Objective:To develop and validate a machine learning (ML) noninvasive model based on routine laboratory parameters to preoperatively predict the microvascular invasion (MVI) in patients with hepatocellular carcinoma (HCC).Methods:A total of 629 HCC patients who underwent hepatectomy at the First Affiliated Hospital of Fujian Medical University between January 2019 and December 2023 were retrospectively enrolled in this study and were divided chronologically into a training set ( n=464) and internal validation set ( n=165). A cohort with 190 HCC patients from Fujian Provincial Hospital were used as an external validation set. Preoperatively demographic features, tumor size and routine laboratory data were collected. All patients were divided into MVI-positive or MVI-negative group. The Boruta algorithm and LASSO regression algorithm were used to screen out related features in the training set. Eight different ML algorithms including multivariate logistic regression, decision tree (DT), random forest (RF), extreme gradient boosting (XGboost), k-nearest neighbor (KNN), support vector machine (SVM), light gradient boosting machine (LGBM) and Naive Bayes were used to construct the prediction models. The predictive performances of these models on training and internal validation sets were evaluated by the receiver operating characteristic (ROC) curve with the area under the curve (AUC). The ML model with the highest AUC values was defined as the optimal model and its performance was further validated in the external validation set. The calibration curve showed that the probability value curve was close to the actual occurrence probability curve, and the DCA showed that it could be applied within the threshold probability range of 0.3-0.8 to obtain net benefits. Results:After screening, eight parameters including α-fetoprotein (AFP), protein induced by vitamin K absence Ⅱ (PIVKA-Ⅱ), tumor size, eosinophil count, neutrophil count, creatinine, ApoA1 and total bilirubin were finally selected for the construction of the preoperative prediction model for MVI in HCC. Among all the tested eight ML algorithms, XGboost obtained the optimal performance with an AUC of 0.820 in training set, an AUC of 0.803 in internal testing set and an AUC of 0.758 in external testing set. Further stratified analysis showed that the AUC for preoperatively predicting MVI by XGboost was 0.817 for HCC patients with positive hepatitis B surface antigen, 0.779 for male patients and 0.790 for elder patients. The calibration curves showed good agreement between observed and predicted values and the decision curve analysis curve showed relatively higher net benefits.Conclusions:We successfully established and verified a novel XGboost model based on eight routine laboratory parameters with relatively high and reliable predictive accuracy to preoperatively predict MVI in HCC.

Objective:To investigate the effect of growth stimulation expressed gene 2(ST2)on intestinal microflora of mice with ConA-induced autoimmune hepatitis(AIH)by knockout.Methods:ST2 gene knockout mice were constructed by gene knockout technique.On this basis,AIH mouse model was established by ConA induction.The relative expression of cytokines was detected by real-time fluorescence quantitative PCR(RT-PCR).Intestinal microorganisms were sequenced using the 16S rRNA high-through put sequencing method,and the sequencing results were analyzed using Microbial Ecology related software and database.Results:Com-pared with the control group,ALT and AST were decreased,the damage of liver was mild,and the relative expression of IL-6 was also decreased in the liver of mice with ConA-induced AIH after ST2 gene knockout.There was no significant difference in the ACE and CHAO indices of the abundance of reactive flora.There was no significant difference between Shannon index and Simpson index re-flecting bacterial diversity.Psychrobacter,unclassified_Clostridia,Halothiobacillus,Clostridium_XlVa and Haemophilus genus content increased;Romboutsia,Rikenella,Parabacteroides and unclassified_Clostridiales bacterial content reduce.The areas under the receiv-er operator characteristic(ROC)curves were 0.88,0.89,0.88,0.80,0.90,0.90,0.84,0.91 and 0.84,respectively.Conclusion:Knockout of ST2 gene expression in mice with AIH can reduce liver injury and inflammation,and does not affect the distribution abun-dance and diversity of intestinal flora.However,there are differences among bacteria genera,and it has good diagnostic value.

Objective:To investigate the effect of growth stimulation expressed gene 2(ST2)on intestinal microflora of mice with ConA-induced autoimmune hepatitis(AIH)by knockout.Methods:ST2 gene knockout mice were constructed by gene knockout technique.On this basis,AIH mouse model was established by ConA induction.The relative expression of cytokines was detected by real-time fluorescence quantitative PCR(RT-PCR).Intestinal microorganisms were sequenced using the 16S rRNA high-through put sequencing method,and the sequencing results were analyzed using Microbial Ecology related software and database.Results:Com-pared with the control group,ALT and AST were decreased,the damage of liver was mild,and the relative expression of IL-6 was also decreased in the liver of mice with ConA-induced AIH after ST2 gene knockout.There was no significant difference in the ACE and CHAO indices of the abundance of reactive flora.There was no significant difference between Shannon index and Simpson index re-flecting bacterial diversity.Psychrobacter,unclassified_Clostridia,Halothiobacillus,Clostridium_XlVa and Haemophilus genus content increased;Romboutsia,Rikenella,Parabacteroides and unclassified_Clostridiales bacterial content reduce.The areas under the receiv-er operator characteristic(ROC)curves were 0.88,0.89,0.88,0.80,0.90,0.90,0.84,0.91 and 0.84,respectively.Conclusion:Knockout of ST2 gene expression in mice with AIH can reduce liver injury and inflammation,and does not affect the distribution abun-dance and diversity of intestinal flora.However,there are differences among bacteria genera,and it has good diagnostic value.

