Objective To discuss the value of clinical radiomic nomogram(CRN)and deep convolutional neural network(DCNN)in distinguishing atypical pulmonary hamartoma(APH)from atypical lung adenocarcinoma(ALA).Methods A total of 307 patients were retrospectively recruited from two institutions.Patients in institu-tion 1 were randomly divided into the training(n=184:APH=97,ALA=87)and internal validation sets(n=79:APH=41,ALA=38)in a ratio of 7∶3,and patients in institution 2 were assigned as the external validation set(n=44:APH=23,ALA=21).A CRN model and a DCNN model were established,respectively,and the performances of two models were compared by delong test and receiver operating characteristic(ROC)curves.A human-machine competition was conducted to evaluate the value of AI in the Lung-RADS classification.Results The areas under the curve(AUCs)of DCNN model were higher than those of CRN model in the training,internal and external validation sets(0.983 vs 0.968,0.973 vs 0.953,and 0.942 vs 0.932,respectively),however,the differences were not statistically significant(p=0.23,0.31 and 0.34,respectively).With a radiologist-AI com-petition experiment,AI tended to downgrade more Lung-RADS categories in APH and affirm more Lung-RADS cat-egories in ALA than radiologists.Conclusion Both DCNN and CRN have higher value in distinguishing APH from ALA,with the former performing better.AI is superior to radiologists in evaluating the Lung-RADS classification of pulmonary nodules.

Objective To explore the value of combined prediction model based on clinical and CT radiomics features in discriminating atypical pulmonary hamartoma(APH)from atypical lung adenocarcinoma(ALA).Methods A total of 290 patients with APH and ALA confirmed by pathology were retrospectively selected.250 patients from the First Affiliated Hospital of Anhui Medical University were randomly assigned into a training set(APH=91,ALA=84)and an internal validation set(APH=39,ALA=36)at a ratio of 7∶3,and other 40 patients from the First Affiliated Hospital of USTC were assigned as an external validation set(APH=21,ALA=19).The independent model and multivariate logistic regression combined model were constructed using the selected clinical-CT features and radiomics features,respectively,and a nomogram was drawn.Receiver operating characteristic(ROC)curve and DeLong test were used to evaluate and compare the performances of the models.Results The area under the curve(AUC)of the combined model established by 3 clinical-CT features and 4 radiomics features in the training set was 0.980,which was higher than that of clinical-CT model(AUC=0.885,P＜0.001)and radiomics model(AUC=0.975,P=0.042).The AUC of the combined model in the internal and external validation sets(0.963 vs 0.917)were also higher than those of clinical-CT model(0.858 vs 0.774)and radiomics model(0.953 vs 0.897),respectively.Conclusion The combined prediction model based on clinical and CT radiomics features can improve the differential diagnosis ability of APH and ALA.

Objective To investigate the value of a nomogram model based on contrast-enhanced CT radiomics in predicting the histological type of thymoma.Methods A total of 154 patients(101 in low-risk group and 53 in high-risk group)with thymoma confirmed by pathology were retrospectively selected.The cases were randomly divided into training set(n=107)and validation set(n=47)at a ratio of 7∶3.The three-dimensional volume of interest(VOI)of the whole lesion on the image from the arterial phase of contrast-enhanced CT was manually delineated,and the radiomics features were extracted.Based on the selected radiomics features,the radiomics model was constructed and the model Radiomics score(Radscore)was calculated.Clinical risk factors were screened to construct a clinical model,and a nomogram model was constructed by fusing Radscore and clinical risk factors.The receiver operating characteristic(ROC)curve,area under the curve(AUC),accuracy,sensitivity and specificity were compared to analyze the predictive efficacy and difference of different models for high-risk and low-risk thymoma.The decision curve and calibration curve were drawn to evaluate the clinical value and fitting performance of the nomogram model.Results Eleven radiomics features were selected to construct the radiomics model,and five clinical risk factors[myasthenia gravis(MG),morphology,border,surrounding tissue invasion and CT value in arterial phase]were used to construct the clinical model.In the training set,the AUC of the nomogram model(0.88)was higher than that of the radiomics model(0.80)and the clinical model(0.79),and the difference was statistically significant(Z=2.233,2.713,P=0.026,0.007,respectively).In the validation set,the AUC of the nomogram model was higher than that of the radiomics and clinical models,but the difference was not statistically significant.The calibration curve showed that the nomogram model had good fitting performance,and the decision curve showed that the nomogram model had high clinical benefit.Conclusion The nomogram model based on contrast-enhanced CT can effectively predict high-risk and low-risk thymoma,which is helpful to guide clinicians to make relevant decisions.

