ObjectiveTo evaluate the clinical efficacy and economic value of the Jiangsu Traditional Chinese Medicine （TCM） Diagnosis and Treatment Protocol for Dominant Diseases （abbreviated as the Diagnosis and Treatment Protocol） in adult patients with non-severe community-acquired pneumonia （CAP） based on real-world clinical data. MethodsA retrospective real-world cohort study was conducted using electronic medical records of adult patients hospitalized for non-severe CAP from September 1st， 2023 to December 31st， 2024 across 10 TCM hospitals in Jiangsu province. Patients were classified into an exposure group and a non-exposure group based on whether they received Chinese herbal medicine （CHM） according to the Diagnosis and Treatment Protocol. The non-exposure group received only conventional western medicine， while the exposure group additionally received differentiated CHM for at least five consecutive days. Outcomes were compared between two patient groups， including cough resolution rate， sputum resolution rate （assessed by volume， color， and consistency）， incidence of abnormal C-reactive protein （CRP）， incidence of abnormal white blood cell （WBC） count， and radiographic resolution rate of pulmonary infiltrates on chest imaging. Multivariable logistic regression was performed to identify factors influencing clinical efficacy. Subgroup analyses were conducted according to age， gender， smoking status， history of hypertension， and pneumonia severity score （CURB-65）， and the efficacy of treatment for cough and sputum was analyzed within each subgroup. Cost-effectiveness analysis was conducted using cough resolution rate as the outcome measure， evaluating the pharmacoeconomics of the two groups. ResultsA total of 1688 patients were included with 1293 in the exposure group and 395 in the non-exposure group. Compared to the non-exposure group， the exposure group demonstrated significantly higher resolution rates of cough， sputum volume， color， and consistency， as well as a significantly lower incidence of abnormal CRP （P＜0.05）. No statistically significant difference was observed between the groups in terms of abnormal WBC count and radiographic resolution rate of pulmonary infiltrates （P＞0.05）. Logistic regression analysis showed that the cough resolution rate in the exposure group was 1.83 times that of the non-exposure group， while the probabilities of resolution in sputum volume， color， and consistency were 1.37， 2.09， and 1.56 times those of the non-exposure group， respectively （P＜0.05）. Subgroup analyses showed that the exposure group achieved significantly higher cough resolution rates across most subgroups except for populations with a CURB-65 score ≥2 or those with a history of hypertension （P＜0.05）. Specifically， among females， patients aged ≥18 and ＜65 years， non-smokers， those without hypertension， and those with a CURB-65 score of 0， the exposure group showed a higher cough resolution rate than the non-exposure group （P＜0.05）. From an economic perspective， total hospitalization cost， length of stay， antibiotic cost， and CHM cost all differed significantly between groups （P＜0.05）. The cost-effectiveness ratio （CER） was 10,788.80 CNY/case in the exposure group， while 22,513.80 CNY/case in the non-exposure group. This implies that， compared with the exposure group， the non-exposure group incurred an additional 17,302.27 CNY to achieve one case of cough resolution. When the willingness-to-pay threshold ranged from 0 to 50,000 CNY， the probability of economic advantage was consistently higher in the exposure group than in the non-exposure group. ConclusionOn the basis of conventional western medicine， the addition of CHM in accordance with the Diagnosis and Treatment Protocol can effectively improve clinical symptoms， reduce inflammatory markers， promote clinical recovery， and is more cost-effective in treating adults with non-severe CAP.

