutang （HLJDT） on cognitive function in mice with ischemic stroke （IS） and to elucidate whether its neuroprotective effects are mediated by inhibition of the nuclear factor-κB （NF-κB） signaling pathway and subsequent suppression of NF-κB-regulated neuronal apoptosis. MethodsAn IS model was established using middle cerebral artery occlusion （MCAO）. Sixty C57BL/6J mice were randomly assigned to five groups （n =12 per group）， i.e.， sham operation， model， HLJDT low-dose （3.9 g·kg^-1·d^-1）， HLJDT high-dose （7.8 g·kg^-1·d^-1）， and Ginkgo biloba extract （GBE， 31.2 mg·kg^-1·d^-1）. Post-operatively， neurological deficit scores （Longa score）， cerebral infarct volume assessed by 2，3，5-triphenyltetrazolium chloride （TTC） staining， and brain water content were evaluated. Learning and memory were assessed using new object recognition （NOR） and fear conditioning （FC） tests. Hippocampal pathology was examined via hematoxylin and eosin （HE） staining. Immunofluorescence detected expression of glial fibrillary acidic protein （GFAP， astrocyte marker）， cellular oncogene Fos （c-Fos， neuronal activation marker）， and glutamate decarboxylase 65 （GAD65）. Western blot measured nuclear factor-κB inhibitor protein α （IκBα）， phosphorylated IκBα （p-IκBα）， NF-κB p65， phosphorylated NF-κB p65 （p-NF-κB p65）， ionic calcium binding adapter molecule 1 （Iba-1）， tumor necrosis factor （TNF）-α， interleukin （IL）-1β， and apoptosis-related proteins， such as cleaved cysteinyl aspartate-specific protease 3 （Caspase-3）， B-cell lymphoma 2 （Bcl-2）， and Bcl-2-associated X protein （Bax）. Real-time quantitative PCR （Real-time PCR） was used to assess mRNA levels of Iba-1， TNF-α， IL-1β， NF-κB p65， cleaved Caspase-3， Bax， and Bcl-2. ResultsCompared with the sham group， the model group exhibited significantly increased neurological deficit scores， brain water content， and cerebral infarct volume （P<0.01）. Hippocampal CA1 neurons were disorganized， showing nuclear pyknosis and karyolysis. NOR exploration time and FC freezing time were significantly reduced （P<0.01）. GFAP and c-Fos expression were increased， while GAD65 expression was decreased （P<0.01）. Cleaved Caspase-3 and Bax were upregulated， Bcl-2 was downregulated， and the Bax/Bcl-2 ratio was elevated （P<0.01）. Expression levels of p-IκBα， p-NF-κB p65， IL-1β， TNF-α， and Iba-1 were significantly increased （P<0.01）. Compared with the model group， HLJDT high-dose， low-dose， and GBE groups showed significant improvements in all parameters （P<0.01）. Among them， the HLJDT high-dose group showed the most pronounced neuronal structural recovery and superior performance in NOR and FC tests （P<0.01）. In this group， GFAP and c-Fos decreased， GAD65 increased （P<0.01）， apoptosis-related protein expression was reversed， and NF-κB signaling and related inflammatory factor expression were suppressed （P<0.01）. ConclusionHLJDT ameliorates cognitive dysfunction in mice after IS， potentially by inhibiting the NF-κB signaling pathway， thereby reducing neuroinflammation and hippocampal neuronal apoptosis.

