Objective:To investigate the clinical value of the prediction models constructed by CT based imaging features and radiomics for World Health Organization/International Society of Urological Pathology (WHO/ISUP) grading in pre-operative patients with T1 clear cell renal cell carcinoma (ccRCC).Methods:Ninety patients with ccRCC diagnosed at Hangzhou Hospital of Traditional Chinese Medicine from January 2016 to December 2023 were enrolled as the training set, and 43 patients diagnosed at the Sir Run Run Shaw Hospital from January 2017 to December 2018 were enrolled as the external validation set. According to the WHO/ISUP grading system, grades Ⅰ and Ⅱ were defined as the low grade group, and grades Ⅲ and Ⅳ were defined as the high grade group. In the training set, 64 patients were in the low grade group and 26 patients in the high grade group. In the external validation set, 33 patients were in the low grade group and 10 patients in the high grade group. The multivariate logistic regression was used to establish an imaging factor model based on CT imaging features in the training set. The 3-dimensional regions of interest were manually contoured at the cortical phase of enhanced CT, and the radiomics features were extracted. Linear correlation between features and L1 regularization were used for feature selection, and then linear support vector classification was used to construct the radiomics model. After that, a combined diagnostic model of nomogram combining the radiomics score and imaging factors was constructed using multivariate logistic regression analysis. The receiver operating characteristic (ROC) curve was used to evaluate the effectiveness of each model. The Delong test was used for comparison of the areas under the ROC curve.Results:The imaging factor model, the radiomics model, and the combined diagnostic model of nomogram were successfully constructed to predict the WHO/ ISUP grading in stage T1 ccRCC. The AUC value of the imaging factor model in the training and validation sets was 0.742 (95% CI: 0.623-0.860) and 0.664 (95% CI: 0.448-0.879), respectively. The AUC values of the radiomics model in the two sets were 0.914 (95% CI: 0.844-0.983) and 0.879 (95% CI: 0.718-1.000), and of the combined diagnostic model of nomogram in the two sets were 0.929 (95% CI: 0.858-0.999) and 0.882 (95% CI: 0.710-1.000), respectively. The AUCs of the radiomics model and combined diagnostic model of nomogram were significantly higher than that of the imaging factor model (both P＜0.05). There was no statistical difference in the AUCs between the combined diagnostic model of nomogram and the radiomics model (both P＞0.05). Conclusion:The CT-based radiomics model and combined diagnostic model of nomogram incorporating radiomics signature and imaging features showed favorable predictive efficacy for the preoperative prediction of WHO/ISUP grading in stage T1 ccRCC.

[Objective]To investigate the value of radiomics based on contrast-enhanced computed tomography(CT)image in predicting the traditional Chinese medicine(TCM)differentiation typing of primary non-small cell lung cancer(NSCLC).[Methods]A total of 130 patients diagnosed as NSCLC by pathology from July 2018 to October 2023 in Hangzhou TCM Hospital Affiliated to Zhejiang Chinese Medical University were retrospectively analyzed.According to the diagnostic criteria of TCM,all the enrolled patients were divided into deficiency syndrome group(67 cases)and excess syndrome group(63 cases),and then assigned to training cohort(91 cases)and validation cohort(39 cases)in a ratio of 7:3.The largest diameter slice of lesion on cross-sectional images was selected and the regions of interest were contoured at unenhanced,arterial and venous phases respectively,and then the radiomics features were extracted.The linear correlation among features and L1 regularization were used for feature selection,and then logistic regression was used to construct the radiomics model based on radiomics features of each phase.The receiver operating characteristic(ROC)curve was used to evaluate the effectiveness of the model in predicting deficiency and excess syndromes of NSCLC.The Delong test was used for comparison of area under curve(AUC)between the two models.[Results]In the training cohort,a total of 7 radiomics models were constructed,including three single-phase radiomics models,three two-phase combination radiomics models and one three-phase combination radiomics model.The AUC of combination radiomics model was higher than that of the single-phase radiomics model.The AUC of three-phase combination radiomics model was the largest,which was 0.876[95%confidence interval(CI)(0.807～0.945)]and 0.755[95%CI(0.603～0.908)]in the training cohort and validation cohort respectively.[Conclusion]The radiomics model based on contrast-enhanced CT image has high efficacy in predicting the TCM differentiation typing of NSCLC,and the three-phase combination radiomics model demonstrates the best diagnostic efficacy.