ObjectiveTo observe the real‑world effectiveness and economic outcomes of Weishi Qingjin Formula （苇石清金方， WQF） in the treatment of adult community‑acquired pneumonia （CAP） with phlegm‑heat obstructing the lung syndrome. MethodsBased on a multicenter， real-world retrospective cohort study， clinical data were collected from hospitalized adult patients diagnosed with non‑severe CAP and phlegm‑heat obstructing the lung syndrome in 10 traditional Chinese medicine （TCM） hospitals in Jiangsu province. Patients were divided into an exposure group （those who received oral WQF） and a non‑exposure group （those who did not）. The following outcomes were compared between the two groups before and after treatment， which were remission rates of clinical symptoms including cough， expectoration （sputum volume， color， consistency）， and chest pain， levels of inflammatory markers including C‑reactive protein （CRP） and white blood cell count （WBC）， and the rate of pulmonary inflammatory absorption on chest CT. Subgroup analyses were performed based on age， gender， smoking status， presence of hypertension， and the severity of community-acquired pneumonia （CURB‑65） score， comparing the two groups in terms of cough remission rate， chest pain remission rate， and chest CT absorption rate. For health economic evaluation， cost‑effectiveness analysis was used to calculate the cost‑effectiveness ratio （CER） and incremental cost‑effectiveness ratio （ICER）. Univariate sensitivity analysis and probabilistic sensitivity analysis were performed to test the robustness of the results. ResultsA total of 647 patients in the exposure group and 1491 patients in the non-exposure group were included in the final statistical analysis. There was no statistically significant difference in length of hospital stay， gender， marital status， smoking history， bronchoscopy history， and comorbidities between the groups （P＞0.05）， but age， CURB-65 score， and antibiotic use. The exposure group had significantly higher remission rates of cough and sputum consistency than the non-exposure group （P＜0.05）. After adjusting for confounders using propensity score matching and logistic regression， the cough remission rate in the exposure group was 1.49 times that of the non-exposure group （P＜0.01）. No significant difference was observed between groups in the reduction rates of CRP and WBC， and in the rate of pulmonary inflammatory absorption on chest CT （P＞0.05）. Subgroup analyses revealed that the cough remission rate in the exposure group was significantly better than that in the non-exposure group except for patients aged ≥65 years， smokers， hypertensive patients， those using other type antibiotics or not using antibiotics， and those with a CURB-65 score ≥1 （P＜0.05）. Among smokers， the chest pain remission rate in the exposure group was 4.38 times that of the non-exposure group （P＜0.01）. No significant difference in chest CT absorption rate was found between groups across subgroups of gender， age， hypertension status， or antibiotic type （P＞0.05）. In terms of economic evaluation， CER was 10,877.60 CNY/case in the exposure group and 16,773.10 CNY/case in the non-exposure group. Compared to the exposure group， the non-exposure group incurred an additional 15,034.26 CNY to achieve one case of cough resolution， indicating a more favorable cost-effectiveness profile. Probabilistic sensitivity analysis yielded results consistent with the cost-effectiveness analysis， confirming the robustness of the findings. ConclusionWQF demonstrates significant efficacy in improving cough symptoms in the treatment of adult CAP with phlegm-heat obstructing the lung syndrome， and also exhibits favorable economic benefits.