Objective:To develop and validate a machine learning (ML) noninvasive model based on routine laboratory parameters to preoperatively predict the microvascular invasion (MVI) in patients with hepatocellular carcinoma (HCC).Methods:A total of 629 HCC patients who underwent hepatectomy at the First Affiliated Hospital of Fujian Medical University between January 2019 and December 2023 were retrospectively enrolled in this study and were divided chronologically into a training set ( n=464) and internal validation set ( n=165). A cohort with 190 HCC patients from Fujian Provincial Hospital were used as an external validation set. Preoperatively demographic features, tumor size and routine laboratory data were collected. All patients were divided into MVI-positive or MVI-negative group. The Boruta algorithm and LASSO regression algorithm were used to screen out related features in the training set. Eight different ML algorithms including multivariate logistic regression, decision tree (DT), random forest (RF), extreme gradient boosting (XGboost), k-nearest neighbor (KNN), support vector machine (SVM), light gradient boosting machine (LGBM) and Naive Bayes were used to construct the prediction models. The predictive performances of these models on training and internal validation sets were evaluated by the receiver operating characteristic (ROC) curve with the area under the curve (AUC). The ML model with the highest AUC values was defined as the optimal model and its performance was further validated in the external validation set. The calibration curve showed that the probability value curve was close to the actual occurrence probability curve, and the DCA showed that it could be applied within the threshold probability range of 0.3-0.8 to obtain net benefits. Results:After screening, eight parameters including α-fetoprotein (AFP), protein induced by vitamin K absence Ⅱ (PIVKA-Ⅱ), tumor size, eosinophil count, neutrophil count, creatinine, ApoA1 and total bilirubin were finally selected for the construction of the preoperative prediction model for MVI in HCC. Among all the tested eight ML algorithms, XGboost obtained the optimal performance with an AUC of 0.820 in training set, an AUC of 0.803 in internal testing set and an AUC of 0.758 in external testing set. Further stratified analysis showed that the AUC for preoperatively predicting MVI by XGboost was 0.817 for HCC patients with positive hepatitis B surface antigen, 0.779 for male patients and 0.790 for elder patients. The calibration curves showed good agreement between observed and predicted values and the decision curve analysis curve showed relatively higher net benefits.Conclusions:We successfully established and verified a novel XGboost model based on eight routine laboratory parameters with relatively high and reliable predictive accuracy to preoperatively predict MVI in HCC.

Objective:To analyze the serum and urinary amino acid (AA) profiles of urolithiasis patients to explore the potential biomarkers for clinical screening and early diagnosis.Methods:Case-control study. Serum and urine samples were collected from 74 urolithiasis patients (aged 20-82 years, 41 men, 33 female) in the department of urology of the First Affiliated Hospital of Fujian Medical University and 35 healthy controls (HC, aged 22-80 years old, 20 men, 15 female) from the health examination center from February 2015 to October 2017. Serum and urinary AA levels of patients and HC were analyzed using Gas Chromatography-Mass Spectrometry (GC-MS) based metabolomic strategy. The multivariate statistical analysis methods of principal component analysis (PCA) and orthogonal partial least squares discrimination analysis (OPLS-DA) were employed for modeling. The variable importance projection (VIP) value of OPLS-DA model>1 and P<0.05 of t test were selected to screen the differential amino acid metabolites. The diagnostic capabilities of potential markers were evaluated by receiver operating characteristic (ROC) curve and binary logistic regression analysis. Results:Five AA metabolites including serine, glutamate, aspartic acid, isoleucine and glycine were found, which had statistically significant differences between the patient group and the control group ( P<0.05) and were associated with seven metabolic pathways. Serum serine, glutamate, aspartic acid, isoleucine and urine glycine and aspartic acid were combined into an integrated marker panel whose AUC value was 0.890, the sensitivity was 78.0%, and the specificity was 96.4%. Conclusion:Five amino acids in serum and urine could be used as an integrated biomarker panel for the clinical screening and early diagnosis of urolithiasis, which could provide some experimental basis for molecular urolithiasis research.