Objective:To investigate the value of nomograms based on clinical parameters, apparent diffusion coefficient (ADC) and MRI-derived radiomics in predicting survival of patients with locally advanced cervical cancer (LACC) after concurrent chemoradiotherapy (CCRT).Methods:Clinical data of 423 patients with IB-IVA cervical cancer treated with CCRT at Anhui Provincial Hospital Affiliated to Anhui Medical University from March 2014 to March 2020 were retrospectively analyzed and randomly divided into the training and validation groups at a ratio of 2∶1 using the simple randomization method. The values of ADC min, ADC mean, ADC max and 3D texture parameters of diffusion weighted imaging (DWI), T 2WI, T 2WI-fat suppression of pre-treatment primary lesions in all patients were measured. The least absolute shrinkage and selection operator (LASSO) algorithm and logistic regression analysis were used to screen the texture features and calculate radiomics score (Rad-score). Cox regression analysis was employed to construct nomogram models for predicting overall survival (OS) and cancer-specific survival (CS) of patients with LACC after CCRT, which were subject to internal and external validation. Results:Squamous cell carcinoma antigen (SCC-Ag), external beam radiotherapy dose, ADCmin and Rad-score were the independent prognostic factors for OS and CS of LACC patients after CCRT and constituted predictive models for OS and CS. The area under the receiver operating characteristic (ROC) curve (AUC) of two models in predicting 1-year, 3-year, 5-year OS and CS was 0.906, 0.917, 0.916 and 0.911, 0.918, 0.920, with internally validated consistency indexes (C-indexes) of 0.897 and 0.900. Then, models were brought into the validation group for external validation with AUC of 0.986, 0.942, 0.932 and 0.986, 0.933, 0.926 in predicting 1-year, 3-year, 5-year OS and CS.Conclusion:The nomograms based on clinical parameters, ADC values and MRI-derived radiomics are of high clinical value in predicting OS and CS of patients with LACC after CCRT, which can be used as prognostic markers for patients with cervical cancer to certain extent.

Objective:To investigate the value of nomogram based on intravoxel incoherent motion diffusion weighted imaging (IVIM-DWI) and MRI-derived radiomics for predicting recurrence after concurrent chemoradiotherapy (CCRT) in patients with locally advanced cervical cancer (LACC).Methods:Clinical data of 111 patients with ⅠB-ⅣA cervical cancer who underwent CCRT at Anhui Provincial Hospital from December 2014 to December 2019 and were continuously followed up were retrospectively analyzed. Pre-treatment IVIM-DWI parameters (ADC, D, D * and f) and pre- and post-treatment 3D texture parameters (from axial T 2WI) of the primary lesions were measured. Least absolute shrinkage and selection operator (LASSO) algorithm and multivariate logistic regression analysis were used to filter texture features and calculate radiomics score (Rad-score). A Cox regression model was used to analyze independent risk factors for recurrence after CCRT in patients with LACC and construct a nomogram. Results:External beam radiotherapy dose, f value , pre-treatment Rad-score and post-treatment Rad-score ( HR=0.204, 3.253, 2.544, 7.576) were the independent prognostic factors for recurrence after CCRT in cervical cancer patients and jointly formed the nomogram. The area under curve (AUC) of the nomogram for predicting 1-, 3- and 5-year disease-free survival (DFS) was 0.895, 0.888 and 0.916, with internal validation C-indexes of 0.859, 0.903 and 0.867, respectively. The decision curves analysis showed that the nomogram has a higher net clinical benefit compared to other models, and the clinical impact curves further visualized its predictive accuracy. Conclusions:The nomogam based on IVIM-DWI and radiomics has high clinical value in predicting recurrence after CCRT in patients with LACC, providing reference for prognostic assessment and individualized treatment of cervical cancer patients.

