Objective:To evaluate the diagnostic performance of artificial intelligence-assisted diagnostic systems in cervical cytopathological examination.Methods:Cervical cytology slide data were retrospectively collected from four hospitals for the external validation of the developed artificial intelligence-assisted diagnostic system. Subsequently, prospective data collection was conducted for human-machine assisted studies.Results:In the retrospective study, a total of 3 162 valid samples were collected as external validation data. The system showed an area under the curve (AUC) of 0.890 (95% CI: 0.878-0.902), accuracy of 0.885 (95% CI: 0.873-0.896), sensitivity of 0.928 (95% CI: 0.914-0.941), and specificity of 0.852 (95% CI: 0.834-0.867). In the prospective study, 212 valid samples were collected, and five junior cytologists participated in the human-machine assisted study. Without artificial intelligence assistance, the average AUC for the five cytologists was 0.686 (95% CI: 0.650-0.722), the accuracy was 0.699 (95% CI: 0.671-0.727), the sensitivity was 0.653 (95% CI: 0.599-0.703), the specificity was 0.719 (95% CI: 0.685-0.750), the Fleiss κ value was 0.510, and the reading time was 223 seconds. With artificial intelligence assistance, the AUC, accuracy, sensitivity, and specificity increased by 0.166, 0.143, 0.225, and 0.107, respectively. Additionally, Fleiss κ was 0.730 and the reading time decreased by 188 seconds. All differences were statistically significant (all P<0.001). Conclusions:Artificial intelligence-assisted diagnosis system shows excellent performance and good generalizability, significantly improving the diagnostic accuracy, consistency, and efficiency of junior cytologists. It can be an effective auxiliary tool for junior cytologists in clinical practice.

Objective:To discuss the effect of KIAA1522 on the proliferation,migration,and invasion of lung cancer cells,and to clarify its signaling mechanism.Methods:Bioinformatics analysis was used to detect the expression levels of KIAA1522 mRNA and protein in 75 cases of human non-small cell lung cancer(NSCLC)tissues and adjacent normal lung tissues;immunohistochemical staining was used to detect the expression of KIAA1522 protein in NSCLC tissue and adjacent normal lung tissues;Western blotting method was used to detect the expression level of KIAA1522 protein in various lung cancer cell lines.KIAA1522-small interfering(siRNA)and over-expression plasmids were transfected into the lung cancer H1299 and A549 cells,respectively.The KIAA1522-siRNA experiment was divided into blank group,negative control group(si-NC group),KIAA1522-siRNA#1 group,and KIAA1522-siRNA#2 group.The KIAA1522 over-expression experiment was divided into control group,empty vector control group(OE-NC group,transfected with KIAA1522 over-expression empty vector plasmid),KIAA1522 overexpression group(OE-KIAA1522 group,transfected with KIAA1522 over-expression plasmid),KIAA1522 over-expression+MK2206 group[OE-KIAA1522+MK2206 group,co-transfected with KIAA1522 over-expression plasmid and protein kinase B(AKT)signaling pathway inhibitor MK2206],and MK2206 group(transfected with MK2206).Western blotting method was used to verify the transfection efficiencies of the cells in various groups;MTT assay was used to detect the proliferation activities of the lung cancer cells in various groups;cell scratch assay was used to detect the migration rates of lung cancer cells in various groups;Transwell chamber assay was used to detect the numbers of invasion lung cancer cells in various groups;Western blotting method was used to detect the expression levels of phosphorylated AKT(p-AKT),total AKT(t-AKT),cyclin D1(Cyclin D1),vascular endothelial growth factor(VEGF),and epithelial-mesenchymal transition(EMT)-related proteins[vimentin(Vimentin),N-cadherin(N-cadherin),and E-cadherin(E-cadherin)]proteins in the cells in various groups.Results:The bioinformatics analysis results showed that compared with adjacent normal lung tissue,the expression levels of KIAA1522 mRNA and protein in NSCLC tissue were