ObjectiveTo evaluate the clinical efficacy and economic value of the Jiangsu Traditional Chinese Medicine （TCM） Diagnosis and Treatment Protocol for Dominant Diseases （abbreviated as the Diagnosis and Treatment Protocol） in adult patients with non-severe community-acquired pneumonia （CAP） based on real-world clinical data. MethodsA retrospective real-world cohort study was conducted using electronic medical records of adult patients hospitalized for non-severe CAP from September 1st， 2023 to December 31st， 2024 across 10 TCM hospitals in Jiangsu province. Patients were classified into an exposure group and a non-exposure group based on whether they received Chinese herbal medicine （CHM） according to the Diagnosis and Treatment Protocol. The non-exposure group received only conventional western medicine， while the exposure group additionally received differentiated CHM for at least five consecutive days. Outcomes were compared between two patient groups， including cough resolution rate， sputum resolution rate （assessed by volume， color， and consistency）， incidence of abnormal C-reactive protein （CRP）， incidence of abnormal white blood cell （WBC） count， and radiographic resolution rate of pulmonary infiltrates on chest imaging. Multivariable logistic regression was performed to identify factors influencing clinical efficacy. Subgroup analyses were conducted according to age， gender， smoking status， history of hypertension， and pneumonia severity score （CURB-65）， and the efficacy of treatment for cough and sputum was analyzed within each subgroup. Cost-effectiveness analysis was conducted using cough resolution rate as the outcome measure， evaluating the pharmacoeconomics of the two groups. ResultsA total of 1688 patients were included with 1293 in the exposure group and 395 in the non-exposure group. Compared to the non-exposure group， the exposure group demonstrated significantly higher resolution rates of cough， sputum volume， color， and consistency， as well as a significantly lower incidence of abnormal CRP （P＜0.05）. No statistically significant difference was observed between the groups in terms of abnormal WBC count and radiographic resolution rate of pulmonary infiltrates （P＞0.05）. Logistic regression analysis showed that the cough resolution rate in the exposure group was 1.83 times that of the non-exposure group， while the probabilities of resolution in sputum volume， color， and consistency were 1.37， 2.09， and 1.56 times those of the non-exposure group， respectively （P＜0.05）. Subgroup analyses showed that the exposure group achieved significantly higher cough resolution rates across most subgroups except for populations with a CURB-65 score ≥2 or those with a history of hypertension （P＜0.05）. Specifically， among females， patients aged ≥18 and ＜65 years， non-smokers， those without hypertension， and those with a CURB-65 score of 0， the exposure group showed a higher cough resolution rate than the non-exposure group （P＜0.05）. From an economic perspective， total hospitalization cost， length of stay， antibiotic cost， and CHM cost all differed significantly between groups （P＜0.05）. The cost-effectiveness ratio （CER） was 10,788.80 CNY/case in the exposure group， while 22,513.80 CNY/case in the non-exposure group. This implies that， compared with the exposure group， the non-exposure group incurred an additional 17,302.27 CNY to achieve one case of cough resolution. When the willingness-to-pay threshold ranged from 0 to 50,000 CNY， the probability of economic advantage was consistently higher in the exposure group than in the non-exposure group. ConclusionOn the basis of conventional western medicine， the addition of CHM in accordance with the Diagnosis and Treatment Protocol can effectively improve clinical symptoms， reduce inflammatory markers， promote clinical recovery， and is more cost-effective in treating adults with non-severe CAP.

ObjectiveTo observe the real‑world effectiveness and economic outcomes of Weishi Qingjin Formula （苇石清金方， WQF） in the treatment of adult community‑acquired pneumonia （CAP） with phlegm‑heat obstructing the lung syndrome. MethodsBased on a multicenter， real-world retrospective cohort study， clinical data were collected from hospitalized adult patients diagnosed with non‑severe CAP and phlegm‑heat obstructing the lung syndrome in 10 traditional Chinese medicine （TCM） hospitals in Jiangsu province. Patients were divided into an exposure group （those who received oral WQF） and a non‑exposure group （those who did not）. The following outcomes were compared between the two groups before and after treatment， which were remission rates of clinical symptoms including cough， expectoration （sputum volume， color， consistency）， and chest pain， levels of inflammatory markers including C‑reactive protein （CRP） and white blood cell count （WBC）， and the rate of pulmonary inflammatory absorption on chest CT. Subgroup analyses were performed based on age， gender， smoking