Objective:To investigate the value of intravoxel incoherent motion diffusion weighted imaging (IVIM-DWI) parameters combined with T 2WI texture analysis of primary lesions of rectal adenocarcinoma in preoperative prediction of lymph node metastasis with short diameter ≤9 mm. Methods:Retrospective analysis was performed on 115 cases of rectal adenocarcinoma confirmed by surgical pathology in Affiliated Provincial Hospital of Anhui Medical University from June 2015 to October 2020. All patients underwent total mesorectal resection and received conventional rectal MRI and IVIM-DWI scan before surgery. According to the pathological results of lymph node, the patients were divided into lymph node metastatic group ( n=44) and non-metastatic group ( n=71). IVIM-DWI parameters of primary rectal adenocarcinoma were measured including apparent diffusion coefficient (ADC), diffusion coefficient (D), pseudo diffusion coefficient (D *) and perfusion fraction (f). The region of interest (ROI) of the whole lesion of rectal adenocarcinoma was delineated on axial T 2WI; then the ROIs were imported into GE Analysis Kit software to extract 3D texture feature. The differences of IVIM-DWI parameters and texture feature parameters were compared between two groups using independent sample t test or Mann-Whitney U test. The optimal texture feature parameters with independent predictive function were screened by multivariate logistic regression. Then the texture feature model and combined model based IVIM-DWI and texture feature parameters were established. The receiver operating characteristic (ROC) curves were used to evaluate the performances of IVIM-DWI, texture feature parameters, texture feature model and combined model in predicting lymph node metastasis in patients with rectal adenocarcinoma. The area under the ROC curve (AUC) were compared with DeLong test. Results:Among all the IVIM-DWI parameters, the D * and f values of primary rectal adenocarcinoma were significantly different between the lymph node metastasis group and the non-lymph node metastasis group ( Z=3.39, P=0.001, Z=-3.06, P=0.002); no statistical significance was found in the ADC and D values between two groups (both P>0.05). A total of 828 texture feature parameters were obtained based on T 2WI of primary lesion of rectal adenocarcinoma, among which 3 optimal texture feature parameters were selected, including firstorder_Skewness, shape_Sphericity and glcm_Idn. The ROC curve results showed that the AUC of D * and f were 0.689 and 0.670, respectively. The AUC of 3 texture feature parameters were 0.651, 0.628, 0.631, respectively. The AUC of texture feature model and the combined model were 0.775 and 0.803. The AUC of combined model was larger than D *, f and the three texture feature parameters (all P<0.05). Conclusion:IVIM-DWI parameters combined with T 2WI texture feature parameters in primary lesion of rectal adenocarcinoma show good diagnostic efficacy in preoperative prediction of lymph node metastasis with short diameter≤9 mm.

Objective:To investigate the value of ADC derived from DWI combined with texture analysis derived from T 2WI fat suppressed images in distinguishing benign and malignant soft tissue tumors. Methods:The MRI and DWI images of 94 patients with soft tissue tumors (44 cases with malignant and 50 cases with benign) confirmed by pathology were analyzed retrospectively in the First Affiliated Hospital of USTC West District. ADC values of solid components were measured at GE ADW4.6 workstation. The texture features were extracted by manually drawing the ROI on the maximum level of the T 2WI fat suppressed images; the ADC values and texture parameters between the two groups were statistically analyzed by SPSS17.0, and the multivariate logistic regression model were conducted to analyze and calculate the diagnostic performance. Results:ADC value of benign and malignant soft tissue tumors was (1.6±0.3)×10 -3 mm 2/s, (1.2±0.5)×10 -3 mm 2/s, respectively, and the difference was statistically significant( t=-5.382, P<0.05). Taking 1.28×10 -3 mm 2/s as the critical value, the area under curve (AUC) for the diagnosis of benign and malignant soft tissue tumors was 0.783, the sensitivity was 92.00%, and the specificity was 65.91%. Among the texture features, the AUC of frequency size, skewness, Inertia All Direction_offset7, Inverse Difference Moment angle0_offset1, Inverse Difference Moment angle0_offset7 and Haralick Correlation All Direction_offset4_SD distinguishing benign and malignant soft tissue tumors were 0.825, 0.739, 0.826, 0.816, 0.820 and 0.783, respectively. The AUC, sensitivity and specificity of the best predictive model distinguishing benign and malignant soft tissue tumors were 0.930, 88.00% and 86.36% respectively using multivariate logistic regression analysis. Conclusion:ADC combined with texture analysis is of great value in preoperative differentiation of benign and malignant soft tissue tumors.

Objective:To investigate the value of quantitative CT (QCT) body component parameters before and after transcatheter arterial chemoembolization (TACE) as prognostic indicator for patients with hepatic cell carcinoma (HCC).Methods:Retrospective analysis was performed on 40 patients with advanced HCC who received TACE treatment in Anhui Provincial Hospital Affiliated to Anhui Medical University from November 2013 to May 2017, all of them received QCT scanning before and after treatment. The information were recorded, including gender, age, alpha-fetoprotein (AFP), TNM stage, liver function Child-Pugh grade, portal venous thromboembolism, cirrhosis, maximum tumor diameter, tumor type, and frequency of interventional therapy. QCT parameters were measured before and after treatment, including L1, L2 bone mineral density (BMD), L3-level paravertebral muscle area (MA), subcutaneous fat area (SFA) and visceral fat area (VFA), and the change rate of QCT parameters (ΔBMD, ΔMA, ΔSFA, ΔVFA) before and after TACE were calculated after the QCT scan interval was standardized. The cut-off values of ΔBMD, ΔMA, ΔSFA and ΔVFA to diagnose the prognosis of HCC patients after TACE were obtained by drawing the ROC curves. The Kaplan-Meier method was used to calculate the survival rate, the Log-rank method was used for univariate analysis, and the Cox regression analysis model was used for multivariate analysis to screen out independent factors affecting the prognosis of HCC patients after TACE.Results:ROC curve analysis showed that the cut-off values of ΔBMD, ΔMA, ΔSFA and ΔVFA to diagnose the prognosis of HCC patients after TACE were -8.64%, -6.84%, -9.84% and 5.70%, respectively. Univariate analysis showed that AFP, TNM stage, liver function Child-Pugh grade, portal venous thrombosis, tumor type and ΔMA, ΔSFA, ΔVFA had statistically significant effects on prognosis ( P<0.1). Multivariate analysis showed that ΔMA, ΔVFA and portal venous thromboembolism were independent influencing factors for the prognosis of HCC patients after TACE treatment ( P<0.05). Conclusions:ΔMA, ΔVFA and portal venous thromboembolism have reference value for prognosis assessment of TACE treatment for HCC patients, and QCT body composition analysis is helpful to evaluate the prognosis of HCC patients.