Objective:To investigate the relationship between the serum levels of vascular endothelial growth factor (VEGF), matrix metalloproteinase-9 (MMP-9), S100 calcium binding protein B (S100B) with glycolipid metabolism, pregnancy outcome in pregnant women with gestational diabetes.Methods:The clinical data of 153 pregnant women with gestational diabetes (research group) and 153 healthy pregnant women (control group) in the Second Affiliated Hospital of Xi ′an Medical University from January 2020 to October 2023 were retrospectively analyzed. The serum levels of VEGF, MMP-9 and S100B were measured by enzyme linked immunosorbent assay, and the fasting blood glucose, triglyceride, total cholesterol, high-density lipoprotein cholesterol (HDL-C), low-density lipoprotein cholesterol (LDL-C), fasting insulin and glycated hemoglobin were measured, and the homeostasis model assessment insulin resistance index (HOMA-IR) was calculated. The adverse outcomes of pregnant women with gestational diabetes were recorded. Pearson method was used to analyze the correlation between glycolipid metabolism indexes and VEGF, MMP-9, S100B in pregnant women with gestational diabetes. Multivariate Logistic regression was used to analyze the independent risk factors of adverse pregnancy outcome in pregnant women with gestational diabetes. Receiver operating characteristic (ROC) curve was drawn to analyze the predictive value of VEGF, MMP-9 and S100B on adverse pregnancy outcome in pregnant women with gestational diabetes. Results:The fasting blood glucose, fasting insulin, glycated hemoglobin, HOMA-IR, triglyceride, total cholesterol, LDL-C, VEGF, MMP-9 and S100B in research group were significantly higher than those in control group: (9.42 ± 0.65) mmol/L vs. (4.13 ± 0.46) mmol/L, (16.58 ± 2.37) mU/L vs. (13.41 ± 2.05) mU/L, (7.28 ± 0.46)% vs. (4.35 ± 0.39)%, 4.83 ± 0.42 vs. 2.71 ± 0.37, (3.41 ± 0.67) mmol/L vs. (2.85 ± 0.63) mmol/L, (5.54 ± 1.56) mmol/L vs. (5.12 ± 1.50) mmol/L, (3.14 ± 0.97) mmol/L vs. (2.86 ± 0.93) mmol/L, (184.02 ± 30.25) ng/L vs. (156.33 ± 26.41) ng/L, (45.78 ± 7.56) μg/L vs. (29.36 ± 5.03) μg/L and (117.51 ± 25.12) ng/L vs. (89.74 ± 22.46) ng/L, the HDL-C was significantly lower than that in control group: (1.34 ± 0.27) mmol/L vs. (1.42 ± 0.30) mmol/L, and there were statistical differences ( P<0.01 or <0.05). Pearson correlation analysis result showed that VEGF, MMP-9, S100B in pregnant women with gestational diabetes were positively correlated with fasting blood glucose, fasting insulin, glycated hemoglobin, HOMA-IR, triglyceride, total cholesterol and LDL-C ( P<0.01), negatively correlated with HDL-C ( P<0.01). Among 153 pregnant women with gestational diabetes, 49 had adverse pregnancy outcome, and 104 had good pregnancy outcome. The VEGF, MMP-9 and S100B in pregnant women with adverse pregnancy outcome were significantly higher than those in pregnant women with good pregnancy outcome: (212.75 ± 28.63) ng/L vs. (170.49 ± 26.58) ng/L, (52.37 ± 7.14) μg/L vs. (42.68 ± 6.35) μg/L and (136.83 ± 23.62) ng/L vs. (108.41 ± 21.35) ng/L, and there were statistical differences ( P<0.01). Multivariate Logistic regression analysis result showed that VEGF, MMP-9 and S100B were independent risk factors for adverse pregnancy outcome in pregnant women with gestational diabetes ( OR = 7.013, 5.382 and 6.129; 95% CI 5.206 to 9.447, 3.449 to 8.398 and 3.520 to 10.673; P<0.01). ROC curve analysis result showed that the area under the curve of VEGF, MMP-9 combined S100B in predicting adverse pregnancy outcome in pregnant women with gestational diabetes was significantly larger than that of VEGF, MMP-9 and S100B alone (0.945 vs. 0.863, 0.847 and 0.801; P<0.05 or <0.01), with sensitivity of 89.80% and specificity of 91.30%. Conclusions:The high serum levels of VEGF, MMP-9 and S100B are associated with abnormal glycolipid metabolism and adverse pregnancy outcome in pregnant women with gestational diabetes, and the combination of the three indexes has a high predictive value for adverse pregnancy outcome.