significantly increased(P<0.05 or P<0.01).The immunohistochemistry staining results showed that compared with adjacent normal lung tissue,the positive expression rate of KIAA1522 protein in NSCLC tissue was significantly increased(P<0.05)and was associated with TNM stage(P<0.01).The Western blotting results showed that compared with normal lung epithelial cells BEAS-2B,the expression levels of KIAA1522 protein in lung cancer cell lines PC9,H1299,H460,A549,H1975,and H226 were significantly increased(P<0.05 or P<0.01).Compared with si-NC group,the expression levels of KIAA1522 protein in the H1299 cells in KIAA1522-siRNA#1 group and KIAA1522-siRNA#2 group were significantly decreased(P<0.01);compared with OE-NC group,the expression level of KIAA1522 protein in the A549 cells in OE-KIAA1522 group was significantly increased(P<0.01).The MTT results showed that at 24,48,and 72 h of cell culture,compared with si-NC group,the proliferation activities of the H1299 cells in KIAA1522-siRNA#1 group and KIAA1522-siRNA#2 group were significantly decreased(P<0.01);compared with OE-NC group,the proliferation activity of the A549 cells in OE-KIAA1522 group was significantly increased(P<0.05);compared with OE-KIAA1522 group,the proliferation activity of the A549 cells in OE-KIAA1522+MK2206 group was significantly decreased(P<0.01);compared with OE-KIAA1522+MK2206 group,the proliferation activity of the A549 cells in MK2206 group was significantly decreased(P<0.05).The cell scratch assay results showed that compared with si-NC group,the migration rates of the H1299 cells in KIAA1522-siRNA#1 group and KIAA1522-siRNA#2 group were significantly decreased(P<0.01);compared with OE-NC group,the migration rate of the A549 cells in OE-KIAA1522 group was significantly increased(P<0.01);compared with OE-KIAA1522 group,the migration rate of the A549 cells in OE-KIAA1522+MK2206 group was significantly decreased(P<0.05);compared with OE-KIAA1522+MK2206 group,the migration rate of the A549 cells in MK2206 group was significantly decreased(P<0.05).The Transwell chamber assay results showed that compared with si-NC group,the numbers of invasion H1299 cells in KIAA1522-siRNA#1 group and KIAA1522-siRNA#2 group were significantly decreased(P<0.01);compared with OE-NC group,the number of invasion A549 cells in OE-KIAA1522 group was significantly increased(P<0.01);compared with OE-KIAA1522 group,the number of invasion A549 cells in OE-KIAA1522+MK2206 group was significantly decreased(P<0.01);compared with OE-KIAA1522+MK2206 group,the number of invasion A549 cells in MK2206 group was significantly decreased(P<0.01).The Western blotting results showed that compared with si-NC group,the expression levels of p-AKT,Cyclin D1,Vimentin,N-cadherin,and VEGF proteins in the H1299 cells in KIAA1522-siRNA#1 group and KIAA1522-siRNA#2 group were significantly decreased(P<0.05 or P<0.01),while the expression level of E-cadherin protein was significantly increased(P<0.01);compared with OE-NC group,the expression levels of p-AKT,Cyclin D1,Vimentin,N-cadherin,and VEGF proteins in the A549 cells in OE-KIAA1522 group were significantly increased(P<0.05 or P<0.01),while the expression level of E-cadherin protein was significantly decreased(P<0.05);compared with OE-KIAA1522 group,the expression levels of p-AKT,Cyclin D1,Vimentin,N-cadherin,and VEGF proteins in the A549 cells in OE-KIAA1522+MK2206 group were significantly decreased(P<0.05 or P<0.01),while the expression level of E-cadherin protein was significantly increased(P<0.05);compared with OE-KIAA1522+MK2206 group,the expression levels of Cyclin D1,Vimentin,N-cadherin,and VEGF proteins in the A549 cells in MK2206 group were significantly decreased(P<0.05 or P<0.01),while the expression level of E-cadherin protein was significantly increased(P<0.05).Conclusion:The KIAA1522 protein upregulates the expression of Cyclin D1,EMT-related proteins,and VEGF protein in lung cancer cells,promoting the proliferation,migration,and invasion of lung cancer cells,and its mechanism is related to the activation of the AKT signaling pathway.