status， presence of hypertension， and the severity of community-acquired pneumonia （CURB‑65） score， comparing the two groups in terms of cough remission rate， chest pain remission rate， and chest CT absorption rate. For health economic evaluation， cost‑effectiveness analysis was used to calculate the cost‑effectiveness ratio （CER） and incremental cost‑effectiveness ratio （ICER）. Univariate sensitivity analysis and probabilistic sensitivity analysis were performed to test the robustness of the results. ResultsA total of 647 patients in the exposure group and 1491 patients in the non-exposure group were included in the final statistical analysis. There was no statistically significant difference in length of hospital stay， gender， marital status， smoking history， bronchoscopy history， and comorbidities between the groups （P＞0.05）， but age， CURB-65 score， and antibiotic use. The exposure group had significantly higher remission rates of cough and sputum consistency than the non-exposure group （P＜0.05）. After adjusting for confounders using propensity score matching and logistic regression， the cough remission rate in the exposure group was 1.49 times that of the non-exposure group （P＜0.01）. No significant difference was observed between groups in the reduction rates of CRP and WBC， and in the rate of pulmonary inflammatory absorption on chest CT （P＞0.05）. Subgroup analyses revealed that the cough remission rate in the exposure group was significantly better than that in the non-exposure group except for patients aged ≥65 years， smokers， hypertensive patients， those using other type antibiotics or not using antibiotics， and those with a CURB-65 score ≥1 （P＜0.05）. Among smokers， the chest pain remission rate in the exposure group was 4.38 times that of the non-exposure group （P＜0.01）. No significant difference in chest CT absorption rate was found between groups across subgroups of gender， age， hypertension status， or antibiotic type （P＞0.05）. In terms of economic evaluation， CER was 10,877.60 CNY/case in the exposure group and 16,773.10 CNY/case in the non-exposure group. Compared to the exposure group， the non-exposure group incurred an additional 15,034.26 CNY to achieve one case of cough resolution， indicating a more favorable cost-effectiveness profile. Probabilistic sensitivity analysis yielded results consistent with the cost-effectiveness analysis， confirming the robustness of the findings. ConclusionWQF demonstrates significant efficacy in improving cough symptoms in the treatment of adult CAP with phlegm-heat obstructing the lung syndrome， and also exhibits favorable economic benefits.

Objective:To construct a model of mitochondria-related genes (Mito-RGs) in hepatocellular carcinoma (HCC), and predict the immune, prognostic and therapeutic characteristics of HCC based on the model, so as to provide a new idea for the diagnosis and treatment of HCC.Methods:The expression profiles of HCC and corresponding clinical information were obtained from the Cancer Genome Atlas (TCGA) database. Univariate Cox regression, Least Absolute Shrinkage and Selection Operator (LASSO) regression and multivariate Cox regression were used to construct a prognostic model of HCC based on Mito-RGs, and the International Cancer Genome Consortium-Liver Cancer-RIKEN-Japan ICGC-LIRI-JP dataset were used for validation. GO and KEGG analyses were performed to investigate the signaling pathways enriched for differentially expressed genes in the high- and low-risk groups. Immune infiltration was assessed using CIBERSORT. Single-cell data were used to study the proportion of immune cells in high- and low-risk groups of HCC samples and the relationship with cell proliferation. Cellminer was used to assess the relationship between risk score models and drug sensitivity.Results:A risk-prognostic model of HCC containing seven Mito-RGs ( DTYMK, ACADS, HMGCS2, CYP27A1, TOMM40L, STOM, and AKR1B10) was constructed. High-risk HCC patients had a worse prognosis. Genes upregulated in the high- and low-risk groups of differentially expressed genes were enriched in cell cycle and metabolism-related pathways. Single-cell data showed higher proportions of CD8 + T cells, macrophages and monocytes, and proliferating cells in the high-risk group. CIBERSORT analysis suggested that Treg cells and M0 macrophages were more abundant in the high-risk group, whereas CD8 + T cells and CD4 + memory T cells were less abundant. Patients in the high-risk group were more sensitive to myeloid cell leukemia sequence 1 inhibitor, vincristine, phosphatidylinositol kinase beta subunit inhibitor, and aurora kinase A, while trametinib, selumetinib, extracellular regulated protein kinase, and mitogen-activated extracellular signal-regulated kinase were more effective in the low-risk group. Conclusion:The constructed Mito-RGs model is capable of providing a more accurate assessment of the prognosis and the degree of immune cell infiltration in HCC patients.