[Objective]To investigate the value of radiomics based on contrast-enhanced computed tomography(CT)image in predicting the traditional Chinese medicine(TCM)differentiation typing of primary non-small cell lung cancer(NSCLC).[Methods]A total of 130 patients diagnosed as NSCLC by pathology from July 2018 to October 2023 in Hangzhou TCM Hospital Affiliated to Zhejiang Chinese Medical University were retrospectively analyzed.According to the diagnostic criteria of TCM,all the enrolled patients were divided into deficiency syndrome group(67 cases)and excess syndrome group(63 cases),and then assigned to training cohort(91 cases)and validation cohort(39 cases)in a ratio of 7:3.The largest diameter slice of lesion on cross-sectional images was selected and the regions of interest were contoured at unenhanced,arterial and venous phases respectively,and then the radiomics features were extracted.The linear correlation among features and L1 regularization were used for feature selection,and then logistic regression was used to construct the radiomics model based on radiomics features of each phase.The receiver operating characteristic(ROC)curve was used to evaluate the effectiveness of the model in predicting deficiency and excess syndromes of NSCLC.The Delong test was used for comparison of area under curve(AUC)between the two models.[Results]In the training cohort,a total of 7 radiomics models were constructed,including three single-phase radiomics models,three two-phase combination radiomics models and one three-phase combination radiomics model.The AUC of combination radiomics model was higher than that of the single-phase radiomics model.The AUC of three-phase combination radiomics model was the largest,which was 0.876[95%confidence interval(CI)(0.807～0.945)]and 0.755[95%CI(0.603～0.908)]in the training cohort and validation cohort respectively.[Conclusion]The radiomics model based on contrast-enhanced CT image has high efficacy in predicting the TCM differentiation typing of NSCLC,and the three-phase combination radiomics model demonstrates the best diagnostic efficacy.

Objective:To investigate the clinical value of the prediction models constructed by CT based imaging features and radiomics for World Health Organization/International Society of Urological Pathology (WHO/ISUP) grading in pre-operative patients with T1 clear cell renal cell carcinoma (ccRCC).Methods:Ninety patients with ccRCC diagnosed at Hangzhou Hospital of Traditional Chinese Medicine from January 2016 to December 2023 were enrolled as the training set, and 43 patients diagnosed at the Sir Run Run Shaw Hospital from January 2017 to December 2018 were enrolled as the external validation set. According to the WHO/ISUP grading system, grades Ⅰ and Ⅱ were defined as the low grade group, and grades Ⅲ and Ⅳ were defined as the high grade group. In the training set, 64 patients were in the low grade group and 26 patients in the high grade group. In the external validation set, 33 patients were in the low grade group and 10 patients in the high grade group. The multivariate logistic regression was used to establish an imaging factor model based on CT imaging features in the training set. The 3-dimensional regions of interest were manually contoured at the cortical phase of enhanced CT, and the radiomics features were extracted. Linear correlation between features and L1 regularization were used for feature selection, and then linear support vector classification was used to construct the radiomics model. After that, a combined diagnostic model of nomogram combining the radiomics score and imaging factors was constructed using multivariate logistic regression analysis. The receiver operating characteristic (ROC) curve was used to evaluate the effectiveness of each model. The Delong test was used for comparison of the areas under the ROC curve.Results:The imaging factor model, the radiomics model, and the combined diagnostic model of nomogram were successfully constructed to predict the WHO/ ISUP grading in stage T1 ccRCC. The AUC value of the imaging factor model in the training and validation sets was 0.742 (95% CI: 0.623-0.860) and 0.664 (95% CI: 0.448-0.879), respectively. The AUC values of the radiomics model in the two sets were 0.914 (95% CI: 0.844-0.983) and 0.879 (95% CI: 0.718-1.000), and of the combined diagnostic model of nomogram in the two sets were 0.929 (95% CI: 0.858-0.999) and 0.882 (95% CI: 0.710-1.000), respectively. The AUCs of the radiomics model and combined diagnostic model of nomogram were significantly higher than that of the imaging factor model (both P＜0.05). There was no statistical difference in the AUCs between the combined diagnostic model of nomogram and the radiomics model (both P＞0.05). Conclusion:The CT-based radiomics model and combined diagnostic model of nomogram incorporating radiomics signature and imaging features showed favorable predictive efficacy for the preoperative prediction of WHO/ISUP grading in stage T1 ccRCC.