Objective:To evaluate the diagnostic performance of artificial intelligence-assisted diagnostic systems in cervical cytopathological examination.Methods:Cervical cytology slide data were retrospectively collected from four hospitals for the external validation of the developed artificial intelligence-assisted diagnostic system. Subsequently, prospective data collection was conducted for human-machine assisted studies.Results:In the retrospective study, a total of 3 162 valid samples were collected as external validation data. The system showed an area under the curve (AUC) of 0.890 (95% CI: 0.878-0.902), accuracy of 0.885 (95% CI: 0.873-0.896), sensitivity of 0.928 (95% CI: 0.914-0.941), and specificity of 0.852 (95% CI: 0.834-0.867). In the prospective study, 212 valid samples were collected, and five junior cytologists participated in the human-machine assisted study. Without artificial intelligence assistance, the average AUC for the five cytologists was 0.686 (95% CI: 0.650-0.722), the accuracy was 0.699 (95% CI: 0.671-0.727), the sensitivity was 0.653 (95% CI: 0.599-0.703), the specificity was 0.719 (95% CI: 0.685-0.750), the Fleiss κ value was 0.510, and the reading time was 223 seconds. With artificial intelligence assistance, the AUC, accuracy, sensitivity, and specificity increased by 0.166, 0.143, 0.225, and 0.107, respectively. Additionally, Fleiss κ was 0.730 and the reading time decreased by 188 seconds. All differences were statistically significant (all P<0.001). Conclusions:Artificial intelligence-assisted diagnosis system shows excellent performance and good generalizability, significantly improving the diagnostic accuracy, consistency, and efficiency of junior cytologists. It can be an effective auxiliary tool for junior cytologists in clinical practice.

Background::Thoracic aortic aneurysm (TAA) is a fatal cardiovascular disease, the pathogenesis of which has not yet been clarified. This study aimed to identify and validate the diagnostic markers of TAA to provide a strong theoretical basis for developing new methods to prevent and treat this disease.Methods::Gene expression profiles of the GSE9106, GSE26155, and GSE155468 datasets were acquired from the Gene Expression Omnibus (GEO) database. Differentially expressed genes (DEGs) were identified using the "limma" package in R. Least absolute shrinkage and selection operator (LASSO), support vector machine-recursive feature elimination (SVM-RFE), random forest, and binary logistic regression analyses were used to screen the diagnostic marker genes. Single-sample gene set enrichment analysis (ssGSEA) was used to estimate immune cell infiltration in TAA.Results::A total of 16 DEGs were identified. The enrichment and functional correlation analyses showed that DEGs were mainly associated with inflammatory response pathways and collagen-related diseases. Collagen type I alpha 1 chain ( COL1A1) and synaptotagmin like 2 ( SYTL2) were identified as diagnostic marker genes with a high diagnostic value for TAA. The expression of COL1A1 and SYTL2 was considerably higher in TAA vascular wall tissues than in the corresponding normal tissues, and there were significant differences in the infiltration of immune cells between TAA and normal vascular wall tissues. Additionally, COL1A1 and SYTL2 expression were associated with the infiltration of immune cells in the vascular wall tissue. Single-cell analysis showed that COL1A1 in TAA was mainly derived from fibroblasts and SYTL2 mainly from cluster of differentiation (CD)8 + T cells. In addition, single-cell analysis indicated that fibroblasts and CD8 + T cells in TAA were significantly higher than those in normal arterial wall tissue. Conclusions::COL1A1 and SYTL2 may serve as diagnostic marker genes for TAA. The upregulation of SYTL2 and COL1A1 may be involved in the inflammatory infiltration of the vessel wall and poor extracellular matrix remodeling, promoting the progression of TAA.

Objective:To discuss the effects of monocyte chemoattractant protein-1(MCP-1)on the migration and invasion of lung cancer A549 cells,and to clarify the mechanisms.Methods:Immunohistochemistry method was used to detect the expression of MCP-1 protein in 80 cases of non-small cell lung cancer(NSCLC)and adjacent normal lung tissues.The human lung cancer A549 cells were cultured in vitro.The MCP-1-small interfering RNA(siRNA)experiment was divided into blank group,negative control group(si-NC group),MCP-1-siRNA-1 group,and MCP-1-siRNA-2 group.The MCP-1 over-expression experiment was divided into control group,empty vector control group(OE-NC,transfected with MCP-1 over-expression empty vector),over-expression MCP-1 group(OE-MCP-1 group,transfected with MCP-1 over-expression plasmid),over-expression MCP-1+extracellular