Objective:To investigate the value of habitat radiomic features based on gadolinium ethoxybenzyl diethylenetriamine pentaacetic acid (Gd-EOB-DTPA)-enhanced MRI in establishing a predictive model for cytokeratin 19 (CK19) expression in hepatocellular carcinoma (HCC) and to evaluate its role in prognostic risk stratification.Methods:This multicenter case-control study retrospectively enrolled 489 patients with pathologically confirmed HCC who underwent Gd-EOB-DTPA-enhanced MRI between June 2016 and June 2024. Among them, 346 patients from the First Affiliated Hospital of Soochow University were divided into a training cohort ( n=245) and an internal test cohort ( n=101) via stratified sampling at a 7∶3 ratio. And 143 patients from Nantong Third Hospital Affiliated to Nantong University served as an external validation cohort. The training cohort included 53 CK19-positive and 192 CK19-negative patients. The internal test cohort included 21 CK19-positive and 80 CK19-negative patients. The external validation cohort included 30 CK19-positive and 113 CK19-negative patients. Univariate logistic regression analysis was performed to identify potential factors associated with CK19 expression, and a clinical-radiologic model was constructed. The k-means clustering algorithm was applied to segment target HCC lesions into 3 subregions. Radiomic features were extracted and selected from these habitat subregions. Habitat radiomics models were constructed for the arterial phase (AP), portal venous phase, hepatobiliary phase (HBP), and combined phases (CP). Multivariate logistic regression analysis identified independent clinical and radiologic predictors of CK19 expression, and the optimal habitat model score was integrated to build a clinical-radiologic-habitat combined model. The area under the receiver operating characteristic curve (AUC) was used to evaluate model predictive performance. Recurrence-free survival (RFS) was analyzed using the Kaplan-Meier method and the differences in survival curves were compared with the log-rank test. Results:Univariate logistic regression analysis revealed that alpha-fetoprotein (AFP) ( OR=2.629, 95% CI 1.412-4.896, P=0.002), AP enhancement ( OR=3.636, 95% CI 1.642-8.052, P=0.001), AP peritumoral enhancement ( OR=2.219, 95% CI 1.084-4.542, P=0.029), and HBP peritumoral hypointensity ( OR=2.010, 95% CI 1.004-4.021, P=0.049) were potential factors associated with CK19 expression, which were incorporated into the clinical-radiologic model. In the internal and external validation cohorts, the AUC of the clinical-radiologic model was 0.690 (95% CI 0.590-0.778) and 0.650 (95% CI 0.565-0.727), respectively. The habitat radiomics model based on CP images demonstrated the highest performance. It achieved AUC of 0.729 (95% CI 0.622-0.836) and 0.725 (95% CI 0.607-0.842) in the internal and external validation cohorts, respectively. Multivariate analysis identified AFP ( OR=2.494, 95% CI 1.163-5.348, P=0.019), AP enhancement ( OR=5.230, 95% CI 1.868-14.643, P=0.002) and habitat radiomics model score ( OR=4.105, 95% CI 2.643-6.368, P<0.001) as independent predictors of CK19 positivity. Based on these factors, a combined clinical-radiologic-habitat combined model was established. The clinical-radiologic-habitat combined model achieved AUCs of 0.767 (95% CI 0.671-0.846) and 0.730 (95% CI 0.649-0.801) in the internal and external validation cohorts, respectively. Significant differences in RFS were observed between the CK19-positive group (25.1 month) and CK19-negative group (51.0 month) as predicted by the clinical-radiologic-habitat model ( χ2=4.17, P=0.041). Conclusion:The clinical-radiologic-habitat combined model based on Gd-EOB-DTPA-enhanced MRI habitat radiomics demonstrates good predictive performance for CK19 expression in HCC and offers valuable prognostic stratification for clinical practice.