Objective:To investigate the clinical value of enhanced magnetic resonance imaging (MRI)-based deep learning model in preoperative prediction of proliferative hepatocellular carcinoma (HCC).Methods:The retrospective cohort study was conducted. The clinical data of 906 HCC patients who were admitted to The First Affiliated Hospital of Army Medical University and The Second Affiliated Hospital of Chongqing Medical University from May 2017 to October 2022 were collected. There were 769 males and 137 females, aged (53.2±10.9)years. Of the 906 patients, 815 cases who were admitted to The First Affiliated Hospital of Army Medical University were divided into the training set of 634 patients and the internal validation set of 181 patients using a random number table method with a ratio of 8:2, and 91 patients who were admitted to The Second Affiliated Hospital of Chongqing Medical University were divided into the external validation set. The training set was used to construct the prediction model, while the validation set was used to validate the prediction model. Observation indicators: (1) analysis of factors influencing the pathological classification of HCC patients; (2) deep learning imaging features of HCC patients; (3) evaluation of the efficacy of prediction model for proliferative HCC; (4) validation of the prediction model for proliferative HCC; (5) prognosis of HCC patients. Comparison of measurement data with normal distribution between groups was conducted using the independent sample t test. Comparison of measurement data with skewed distribution between groups was conducted using the Mann-Whitney U test. Comparison of count data between groups was conducted using the chi-square test. Multivariate analysis was conducted using the binary Logistic regression model. The model perfor-mance was evaluated through five-fold cross-validation, and receiver operating characteristic (ROC) curve was plotted to assess the diagnostic value of the model based on the area under curve (AUC), sensitivity, and specificity. The Delong test was used to compare the diagnostic performance of models. The Hosmer-Lemeshow test was employed to evaluate the calibration of models. The optimal cutoff value of the prediction model was determined by the maximum Youden index, with the value >0.175 indicating high-risk patients and value ≤0.175 indicating low-risk patients.The Kaplan-Meier method was used to calculate the survival rate and the Log-rank test was used for survival analysis. Results:(1) Analysis of factors influencing the pathological classification of HCC patients. Of 634 patients in the training set, there were 190 cases of proliferative HCC and 444 cases of non-proliferative HCC. Results of multivariate analysis showed that alpha fetoprotein (AFP) ≥400 μg/L and tumor diameter >5 cm were independent risk factors for pathological type of HCC as proli-ferative [ odds ratio=1.73, 1.88, 95% confidence interval ( CI) as 1.19-2.50, 1.30-2.71, P<0.05]. (2) Deep learning imaging features of HCC patients. In the training set of 634 patients, the probability predicted by MRI-based deep learning model was 84.8%(30.5%,95.4%) for proliferative HCC and 5.8%(3.2%,12.5%) for non-proliferative HCC, showing a significant difference between them ( Z=-16.01, P<0.05). (3) Evaluation of the efficacy of prediction model for proliferative HCC. In the training set, the AUC of clinical prediction model for proliferative HCC was 0.63(95% CI as 0.59-0.68, P<0.05), with sensitivity of 54.74% and specificity of 64.19%. The AUC of MRI-based deep learning prediction model was 0.90(95% CI as 0.87-0.93, P<0.05), with sensitivity of 80.53% and specificity of 86.94%. The AUC of combined MRI-based deep learning with clinical prediction model was 0.90 (95% CI as 0.87-0.93, P<0.05), with sensitivity of 83.16% and specificity of 86.04%. Results of Delong test showed that there was a significant difference between the combined MRI-based deep learning with clinical prediction model and the clinical prediction model ( P<0.05), and there was no signifi-cant difference between the combined MRI-based deep learning with clinical prediction model and the MRI-based deep learning prediction model ( P>0.05). Results of Hosmer-Lemeshow test showed good calibration for the clinical prediction model, the MRI-based deep learning prediction model and the combined MRI-based deep learning with clinical prediction