Objective The study aims to investigate the associationbetweencholesterol 7α-hydroxylase (CYP7A1) gene polymorphism and different clinical outcomes after Hepatitis B virus (HBV)infection in Fujian Han population and lay a foundation for understanding the mechanisms of genesis anddevelopment of HBV-related diseases.Methods Case-control study was conducted.586 patients of HBVpersistent infection without antiviral therapy and 225 HBV rehabilitation patients (35-55 years old) werecollected from May 2015 to June 2016 in the Liverish Center of First Clinical College of Fujian MedicalUniversity.The group of HBV persistent infection without antiviral therapy included 246 patients with chronichepatitis B, 177 patients with hepatitis B-related cirrhosis, and 163 patients with hepatitis B-related liver cancer.The rs3824260, rs4738687and rs8192871 loci of CYP7A1 gene were detected by improved multipleligase detection reaction (iMLDR).Logistic regression analysis and chi-square test were used to analyze thegenotyping results.Results Three SNPs ( single nucleotide polymorphisms ) of CYP7A1 gene wereselected and compared between HBV persistent infection group and HBV rehabilitation group and betweenchronic hepatitis B subgroup, liver cirrhosis subgroup and liver cancer subgroup.After adjustment for factorsincluding age andgender, there was no significant difference in the distribution of rs3824260 genotype amongthe groups(χ2 =1.565,P =0.459), however,the frequency of allele C in HBV rehabilitation group wassignificantly higher than in HBV persistent in fectiongroup for men (χ2 =4.365,P =0.037), whereas thefrequency of rs3824260 CC and CT was more likely to be observed in liver cancer group than in non -livercancer group (chronic hepatitis B subgroup and liver cirrhosis subgroup ) for women (χ2 =5.768,P =0.012;χ2 =10.130,P =0.001).The frequency of rs4738687 GG genotype was more likely to be observed innon-liver cancer group than in liver cancer group (χ2 =4.403,P =0.041;χ2 =6.940,P =0.009).Theresults of gender stratification showed that there were significant differences in the distribution of rs 4738687among the HBV persistent infection groups for men (χ2 =10.697,P =0.030), however, there was nosignificant difference in the distribution of rs4738687 among the HBV persistent infection groups for women(χ2 =4.627,P =0.329), and there was no significant difference in the distribution of genotype frequencyand allele frequency among all groups(χ2 =0.489,P =0.792).There was no significant difference after sexstratification either (χ2 =1.282, P =0.526;χ2 =1.565,P =0.465) .Conclusions These findingssuggested that CYP7A1 gene polymorphism was related todifferent clinical outcomes in Fujian Hanpopulation.The rs3824260 mutation had a certain gender preference and the mutation allele was detected ina higher proportion in male patients.Male HBV patients with rs3824260 C allele had more chance ofswitching to rehabilitation.The rs4738687 was likely to be related to the occurrence of liver cancer in FujianHan population, and GG genotype may delay the occurrence and development of liver cancer especially in themale group.The rs8192871 was not found to be related to the different clinical outcomes of HBV infection.

This study was conducted to explore a proper training model of interns' clinical thinking ability under the construction of a new four-year system of medical laboratory technology courses, combined with the establishment of innovative standard whole process practice mode. Multi-teaching methods of clinical thinking, such as explanation of laboratory sheet, interactive teaching based on micro digital system, interdisciplinary multiple information system, combined PBL teaching and intern report, were applied and evaluated in the laboratory. Integrated application of these methods remarkably improved the intern's com-prehensive professional quality and their practice performance. All methods received high evaluation from both the interns and teachers.

Objective To investigate the epidemiological and molecular virological characteristics in HBV-infected patients with copositive HBsAg and anti-HBs.Methods HBV serological markers were analyzed in 52 070 specimens.The epidemiological characteristics of HBsAg and anti-HBs simultaneously positive patients (the experimental group) and HBsAg positive and auti-HBs negative patients (the control group) were compared.The S protein of HBV coding region was amplified by semi-nested PCR and sequenced.The statistical differences between the two groups were compared in different gene regions,genotypes and different clinical diagnosis.Results HBsAg was positive in 20.40％ (10 621/52 070) of all specimens.In the patients with positive HBsAg,2.48％ (263/10 621) was positive anti-HBs.The prevalence of co-positive HBsAg and auti-HBs was higher in aged 0 to 9 years and greater than or equal to 80 years than that in other age,and the prevalence of positive HBsAg and negative anti-HBs was completely opposite.The mutation rate of S protein in the experimental group was significantly higher than that in the control group (1.52％ vs 0.81％,P ＜0.01) with the mutation in the major hydrophilic region (MHR) (1.68％ vs 0.57％,P ＜0.01).The mutation rates of S protein of HBV carriers,chronic hepatitis B (CHB) patients and patients with liver cirrhosis (LC) in the experimental group were significantly higher than those in the control group (1.47％ vs 0.65％,1.28％ vs 0.84％,2.21％ vs 0.44％,P ＜0.05,respectively),except for the patients with hepatocellular carcinoma (HCC) (1.97％ vs 2.21％,P ＞ 0.05).Conclusion Co-positive HBsAg and anti-HBs in HBV-infected patients was more common in HBsAg positive patients aged 0 to 9 years and greater than or equal to 80 years than the others.Coexistence of HBsAg and anti-HBs in HBV-infected patients may relate to immune escape caused by mutation of S protein (mainly MHR).The mutation rates of S protein in the two groups of patients,co-positive HBsAg and anti-HBs and the positive HBsAg combined with negative anti-HBs,were associated with the stage of liver disease.