Objective:To explore the value of the deep learning model based on gadolinium ethoxybenzyl diethylenetriamine pentaacetic acid (Gd-EOB-DTPA) enhanced MRI in preoperatively predicting vessels completely encapsulating tumor clusters (VETC) in hepatocellular carcinoma (HCC).Methods:This study adopted a case-control design to retrospectively analyze 420 patients with HCC confirmed by postoperative pathology who underwent Gd-EOB-DTPA enhanced MRI between June 2016 and March 2023. A total of 420 patients were divided into a training set ( n=305) from the First Affiliated Hospital of Soochow University and an external validation set ( n=115) from Affiliated Nantong Hospital 3 of Nantong University. Based on postoperative pathological findings, patients were stratified into VETC-positive and VETC-negative groups. The training set comprised 161 VETC-positive cases and 144 VETC-negative cases, while the external validation set included 55 VETC-positive cases and 60 VETC-negative cases. Tumor regions of interest in arterial, portal venous, and hepatobiliary phases were manually delineated using ITK-SNAP software. Pre-trained Vgg19, Densenet121, and Vision Transformer (ViT) models were employed for transfer learning, extracting deep learning features from each image. Feature data were processed using FAE software, and 12 logistic regression models (arterial phase, portal venous phase, hepatobiliary phase, and combined three-phase models) were constructed to select the optimal deep learning model. Independent predictors in clinical characteristics were identified through univariate and multivariate logistic analyses to establish a clinical model for predicting VETC pattern. Subsequently, a clinical-deep learning fusion model was developed by integrating these clinical predictors with the optimal deep learning features. Model performance in predicting VETC-positive HCC was evaluated using receiver operating characteristic curves, calibration curves, and decision curve analysis (DCA). Results:In the external validation set, the area under the curve (AUC) of the Vgg19 model in the arterial phase, portal venous phase, hepatobiliary phase, and combined three-phase, respectively were 0.799,0.756,0.789,0.821, which were higher than those of Densenet121 (AUC: 0.544,0.581,0.544,0.583) and ViT (AUC: 0.740,0.752,0.785,0.767) model. The three-phase combined Vgg19 model achieved the highest AUC of 0.821 (95% CI 0.746-0.897). Multivariate logistic regression identified alpha-fetoprotein level ( OR=1.826,95% CI 1.069-3.120, P=0.028) and tumor diameter ( OR=1.329,95% CI 1.206-1.466, P<0.001) as independent predictors of VETC-positive HCC, forming the clinical model with an AUC of 0.789 (95% CI 0.703-0.859). The clinical-deep learning fusion model further achieved the AUC of 0.825 (95% CI 0.749-0.900). Calibration curves confirmed high concordance between predicted and actual probabilities for the three-phase Vgg19 model, while DCA revealed greater net clinical benefit for the combined Vgg19 and fusion models compared with the clinical model alone. Conclusions:The deep learning model based on Gd-EOB-DTPA enhanced MRI can be used to predict VETC of HCC preoperatively, among which the three-phase combined Vgg19 model and the clinical-deep learning model provide high predictive value.

Objective:To construct a model of mitochondria-related genes (Mito-RGs) in hepatocellular carcinoma (HCC), and predict the immune, prognostic and therapeutic characteristics of HCC based on the model, so as to provide a new idea for the diagnosis and treatment of HCC.Methods:The expression profiles of HCC and corresponding clinical information were obtained from the Cancer Genome Atlas (TCGA) database. Univariate Cox regression, Least Absolute Shrinkage and Selection Operator (LASSO) regression and multivariate Cox regression were used to construct a prognostic model of HCC based on Mito-RGs, and the International Cancer Genome Consortium-Liver Cancer-RIKEN-Japan ICGC-LIRI-JP dataset were used for validation. GO and KEGG analyses were performed to investigate the signaling pathways enriched for differentially expressed genes in the high- and low-risk groups. Immune infiltration was assessed using CIBERSORT. Single-cell data were used to study the proportion of immune cells in high- and low-risk groups of HCC samples and the relationship with cell proliferation. Cellminer was used to assess the relationship between risk score models and drug sensitivity.Results:A risk-prognostic model of HCC containing seven Mito-RGs ( DTYMK, ACADS, HMGCS2, CYP27A1, TOMM40L, STOM, and AKR1B10) was constructed. High-risk HCC patients had a worse prognosis. Genes upregulated in the high- and low-risk groups of differentially expressed genes were enriched in cell cycle and metabolism-related pathways. Single-cell data showed higher proportions of CD8 + T cells, macrophages and monocytes, and proliferating cells in the high-risk group. CIBERSORT analysis suggested that Treg cells and M0 macrophages were more abundant in the high-risk group, whereas CD8 + T cells and CD4 + memory T cells were less abundant. Patients in the high-risk group were more sensitive to myeloid cell leukemia sequence 1 inhibitor, vincristine, phosphatidylinositol kinase beta subunit inhibitor, and aurora kinase A, while trametinib, selumetinib, extracellular regulated protein kinase, and mitogen-activated extracellular signal-regulated kinase were more effective in the low-risk group. Conclusion:The constructed Mito-RGs model is capable of providing a more accurate assessment of the prognosis and the degree of immune cell infiltration in HCC patients.