model ( χ2=0.84, 6.38, 3.93, P>0.05), indicating that the predicted probabilities of these three prediction models matched the actual risk well. (4) Validation of the prediction model for proliferative HCC. Results of validation of the prediction model in internal validation set showed the AUC of MRI-based deep learning prediction model for proliferative HCC was 0.84(95% CI as 0.77-0.91, P<0.05), with sensitivity of 82.35% and specificity of 77.69%. Results of validation of the prediction model in external validation set showed the AUC of MRI-based deep learning prediction model for proliferative HCC was 0.81(95% CI as 0.71-0.92, P<0.05), with sensitivity of 70.00% and specificity of 81.69%. (5) Prognosis of HCC patients. Of the 906 patients, the 1-, 3-, and 5-year recurrence-free survival rates for 645 proliferative HCC patients were 56.9%, 31.4%, and 29.1%, respectively, and the 1-, 3-, and 5-year recurrence-free survival rates for 261 non-proliferative HCC patients were 88.8%, 68.6%, and 56.0%, respectively. There were significant differences in recurrence-free survival time between proliferative HCC and non-proliferative HCC patients of the training set, internal validation set and external validation set ( P<0.05). The 1-, 3-, 5-year recurrence-free survival rates for 331 high-risk HCC patients were 64.6%, 50.4%, 43.6%, versus 88.5%, 71.9%, 62.7% for 575 low-risk HCC patients. There were significant differences in recurrence-free survival time between high-risk HCC patients and low-risk HCC patients of the training set, internal validation set and external validation set ( P<0.05). Conclusion:The MRI-based deep learning model can effectively predict proliferative HCC and recurrence-free survival of patients before the surgery.

Objective:To evaluate preoperative high-resolution thin-layer cervical enhanced CT used to predict the venous route of the submental flap reflux vein and its relationship with adjacent structures in order to guide the anatomical understanding and protection of submental flap in pharyngeal cancer surgery.Methods:Sixty consecutive patients with pharyngeal cancer who underwent submental flap repair surgery in our department from March 2022 to December 2024, as well as 60 patients who were accepted neck dissection suffering other cancers, were selected. Before surgery, high-resolution cervical enhanced CT scans were performed, and the position of the transverse section of the facial vein in the venous phase horizontal image gradually variation tendency was focused layer by layer. The direction and adjacent relationship of the submental flap reflux veins were determined and recorded. Combined with 60 patients with other head and neck tumors who underwent neck dissection in our department during the same period (a total of 120 cases, 240 sides), the classification and management of the draining veins of Fang′s mental flap were conducted. Type Ⅰ mainly drains into the internal jugular vein; Type Ⅱ mainly drains into the external jugular vein and Type Ⅲ mainly drains into the anterior jugular vein (often accompanied by an external jugular draining branch). The status and proportion of venous drainage were analyzed.Results:Vascular predictive coincidence rate was 98.3% (59/60) among the 60 patients with pharyngeal cancer. Only one patient was predicted to have a simple return to the external jugular vein. However, during the operation, in addition to the main return to the external jugular vein, a small portion also returned to the internal jugular vein. Submental flap reflux vessels were classified into three types based on intraoperative submental flap venous return in 60 cases of laryngopharyngeal cancer, in conjunction with the analysis of venous return patterns from 240 cervical CT scans. Type Ⅰ mainly refluxed to the internal jugular vein, accounting for 42.1%. Type Ⅱ mainly refluxed to the external jugular vein (47.9%). Type Ⅲ mainly refluxed to the anterior jugular vein (10.0%). The total detection rate of CT reading of 240 venous reflux was 98.7% (237/240). Vascular predictive coincidence rate was 97.9%(235/240).Conclusion:The detailed analysis of submental venous return vessels can accurately predict the direction of reflux veins and its surrounding areas by preoperative high-resolution enhanced CT scan. This provides reliable guidance for the anatomy and protection of the submental flap reflux veins during surgery.