Objective To investigate the correlation of hepatitis B patients in Fuzhou between resistance patterns in HBV P region and genotype,HBeAg,the hepatitis B process.Methods This was a retrospective study.The serum and clinic data of 1 115 hepatitis B patients were collected from the inpatient and outpatient Center for Liver Diseases in First Affiliated Hospital of Fujian Medical University between October 2011 and January 2015.HBV DNA was extracted and sequenced using the Sanger method to detect HBV genotype and resistance mutations in P region,HBeAg and HBeAb concentration were detected by chemiluminescent assay.The relationship between P region resistance mutations pattern,HBV genotype,serum HBeAg and the hepatitis B process was analyzed.The x2-test was used to compare the resistance rate and positive rate.Results There were significant differences between 14 kinds of resistance loci and the genotype distribution(x2 =30.788,P =0.004),the C/B genotype ratio of three common resistance loci (rtM204V/I,rtL180M,rtA181T/V) were 85/82,49/25 and 27/9,respectively,which in genotype C was higher than genotype B.The resistance ratio of hepatocellular carcinoma,liver cirrhosis,chronic hepatitis B,hepatitis B carriers was 31.4％ (11/35),37.6％ (65/173),27.3％ (146/535) and 21.8％ (43/197),respectively,which showed significant difference between the four clinical diagnosis (x2 =11.858,P =0.008).The highest percentage of resistance was liver cirrhosis,followed by hepatocellular carcinoma and chronic hepatitis B.There was significant difference in the distribution of HBV genotype between HBeAg (+) group and HBeAg (-) group (x2 =11.093,P =0.001),the HBeAg positive rate in genotype C [37.53％ (295/786)] was higher than in genotype B[35.62％ (280/786)].However,the total resistance rate between HBeAg (+) group and HBeAg(-) group was not significantly different[23.7％ (136/573) and 24.6％ (52/211),respectively,x2 =0.07,P =0.791].Conclusions HBV genotype was related to the resistance rates,HBeAg levels and the progress of hepatitis B.The resistance rate and HBeAg positive rate of genotype C were higher than those of genotype B,and clinical outcomes were worse in genotype C.HBV resistance rates and HBeAg levels were related to the progress of hepatitis B,the higher the resistance rates,the worse clinical outcomes.

Objective To investigated the clinical distributions and antimicrobial susceptibility of Streptococcus agalactia strains isolated from the patients .Methods The identification and susceptibility of the strains were mainly measured by automatic VITEK‐Ⅱ system ,the K‐B disc diffusion tests were used for the resistance test of erythromycin ,meropenem ,and D‐test .Results The iso‐lates were mainly from urine (63 .1% ) ,genital tract(7 .8% ) and wound secretion(6 .7% ) .They were obtained from patients in dif‐ferent situations ,including 110 patients who were older than 50 years old (61 .5% ) ,113 female patients (63 .1% ) ,12 gravidas (6 .7% ) ,3 vertical transmitted newborns(1 .7% ) ,and 82 patients with cancer ,undergoing chemo radiotherapy ,with diabetes ,tuber‐culosis or after operations(45 .8% ) .The resistant rates of the isolated Streptococcus agalactia to erythromycin and clindamycin were 42 .9% -93 .3% and 41 .9% -80 .0% respectively .The positive rate of D‐test was 4 .1% .The strains were highly resistant to tet‐racycline(>80% ) ,while the resistance to penicillin was below 10% except in 2008 .All isolates were susceptible to vancomycin and meropenem .Only one strain was resistant to Quinupristin‐dalfopristin .Conclusion Streptococcus agalactia infection in adults most‐ly cause genitourinary tract ,skin and soft tissue infections .There were more females than males with Streptococcus agalactia infec‐tion .Penicillin andβ‐lactams are still the first choice for the treatment .Erythromycin ,clindamycin and tetracycline should be used with caution under the guidance of laboratory susceptibility test results .