regulated protein kinase(ERK)pathway inhibitor PD98059 group(OE-MCP-1+PD98059 group,co-transfected with MCP-1 over-expression plasmid and PD98059),and PD98059 group(transfected with PD98059).The MCP-1 siRNA and plasmids were transfected into the lung cancer A549 cells;Western blotting method was used to verify the transfection efficiencies of the cells in various groups;the migration rate and the number of invasion cells in various groups were observed by wound healing assay and Transwell chamber assay,respectively;Western blotting method was also used to detect the expression levels of phosphorylated ERK(p-ERK),total ERK(t-ERK),and epithelial-mesenchymal transition(EMT)-related proteins in the A549 cells in various groups.Results:Compared with adjacent tissue,the positive expression rate of MCP-1 protein in NSCLC tissue was significantly increased(P<0.05),and the expression level of MCP-1 protein was related to TNM stage and lymph node metastasis(P<0.05).Compared with si-NC group,the expression level of MCP-1 protein in the cells in MCP-1-siRNA-1 and MCP-1-siRNA-2 groups was significantly decreased(P<0.01).Compared with control group and OE-NC group,the expression level of MCP-1 protein in the cells in OE-MCP-1 group was significantly increased(P<0.01).The wound healing assay results showed that compared with si-NC group,the migration rate of the cells in MCP-1-siRNA-1 and MCP-1-siRNA-2 groups were significantly decreased(P<0.01).Compared with OE-NC group,the migration rate of the cells in OE-MCP-1 group was significantly increased(P<0.01);compared with OE-MCP-1 group,the migration rate of the cells in OE-MCP-1+PD98059 group was significantly decreased(P<0.01).Compared with OE-MCP-1+PD98059 group,the migration rate of the cells in PD98059 group was significantly decreased(P<0.01).The Transwell chamber assay results showed that compared with si-NC group,the number of invasion cells in MCP-1-siRNA-1 and MCP-1-siRNA-2 groups was significantly decreased(P<0.01).Compared with OE-NC group,the number of invasion cells in OE-MCP-1 group was significantly increased(P<0.01);compared with OE-MCP-1 group,the number of invasion cells in OE-MCP-1+PD98059 group was significantly decreased(P<0.01);compared with OE-MCP-1+PD98059 group,the number of invasion cells in PD98059 group was significantly decreased(P<0.01).The Western blotting results showed that compared with si-NC group,the expression levels of p-ERK,Vimentin,and N-cadherin protein in the cells in MCP-1-siRNA-1 and MCP-1-siRNA-2 groups were significantly decreased(P<0.05 or P<0.01),and the expression level of E-cadherin proteins was significantly increased(P<0.01).Compared with OE-NC group,the expression levels of p-ERK,Vimentin,and N-cadherin proteins in the cells in OE-MCP-1 group were significantly increased(P<0.01),and the expression level of E-cadherin protein was significantly decreased(P<0.01).Compared with OE-MCP-1 group,the expression levels of p-ERK,Vimentin,and N-cadherins proteins in the OE-MCP-1+PD98059 group were significantly decreased(P<0.01),and the expression level of E-cadherin protein was significantly increased(P<0.05).Compared with OE-MCP-1+PD98059 group,the expression levels of p-ERK,Vimentin,and N-cadherin proteins in the cells in PD98059 group were significantly decreased(P<0.05 or P<0.01),and the expression level of E-cadherin protein was increased(P<0.01).Conclusion:MCP-1 protein can upregulate the expression of EMT-related proteins in the lung cancer A549 cells,and promote the migration and invasion of the lung cancer A549 cells;its mechanism may be related to the activation of the ERK signaling pathway.

Objective To investigate the clinical features and risk factors of primary biliary cholangitis(PBC)comorbid with metabolic associated fatty liver disease(MAFLD)and the interaction between the two diseases.Methods A total of 187 patients who were diagnosed with PBC,MAFLD,or PBC with MAFLD in The First Affiliated Hospital of Kunming Medical University from January 2019 to December 2022 were enrolled and divided into PBC group with 70 patients,PBC+MAFLD group with 38 patients,and MAFLD group with 79 patients.Related data were collected,including general information,clinical symptoms,serological parameters,transient elastography(FibroScan),and non-invasive fibrosis markers,which were compared between the three groups.A one-way analysis of variance or the Kruskal-Wallis H test was used for comparison of continuous data between groups,the chi-square test or the Fisher's exact test was used for comparison of categorical data between groups,and the binary Logistic regression analysis was used for multivariate analysis.Results There were significant differences between the three groups in sex,age,height,weight,body mass