Objective:To compare the survival outcomes of different subsequent treatment regimens in patients with locally advanced hypopharyngeal squamous cell carcinoma (HPSCC) who achieved partial response (PR) after neoadjuvant chemotherapy based on the gross tumor volume regression rate (GTVRR).Methods:This retrospective study included patients with locally advanced HPSCC treated at the Department of Head and Neck Surgery, Beijing Tongren Hospital, from January 2011 to December 2023. The cohort included 135 males and 3 females, aged from 35 to 77 years. All patients received 2-3 cycles of TPF regimen (paclitaxel+cisplatin+5-fluorouracil) neoadjuvant chemotherapy. Subsequent treatments included concurrent chemoradiotherapy or surgery combined with postoperative adjuvant radiotherapy. The impacts of different subsequent treatment modalities on the survivals and prognoses of patients were compared based on GTVRR thresholds of 50% and 70%. The χ 2 test was used to analyze influencing factors; survival analysis and intergroup comparisons were performed using the Kaplan-Meier method and Log-rank test; prognostic factors were assessed using univariate and multivariate Cox regression analyses. Results:The 5-year OS and PFS rates were 56.5% and 47.9%, respectively, while, the 10-year OS and PFS rates were 25.8% and 21.2%, respectively. The median OS was 75 months, and the median PFS was 48 months. The laryngeal function preservation rate for the entire cohort was 83.3%. The patients who underwent surgery combined with postoperative radiotherapy had significantly better OS and PFS outcomes than those treated with concurrent chemoradiotherapy ( P<0.05). Stratification based on GTVRR revealed that the surgery plus postoperative radiotherapy regimen was particularly effective for PR patients with a GTVRR of 30%-70%, showing significantly better OS and PFS compared to the concurrent chemoradiotherapy group ( P<0.05). Conclusion:The optimal subsequent treatment for PR-HPSCC may be surgery-based comprehensive treatment, particularly for patients with a GTVRR of 30%-70%. This study offers valuable insights for the stratified treatment of HPSCC, which could contribute to improving overall patient prognosis.

Objective To provide methodological examples for related research,the influencing factors of the evolution of Qi deficiency syndrome in chronic obstructive pulmonary disease(COPD)based on a nonlinear mixed effects model was explored.Methods A research questionnaire on the influencing factors of the evolution of Qi deficiency syndrome in chronic obstructive pulmonary disease was developed,and clinical data of 650 COPD patients on the 1st and 14th day of acute exacerbation,the 1st and 28th day of risk window,the first day of stable period,and the 90th day were dynamically collected from 10 tertiary hospitals across the country.8 baseline data including gender and age were collected through the PROC NLMIXED process by SAS 9.4 software.Coronary heart disease,diabetes and hypertension accounted for the highest proportion.Nine concurrent syndromes including wind cold syndrome and phlegm heat syndrome were used as fixed effects,and individual level was used as random effects to gradually fit the model and screen the influencing factors of Qi deficiency syndrome in the entire process of disease occurrence and development.Results A total of 637 eligible cases were included,and clinical datas were dynamically collected on the 1st and 14th day of acute exacerbation,the 1st and 28th day of the risk window,the 1st and 90th day of the stable period.It was found that the number of acute exacerbations,alcohol consumption,concomitant hypertension,coronary heart disease,blood stasis syndrome,yin deficiency syndrome,yang deficiency syndrome,6-minute walking distance,and the modified Medical Research Council Dyspnea Questionnaire(mMRC)had an impact on the evolution of Qi deficiency syndrome in the previous year(P<0.05).Conclusion The use of a nonlinear mixed effects model revealed the relevant factors affecting the evolution of Qi deficiency syndrome from complex multi temporal dynamic data,providing methodological references for other related studies.