Objective:To evaluate the application value of intravoxel incoherent motion diffusion weighted imaging (IVIM-DWI) and dynamic contrast enhancement MRI (DCE-MRI) in the prediction of the early efficacy of concurrent chemoradiotherapy (CCRT) for cervical squamous cell carcinoma.Methods:Fifty patients with cervical squamous cell carcinoma confirmed by pathology were included. Before CCRT, IVIM-DWI and DCE-MRI were scanned, and the values of quantitative parameters including ADC, D, D * and f of IVIM-DWI and K trans, K ep, V e and V p of DCE-MRI before treatment were measured for all patients. MRI reexamination was performed 1 month after the end of CCRT, and all patients were divided into the cure group and the residual group according to the tumor remission. The parameters of IVIM-DWI and DCE-MRI before treatment were statistically compared between two groups. The optimal predictive parameters and predictive thresholds were determined by drawing the receiver operating characteristic (ROC) curve. Results:Twenty-four patients were assigned into the cure group and twenty-six patients in the residual group. The ADC, D and V e values before treatment in the cure group were significantly lower than those in the residual group (all P<0.05), whereas the f and K trans values were significantly higher than those in the residual group (both P<0.05). The other parameters did not significantly differ between two groups (all P>0.05). The area under ROC curve (AUC=0.823) of D value was the largest, followed by K transvalue (AUC=0.754). The combined prediction efficacy of D and K trans (AUC=0.867) was higher than that of either D or K trans alone. The sensitivity was 88.5%, 85.8% and 88.8%, and the specificity was 70.8%, 66.7% and 79.2%, respectively. Conclusions:Quantitative parameters of IVIM-DWI and DCE-MRI before treatment have certain predictive value for the early efficacy of CCRT in cervical squamous cell carcinoma, among which the predictive efficacy of D value is the highest, and the combined application of D and K trans can improve the predictive efficacy.

Objective To explore the classic MRI appearance of secondary hemochromatosis (SHC)related liver iron overload, and the feasibility of quantitative evaluation of liver iron overload by iterative decomposition of water and fat with echo asymmetry and leastGsquares estimationGiron quantification (IDEALGIQ).Methods 20 patients with SHCGrelated liver iron overload (experimental group)and 20 healthy adults (control group)underwent routine liver MRI and IDEALGIQ.The MRI images were comparatively analyzed to assess the hallmark of liver iron overload.In both two groups,the R2 ? values were measured on R2 ? maps,which were generated by IDEALGIQ,then the differences in age,gender and R2 ? value between two groups were comparatively analyzed.In experimental group,the serum ferritin (SF)was detected,and a correlation analysis was tested with R2 ? value.Results For all of the 20 patients, there was signal drop of liver parenchyma on T1 and T2 Gweighted images,signal loss with susceptibility artifact on DWI images,and signal dropped on inGphase images relative to outGofGphase images.Among the 20 patients,18 cases appeared "a dark liver parenchyma"on T2 G weighted images,and the spleen signal in 3 cases was similar to liver parenchyma’s hallmark.The R2 ? values in experimental group and control group were 395.58±255.75 Hz and 41.18±7.86 Hz (t＝－6.12,P＝0.00),respectively.No significant differences between two groups were found in gender and age (χ2＝0.10,P＝0.10 and t＝0.09,P＝0.93).The liver iron overload R2? value was not correlated with SF (r＝0.1 5 3 , P＝0.15).Conclusion On MRI,the typical appearance of liver iron overload is hypointense on T1 and T2Gweighted images,especially"a dark liver parenchyma"on T2 WI,signal drops on inGphase images relative to outGofGphase images,and signal loss with susceptibility artifact on DWI images.R2 ? value of IDEALGIQ can quantitatively evaluate the liver iron overload,without a correlation with SF.