Objective:To investigate the value of habitat radiomic features based on gadolinium ethoxybenzyl diethylenetriamine pentaacetic acid (Gd-EOB-DTPA)-enhanced MRI in establishing a predictive model for cytokeratin 19 (CK19) expression in hepatocellular carcinoma (HCC) and to evaluate its role in prognostic risk stratification.Methods:This multicenter case-control study retrospectively enrolled 489 patients with pathologically confirmed HCC who underwent Gd-EOB-DTPA-enhanced MRI between June 2016 and June 2024. Among them, 346 patients from the First Affiliated Hospital of Soochow University were divided into a training cohort ( n=245) and an internal test cohort ( n=101) via stratified sampling at a 7∶3 ratio. And 143 patients from Nantong Third Hospital Affiliated to Nantong University served as an external validation cohort. The training cohort included 53 CK19-positive and 192 CK19-negative patients. The internal test cohort included 21 CK19-positive and 80 CK19-negative patients. The external validation cohort included 30 CK19-positive and 113 CK19-negative patients. Univariate logistic regression analysis was performed to identify potential factors associated with CK19 expression, and a clinical-radiologic model was constructed. The k-means clustering algorithm was applied to segment target HCC lesions into 3 subregions. Radiomic features were extracted and selected from these habitat subregions. Habitat radiomics models were constructed for the arterial phase (AP), portal venous phase, hepatobiliary phase (HBP), and combined phases (CP). Multivariate logistic regression analysis identified independent clinical and radiologic predictors of CK19 expression, and the optimal habitat model score was integrated to build a clinical-radiologic-habitat combined model. The area under the receiver operating characteristic curve (AUC) was used to evaluate model predictive performance. Recurrence-free survival (RFS) was analyzed using the Kaplan-Meier method and the differences in survival curves were compared with the log-rank test. Results:Univariate logistic regression analysis revealed that alpha-fetoprotein (AFP) ( OR=2.629, 95% CI 1.412-4.896, P=0.002), AP enhancement ( OR=3.636, 95% CI 1.642-8.052, P=0.001), AP peritumoral enhancement ( OR=2.219, 95% CI 1.084-4.542, P=0.029), and HBP peritumoral hypointensity ( OR=2.010, 95% CI 1.004-4.021, P=0.049) were potential factors associated with CK19 expression, which were incorporated into the clinical-radiologic model. In the internal and external validation cohorts, the AUC of the clinical-radiologic model was 0.690 (95% CI 0.590-0.778) and 0.650 (95% CI 0.565-0.727), respectively. The habitat radiomics model based on CP images demonstrated the highest performance. It achieved AUC of 0.729 (95% CI 0.622-0.836) and 0.725 (95% CI 0.607-0.842) in the internal and external validation cohorts, respectively. Multivariate analysis identified AFP ( OR=2.494, 95% CI 1.163-5.348, P=0.019), AP enhancement ( OR=5.230, 95% CI 1.868-14.643, P=0.002) and habitat radiomics model score ( OR=4.105, 95% CI 2.643-6.368, P<0.001) as independent predictors of CK19 positivity. Based on these factors, a combined clinical-radiologic-habitat combined model was established. The clinical-radiologic-habitat combined model achieved AUCs of 0.767 (95% CI 0.671-0.846) and 0.730 (95% CI 0.649-0.801) in the internal and external validation cohorts, respectively. Significant differences in RFS were observed between the CK19-positive group (25.1 month) and CK19-negative group (51.0 month) as predicted by the clinical-radiologic-habitat model ( χ2=4.17, P=0.041). Conclusion:The clinical-radiologic-habitat combined model based on Gd-EOB-DTPA-enhanced MRI habitat radiomics demonstrates good predictive performance for CK19 expression in HCC and offers valuable prognostic stratification for clinical practice.