Objective To provide methodological examples for related research,the influencing factors of the evolution of Qi deficiency syndrome in chronic obstructive pulmonary disease(COPD)based on a nonlinear mixed effects model was explored.Methods A research questionnaire on the influencing factors of the evolution of Qi deficiency syndrome in chronic obstructive pulmonary disease was developed,and clinical data of 650 COPD patients on the 1st and 14th day of acute exacerbation,the 1st and 28th day of risk window,the first day of stable period,and the 90th day were dynamically collected from 10 tertiary hospitals across the country.8 baseline data including gender and age were collected through the PROC NLMIXED process by SAS 9.4 software.Coronary heart disease,diabetes and hypertension accounted for the highest proportion.Nine concurrent syndromes including wind cold syndrome and phlegm heat syndrome were used as fixed effects,and individual level was used as random effects to gradually fit the model and screen the influencing factors of Qi deficiency syndrome in the entire process of disease occurrence and development.Results A total of 637 eligible cases were included,and clinical datas were dynamically collected on the 1st and 14th day of acute exacerbation,the 1st and 28th day of the risk window,the 1st and 90th day of the stable period.It was found that the number of acute exacerbations,alcohol consumption,concomitant hypertension,coronary heart disease,blood stasis syndrome,yin deficiency syndrome,yang deficiency syndrome,6-minute walking distance,and the modified Medical Research Council Dyspnea Questionnaire(mMRC)had an impact on the evolution of Qi deficiency syndrome in the previous year(P<0.05).Conclusion The use of a nonlinear mixed effects model revealed the relevant factors affecting the evolution of Qi deficiency syndrome from complex multi temporal dynamic data,providing methodological references for other related studies.

Objective This study aims to explore effective approaches for the transfer of medical scientific and techno-logical achievements to promote the development of research-oriented hospitals.Methods The technology transfer achievements of an affiliated hospital of a university in Guangdong Province over the past six years(2018-2023)were statistically analyzed.The challenges faced during the transfer process,the measures taken,and the current achievements were discussed.Results The number of patent authorizations and authorized departments in the hospital has increased year by year.The transfer rate has risen from 0％in 2019 to 6.69％in 2023.Currently,46 projects have been successfully transferred,with a total transfer amount exceeding 30 million yuan,indicating significant effectiveness in the transfer of medical scientific and technological achievements.Conclusion The hospital attaches great importance to and overall manages the transfer process,establishing a sound manage-ment structure,improving incentive and support systems,and regularly conducting special lectures,training,and guidance.These efforts guide researchers to start from clinical problems and ultimately serve clinical diagnosis and treatment,creating a fa-vorable environment for technology transfer,improving the transfer rate,and promoting the development of research-oriented hos-pitals.

Objective:To investigate the relationship between the serum levels of vascular endothelial growth factor (VEGF), matrix metalloproteinase-9 (MMP-9), S100 calcium binding protein B (S100B) with glycolipid metabolism, pregnancy outcome in pregnant women with gestational diabetes.Methods:The clinical data of 153 pregnant women with gestational diabetes (research group) and 153 healthy pregnant women (control group) in the Second Affiliated Hospital of Xi ′an Medical University from January 2020 to October 2023 were retrospectively analyzed. The serum levels of VEGF, MMP-9 and S100B were measured by enzyme linked immunosorbent assay, and the fasting blood glucose, triglyceride, total cholesterol, high-density lipoprotein cholesterol (HDL-C), low-density lipoprotein cholesterol (LDL-C), fasting insulin and glycated hemoglobin were measured, and the homeostasis model assessment insulin resistance index (HOMA-IR) was calculated. The adverse outcomes of pregnant women with gestational diabetes were recorded. Pearson method was used to analyze the correlation between glycolipid metabolism indexes and VEGF, MMP-9, S100B in pregnant women with gestational diabetes. Multivariate