index(BMI),and history of autoimmune diseases(P<0.05).In the PBC+MAFLD group,female patients accounted for 89.5%,with a mean age of 57.26±12.72 years and a BMI of 23.35±3.70 kg/m2,and in the PBC group,the detection rate of autoimmune diseases was 25.7%(18 patients).There were significant differences between the three groups in the incidence rates of weakness,poor appetite,pruritus,jaundice,varices,ascites,and splenomegaly(all P<0.05).The PBC+MAFLD group had the common symptoms of weakness in 18 patients(47.4%),poor appetite in 15 patients(39.5%),abdominal pain in 14 patients(36.8%),and abdominal distension in 16 patients(42.1%);the MAFLD group had the common symptoms of abdominal pain in 34 patients(43%)and abdominal distension in 32 patients(40.5%);the PBC group had the common symptoms of weakness in 37 patients(52.9%),poor appetite in 25 patients(35.7%),jaundice in 25 patients(35.7%),abdominal pain in 18 patients(25.7%),abdominal distension in 25 patients(35.7%),varices in 19 patients(27.9%),ascites in 23 patients(32.9%),and splenomegaly in 44 patients(62.9%).The PBC+MAFLD group had a controlled attenuation parameter(CAP),which was higher than that of the PBC group,and the PBC group had significantly higher levels of liver stiffness measurement,aspartate aminotransferase-to-platelet ratio index(APRI),and fibrosis-4(FIB-4)than the MAFLD group(all P<0.05).The factors without multicollinearity were included in the regression analysis,and with the PBC group as the reference group,FIB-4(odds ratio[OR]=0.218,95%confidence interval[CI]:0.069-0.633,P<0.05)and history of autoimmune diseases(OR=0.229,95%CI:0.067-0.810,P<0.05)were influencing factors for the onset of PBC with MAFLD;with the MAFLD group as the reference group,ALT(OR=0.157,95%CI:0.025-1.000,P<0.05)and TBil(OR=0.995,95%CI:0.990-0.999,P<0.05)were influencing factors for the onset of PBC with MAFLD.Conclusion PBC with MAFLD lacks specific clinical manifestations,and PBC patients tend to have more severe clinical manifestations and a higher incidence rate of liver function decompensation.PBC comorbid with MAFLD may not aggravate the disease progression of PBC.

Objective Each item of the Chronic Hepatitis Quality of Life Scale (V2.0) was thoroughly analyzed and evaluated according to classical test theory (CTT) and item response theory (IRT). Methods The QLICD-CH (V2.0) scale was used to assess the life quality of 226 patients with chronic hepatitis who were hospitalized in the Department of Infectious Diseases from March 2019 to January 2020. By using CTT's coefficient of variation method, factor analysis method, Cronbach's alpha coefficient method and correlation coefficient method, macroscopic statistical analysis on the scale results was performed. IRT was used to microscopically analyze the information content, discrimination coefficient and difficulty parameters of each item. Results CTT analysis revealed that 36 items of the QLICD-CH (V2.0) scale satisfied three of the four statistical methods. IRT analysis showed that the average information content of 36 items was greater than 0.108, with the degree of discrimination ranging from -0.387 to 3.574; the degree of discrimination of 40 items was greater than 0.30; the difficulty coefficient of each item ranged from -14.805 to 9.057, with 31 items conformed to [-4, 4] and a monotonically increasing trend from b1 to b4. Conclusion The majority of items on the QLICD-CH (V2.0) scale perform well, but there are a few items that could be improved, optimized, or removed.

Objective To characterize the distribution and assess the exposure to phthalic acid esters （PAEs） in the indoor dust of Shanghai City. Methods Samples were collected from 33 sampling sites， including homes， hotels， offices and public places， in Shanghai in 2018， 2019， and 2020. The samples were pretreated by 100 sieves， extracted and concentrated， and then analyzed by gas chromatography-mass spectrometry in selected ion mode （SIM）. Results Results on the characteristics of PAEs in indoor dust in different places showed that concentrations of PAEs were in a range of <0.01-2 464 mg·kg^-1.The average concentration of 16 PAEs was 613 mg·kg^-1. Bis（2-ethylhexyl） phthalate （DEHP）， di-iso-butyl phthalate （DiBP）， di-n-butyl phthalate （DBP） and di-n-octyl phthalate （DnOP） were the main components of PAEs in indoor dust， accounting for approximately 99.5% of 16 PAEs. The intake of DEHP， DBP， DEP and BBP was lower than the tolerable daily intake （TDI） and reference doses （RfD） set by EU CSTEE and U.S. EPA. Conclusion Average daily dose （ADD） via indoor dust is estimated， and the order of intake through different pathways is hand-oral intake>skin contact>respiratory inhalation. Exposure risk of PAEs in children is greater than that in adults.