Objective:To explore the value of the deep learning model based on gadolinium ethoxybenzyl diethylenetriamine pentaacetic acid (Gd-EOB-DTPA) enhanced MRI in preoperatively predicting vessels completely encapsulating tumor clusters (VETC) in hepatocellular carcinoma (HCC).Methods:This study adopted a case-control design to retrospectively analyze 420 patients with HCC confirmed by postoperative pathology who underwent Gd-EOB-DTPA enhanced MRI between June 2016 and March 2023. A total of 420 patients were divided into a training set ( n=305) from the First Affiliated Hospital of Soochow University and an external validation set ( n=115) from Affiliated Nantong Hospital 3 of Nantong University. Based on postoperative pathological findings, patients were stratified into VETC-positive and VETC-negative groups. The training set comprised 161 VETC-positive cases and 144 VETC-negative cases, while the external validation set included 55 VETC-positive cases and 60 VETC-negative cases. Tumor regions of interest in arterial, portal venous, and hepatobiliary phases were manually delineated using ITK-SNAP software. Pre-trained Vgg19, Densenet121, and Vision Transformer (ViT) models were employed for transfer learning, extracting deep learning features from each image. Feature data were processed using FAE software, and 12 logistic regression models (arterial phase, portal venous phase, hepatobiliary phase, and combined three-phase models) were constructed to select the optimal deep learning model. Independent predictors in clinical characteristics were identified through univariate and multivariate logistic analyses to establish a clinical model for predicting VETC pattern. Subsequently, a clinical-deep learning fusion model was developed by integrating these clinical predictors with the optimal deep learning features. Model performance in predicting VETC-positive HCC was evaluated using receiver operating characteristic curves, calibration curves, and decision curve analysis (DCA). Results:In the external validation set, the area under the curve (AUC) of the Vgg19 model in the arterial phase, portal venous phase, hepatobiliary phase, and combined three-phase, respectively were 0.799,0.756,0.789,0.821, which were higher than those of Densenet121 (AUC: 0.544,0.581,0.544,0.583) and ViT (AUC: 0.740,0.752,0.785,0.767) model. The three-phase combined Vgg19 model achieved the highest AUC of 0.821 (95% CI 0.746-0.897). Multivariate logistic regression identified alpha-fetoprotein level ( OR=1.826,95% CI 1.069-3.120, P=0.028) and tumor diameter ( OR=1.329,95% CI 1.206-1.466, P<0.001) as independent predictors of VETC-positive HCC, forming the clinical model with an AUC of 0.789 (95% CI 0.703-0.859). The clinical-deep learning fusion model further achieved the AUC of 0.825 (95% CI 0.749-0.900). Calibration curves confirmed high concordance between predicted and actual probabilities for the three-phase Vgg19 model, while DCA revealed greater net clinical benefit for the combined Vgg19 and fusion models compared with the clinical model alone. Conclusions:The deep learning model based on Gd-EOB-DTPA enhanced MRI can be used to predict VETC of HCC preoperatively, among which the three-phase combined Vgg19 model and the clinical-deep learning model provide high predictive value.