Logistic regression was used to analyze the independent risk factors of adverse pregnancy outcome in pregnant women with gestational diabetes. Receiver operating characteristic (ROC) curve was drawn to analyze the predictive value of VEGF, MMP-9 and S100B on adverse pregnancy outcome in pregnant women with gestational diabetes. Results:The fasting blood glucose, fasting insulin, glycated hemoglobin, HOMA-IR, triglyceride, total cholesterol, LDL-C, VEGF, MMP-9 and S100B in research group were significantly higher than those in control group: (9.42 ± 0.65) mmol/L vs. (4.13 ± 0.46) mmol/L, (16.58 ± 2.37) mU/L vs. (13.41 ± 2.05) mU/L, (7.28 ± 0.46)% vs. (4.35 ± 0.39)%, 4.83 ± 0.42 vs. 2.71 ± 0.37, (3.41 ± 0.67) mmol/L vs. (2.85 ± 0.63) mmol/L, (5.54 ± 1.56) mmol/L vs. (5.12 ± 1.50) mmol/L, (3.14 ± 0.97) mmol/L vs. (2.86 ± 0.93) mmol/L, (184.02 ± 30.25) ng/L vs. (156.33 ± 26.41) ng/L, (45.78 ± 7.56) μg/L vs. (29.36 ± 5.03) μg/L and (117.51 ± 25.12) ng/L vs. (89.74 ± 22.46) ng/L, the HDL-C was significantly lower than that in control group: (1.34 ± 0.27) mmol/L vs. (1.42 ± 0.30) mmol/L, and there were statistical differences ( P<0.01 or <0.05). Pearson correlation analysis result showed that VEGF, MMP-9, S100B in pregnant women with gestational diabetes were positively correlated with fasting blood glucose, fasting insulin, glycated hemoglobin, HOMA-IR, triglyceride, total cholesterol and LDL-C ( P<0.01), negatively correlated with HDL-C ( P<0.01). Among 153 pregnant women with gestational diabetes, 49 had adverse pregnancy outcome, and 104 had good pregnancy outcome. The VEGF, MMP-9 and S100B in pregnant women with adverse pregnancy outcome were significantly higher than those in pregnant women with good pregnancy outcome: (212.75 ± 28.63) ng/L vs. (170.49 ± 26.58) ng/L, (52.37 ± 7.14) μg/L vs. (42.68 ± 6.35) μg/L and (136.83 ± 23.62) ng/L vs. (108.41 ± 21.35) ng/L, and there were statistical differences ( P<0.01). Multivariate Logistic regression analysis result showed that VEGF, MMP-9 and S100B were independent risk factors for adverse pregnancy outcome in pregnant women with gestational diabetes ( OR = 7.013, 5.382 and 6.129; 95% CI 5.206 to 9.447, 3.449 to 8.398 and 3.520 to 10.673; P<0.01). ROC curve analysis result showed that the area under the curve of VEGF, MMP-9 combined S100B in predicting adverse pregnancy outcome in pregnant women with gestational diabetes was significantly larger than that of VEGF, MMP-9 and S100B alone (0.945 vs. 0.863, 0.847 and 0.801; P<0.05 or <0.01), with sensitivity of 89.80% and specificity of 91.30%. Conclusions:The high serum levels of VEGF, MMP-9 and S100B are associated with abnormal glycolipid metabolism and adverse pregnancy outcome in pregnant women with gestational diabetes, and the combination of the three indexes has a high predictive value for adverse pregnancy outcome.

Objective:To investigate the clinical value of enhanced magnetic resonance imaging (MRI)-based deep learning model in preoperative prediction of proliferative hepatocellular carcinoma (HCC).Methods:The retrospective cohort study was conducted. The clinical data of 906 HCC patients who were admitted to The First Affiliated Hospital of Army Medical University and The Second Affiliated Hospital of Chongqing Medical University from May 2017 to October 2022 were collected. There were 769 males and 137 females, aged (53.2±10.9)years. Of the 906 patients, 815 cases who were admitted to The First Affiliated Hospital of Army Medical University were divided into the training set of 634 patients and the internal validation set of 181 patients using a random number table method with a ratio of 8:2, and 91 patients who were admitted to The Second Affiliated Hospital of Chongqing Medical University were divided into the external validation set. The training set was used to construct the prediction model, while the validation set was used to validate the prediction model. Observation indicators: (1) analysis of factors influencing the pathological classification of HCC patients; (2) deep learning imaging features of HCC patients; (3) evaluation of the efficacy of prediction model for proliferative HCC; (4) validation of the prediction model for proliferative HCC; (5) prognosis of HCC patients. Comparison of measurement data with normal distribution between groups was conducted using the independent sample t test. Comparison of measurement data with skewed distribution between groups was conducted using the Mann-Whitney U test. Comparison of count