Objective To explore the value of artificial intelligence(AI)in the diagnosis of pulmonary nodules in terms of consistency and efficiency compared with two radiologists(physician 1 is a chief physician and physician 2 is a deputy chief physician)in the diagnosis of benign and malignant pulmonary nodules using computed tomography(CT).Methods Retrospective analysis of 201 patients with pulmonary nodules confirmed by surgery pathology at Hangzhou Municipal Hospital affiliated to Zhejiang Chinese Medical University from January 2021 to October 2022,including a total of 229 pulmonary nodules,of which 74 were benign and 155 were malignant.The consistency of AI diagnosis with two radiologists was evaluated by weighted Kappa test,and the diagnostic performance of AI with the two radiologists was evaluated by the receiver operating characteristic curve(ROC).Results In the diagnosis of the benign and malignant nature of partial solid nodules,ground-glass nodules,solid nodules,and partial ground-glass and solid plus ground-glass nodules,the consistency between AI and physician 2 was higher than that between AI and physician 1.Additionally,the area under the curve(AUC)of physician 2 was higher than that of AI and physician 1 with statistically significant differences between the AUCs of ground-glass nodules,solid nodules,and partial ground-glass and solid plus ground-glass nodules(P<0.05).In the diagnosis of the benign and malignant nature of partial solid nodules and ground-glass nodules,the AUC of physician 1 was higher than that of AI,but there was no statistically significant difference between the two(P>0.05).In the diagnosis of the benign and malignant nature of solid nodules and partial ground-glass and solid plus ground-glass nodules,the AUC of AI was higher than that of physician 1 with statistically significant differences between the two(P<0.05).In the diagnosis of the benign and malignant nature of ground-glass nodules,solid nodules,and partial ground-glass and solid plus ground-glass nodules,AI's sensitivity(97%,92%,and 94%)was higher than that of physician 1(58%,89%,and 72%)and physician 2(83%,84%,and 85%).Conclusion AI has a certain diagnostic efficacy in the diagnosis of pulmonary nodules malignancy.The overall diagnostic efficacy of the AI system used in this study is between that of physician 1 and physician 2,but its sensitivity is higher than that of the latter two.

Objective To investigate CT and MRI characteristics of liver primary undifferentiated pleomorphic sarcoma (UPS).Methods Data of 8 cases of pathologically confirmed liver UPS were analyzed retrospectively.Results The tumors located in right liver (n=6),left liver (n=1) or caudate lobe (n=1).Irregular shape and ill-defined boundary were found in 6 cases,while regular shape and well-defined boundary were found in the other 2 cases.The lesions in 7cases showed heterogeneous density,and demonstrated homogeneous density in 1 case.Slight heterogeneous enhancement in arterial phase and progressive enhancement in portal phase were observed in all 8 cases,while continuous enhancement was observed in 6 cases.Tumor thrombi in inferior vena or portal vein were observed in 4 cases.Conclusion CT and MRI manifestations of primary liver UPS are characteristic in certain degree,therefore being helpful to the diagnosis of this disease.

PurposeIt has been reported that women have higher incidence of Modic changes than men and it may be related to the change of female hormone levels during menopause which leads to osteoporosis and other factors. This paper investigated the relationship between vertebral bone mineral density (BMD) of menopausal female suffering from chronic low pain and lumbar vertebral Modic changes on MRI, to explore the effect of vertebral bone mineral density upon Modic changes.Materials and Methods A total of 205 menopausal women with chronic low back pain were enrolled and underwent vertebral bone mineral density measurement and lumbar MRI examination. The bone mass of vertebral body and bone imaging data were observed. All patients were divided into three groups according to their level of bone mass: group of normal bone mass: 128 cases; osteopenia group: 58 cases; osteoporosis group: 19 cases. The incidence rate of Modic changes was compared among the three groups and the relationship between bone mineral density and vertebral Modic changes was further analyzed.Results Among 205 patients, 128 were with normal bone mass, 44 had Modic changes (type I: 19 cases; type II: 22 cases; type III: 3 cases) and the incidence rate was 34.4%; osteopenia occurred in 58 patients, among whom 34 had Modic changes (type I: 15 cases; type II: 17 cases; type III: 2 cases), which showed that the rate was 58.6%; 19 patients presented osteoporosis, 15 of whom appeared Modic changes (type I: 6 cases, type II: 7 cases;type III: 2 cases), with the rate of 78.9%. There was statistically signiifcant difference in incidence rate of Modic changes among the three groups (χ2=18.995,P<0.05). Pearson column connection numberC=0.29<0.40. The osteopenia group and osteoporosis group both had higher incidence rates than the group of normal bone mass (χ2=9.636 and 13.680,P<0.01), and the incidence rate showed no difference between the osteopenia group and osteoporosis group (χ2=2.555,P>0.05).Conclusion Lumbar vertebral bone mineral density is correlated to the incidence of vertebral Modic changes in menopausal women with chronic low back pain. With the loss of vertebral bone mass, the incidence of vertebral Modic changes gradually rise. However, the correlation is rather weak; Modic change is a dynamic process, which is also influenced by other factors except vertebral bone mineral density.