Objective To investigate the mechanism of action of integrin α4 (ITGA4) in liver fibrosis based on the anti-liver fibrosis effect of sticky sugar amino acid (SSAA) in rats. Methods A rat model of liver fibrosis was induced by intraperitoneal injection of CCl 4 , and then colchicine and low-, middle-, and high-dose SSAA were used for intervention, with blank control group and SSAA group as control. After 12 weeks of experimental intervention, serum and liver samples were collected to measure the serum levels of alanine aminotransferase (ALT) and aspartate aminotransferase (AST), and HE staining and Sirius Red staining were used to observe the pathological conditions of liver tissue; quantitative real-time PCR was used to measure the transcriptional level of ITGA4, integrin β1 (ITGB1), transforming growth factor-β1 (TGFβ1), alpha-smooth muscle actin (α-SMA), and TIMP2 in liver tissue; Western blot was used to measure the relative protein expression levels of ITGA4, ITGB1, TGFβ1, α-SMA, MMP2, TIMP1, and TIMP2; immunohistochemistry was used to observe the protein expression of TGFβ1 and α-SMA. A one-way analysis of variance was used for comparison of continuous data between multiple groups, and the least significant difference t -test was used for comparison between two groups. Results There were significant increases in AST and ALT in the CCl 4 model group, and intervention with colchicine or low-, middle-, and high-dose SSAA reduced the levels of AST and ALT, with a significant difference between the CCl 4 model group and the other groups (all P < 0.05). HE staining and Sirius Red staining showed disordered structure of hepatic lobules and an increase in collagen fibers in the CCl 4 model group, and the structure of hepatic lobules was improved after intervention with colchicine or low-, middle-, and high-dose SSAA. The CCl 4 model group had significantly higher transcriptional levels of ITGA4, TGFβ1, α-SMA, and TIMP2 than the other groups, and there were significant reductions in the transcriptional levels of each factor after intervention with colchicine or SSAA, with a significant difference between the CCl 4 model group and the other groups (all P < 0.05). The CCl 4 model group had significantly higher protein expression levels of ITGA4, TGFβ1, α-SMA, TIMP2, and TIMP1 and a significantly lower protein expression level of MMP2 than the other groups, and intervention with colchicine or SSAA inhibited the expression of ITGA4, TGFβ1, α-SMA, TIMP2, and TIMP1 and promoted the expression of MMP2. Immunohistochemistry showed that the CCl 4 model group had significantly higher expression levels of TGFβ1 and α-SMA than the other groups, which was inhibited by intervention with colchicine or SSAA. The high-dose SSAA group had the most significant effect in reducing aminotransferases, improving lobular structure, and inhibiting the protein expression of liver fibrosis factors. Conclusion The high expression of ITGA4 in the liver is associated with the development of liver fibrosis, which is consistent with the increases in the expression of TGFβ1 and α-SMA. Inhibiting the expression of ITGA4 can provide more therapeutic targets for liver fibrosis and expand the anti-liver fibrosis mechanism of SSAA.

Objective    To explore the efficiency of Ki-67 expression and CT imaging features in predicting the degree of invasion of lung adenocarcinoma. Methods    The clinical data of 217 patients with pulmonary nodules, who were diagnosed as suspicious lung cancer by multi-disciplinary treatment clinic of pulmonary nodules in our hospital from September 2017 to August 2021, were retrospectively analyzed. There were 84 males and 133 females, aged 52 (25-84) years. The patients were divided into two groups according to the infiltration degree, including an adenocarcinoma in situ and microinvasive adenocarcinoma group (n=145) and an invasive adenocarcinoma group (n=72). Results    There was no statistical difference in the age and gender between the two groups (P>0.05). The univariate analysis showed that CK-7, P63, P40 and CK56 expressions were not different between the two groups (P=0.172, 0.468, 0.827, 0.313), while Napsin A, TTF-1 and Ki-67 expressions were statistically different (P=0.002, 0.020, <0.001). The multivariate analysis showed that Ki-67 expression was statistically different between the two groups (P<0.001). Ki-67 was positively correlated with malignant features of CT images and the degree of lung adenocarcinoma invasion (P<0.05). Ki-67 and CT imaging features alone could predict the degree of lung adenocarcinoma invasion, but their sensitivity and specificity were not high. Ki-67 combined with CT imaging features could achieve a higher prediction efficiency. Conclusion    Compared with Ki-67 or CT imaging features alone, the combined prediction of Ki-67 and imaging features is more effective, which is of great significance for clinicians to select the appropriate operation occasion.