data between groups was conducted using the chi-square test. Multivariate analysis was conducted using the binary Logistic regression model. The model perfor-mance was evaluated through five-fold cross-validation, and receiver operating characteristic (ROC) curve was plotted to assess the diagnostic value of the model based on the area under curve (AUC), sensitivity, and specificity. The Delong test was used to compare the diagnostic performance of models. The Hosmer-Lemeshow test was employed to evaluate the calibration of models. The optimal cutoff value of the prediction model was determined by the maximum Youden index, with the value >0.175 indicating high-risk patients and value ≤0.175 indicating low-risk patients.The Kaplan-Meier method was used to calculate the survival rate and the Log-rank test was used for survival analysis. Results:(1) Analysis of factors influencing the pathological classification of HCC patients. Of 634 patients in the training set, there were 190 cases of proliferative HCC and 444 cases of non-proliferative HCC. Results of multivariate analysis showed that alpha fetoprotein (AFP) ≥400 μg/L and tumor diameter >5 cm were independent risk factors for pathological type of HCC as proli-ferative [ odds ratio=1.73, 1.88, 95% confidence interval ( CI) as 1.19-2.50, 1.30-2.71, P<0.05]. (2) Deep learning imaging features of HCC patients. In the training set of 634 patients, the probability predicted by MRI-based deep learning model was 84.8%(30.5%,95.4%) for proliferative HCC and 5.8%(3.2%,12.5%) for non-proliferative HCC, showing a significant difference between them ( Z=-16.01, P<0.05). (3) Evaluation of the efficacy of prediction model for proliferative HCC. In the training set, the AUC of clinical prediction model for proliferative HCC was 0.63(95% CI as 0.59-0.68, P<0.05), with sensitivity of 54.74% and specificity of 64.19%. The AUC of MRI-based deep learning prediction model was 0.90(95% CI as 0.87-0.93, P<0.05), with sensitivity of 80.53% and specificity of 86.94%. The AUC of combined MRI-based deep learning with clinical prediction model was 0.90 (95% CI as 0.87-0.93, P<0.05), with sensitivity of 83.16% and specificity of 86.04%. Results of Delong test showed that there was a significant difference between the combined MRI-based deep learning with clinical prediction model and the clinical prediction model ( P<0.05), and there was no signifi-cant difference between the combined MRI-based deep learning with clinical prediction model and the MRI-based deep learning prediction model ( P>0.05). Results of Hosmer-Lemeshow test showed good calibration for the clinical prediction model, the MRI-based deep learning prediction model and the combined MRI-based deep learning with clinical prediction model ( χ2=0.84, 6.38, 3.93, P>0.05), indicating that the predicted probabilities of these three prediction models matched the actual risk well. (4) Validation of the prediction model for proliferative HCC. Results of validation of the prediction model in internal validation set showed the AUC of MRI-based deep learning prediction model for proliferative HCC was 0.84(95% CI as 0.77-0.91, P<0.05), with sensitivity of 82.35% and specificity of 77.69%. Results of validation of the prediction model in external validation set showed the AUC of MRI-based deep learning prediction model for proliferative HCC was 0.81(95% CI as 0.71-0.92, P<0.05), with sensitivity of 70.00% and specificity of 81.69%. (5) Prognosis of HCC patients. Of the 906 patients, the 1-, 3-, and 5-year recurrence-free survival rates for 645 proliferative HCC patients were 56.9%, 31.4%, and 29.1%, respectively, and the 1-, 3-, and 5-year recurrence-free survival rates for 261 non-proliferative HCC patients were 88.8%, 68.6%, and 56.0%, respectively. There were significant differences in recurrence-free survival time between proliferative HCC and non-proliferative HCC patients of the training set, internal validation set and external validation set ( P<0.05). The 1-, 3-, 5-year recurrence-free survival rates for 331 high-risk HCC patients were 64.6%, 50.4%, 43.6%, versus 88.5%, 71.9%, 62.7% for 575 low-risk HCC patients. There were significant differences in recurrence-free survival time between high-risk HCC patients and low-risk HCC patients of the training set, internal validation set and external validation set ( P<0.05). Conclusion:The MRI-based deep learning model can effectively predict proliferative HCC and recurrence-free survival of patients before the surgery.