The purpose of this study was to explore the feasibility of dual-source computed tomography (DSCT) high-pitch scan mode in the preoperative evaluation of severe aortic stenosis (AS) referred to transcatheter aortic valve implantation (TAVI). Thirty patients with severe AS referred for TAVI underwent cervico-femoral artery joint DSCT angiography. Measurement and calculation of contrast, contrast noise ratio (CNR) and noise of aorta and access vessels were performed. The intra-and inter-observer reproducibilities for assessing aortic root and access vessels were evaluated. Evaluation of shape and plagues of aorta and access vessels was performed. The contrast, CNR and noise of aorta and access vessels were 348.2~457.9 HU, 12.2~30.3 HU and 19.1~48.1 HU, respectively. There were good intra-and inter-observer reproducibilities in assessing aortic root and access vessels by DSCT (mean difference:-0.73~0.79 mm, r=0.90~0.98, P < 0.001; mean difference:-0.70~0.73 mm, r=0.90~0.96, P < 0.001). In the 30 patients, the diameters of external iliac artery, femeral artery or subclavian artery were less than 7 mm in 5 cases (16.7%), marked calcification in bilateral common iliac arteries in 1 case (3.3%) and marked soft plaque in left common iliac artery in 1 case (3.3%). DSCT high-pitch scan mode was feasible in the preoperative evaluation of aorta and access vessels in patients with AS referred for TAVI.
This study aims to predict expression of Ki67 molecular marker in pancreatic cystic neoplasm using radiomics. We firstly manually segmented tumor area in multi-detector computed tomography (MDCT) images. Then 409 high-throughput features were automatically extracted and the least absolute shrinkage selection operator (LASSO) regression model was used for feature selection. After 200 bootstrapping repetitions of LASSO, 20 most frequently selected features made up the optimal feature set. Then 200 bootstrapping repetitions of support vector machine (SVM) classifier with 10-fold cross-validation were used to avoid overfitting and accurately predict the Ki67 expression. The highest prediction accuracy could achieve 85.29% and the highest area under the receiver operating characteristic curve (AUC) was 91.54% with a sensitivity (SENS) of 81.88% and a specificity (SPEC) of 86.75%. According to the results of experiment, the feasibility of predicting expression of Ki67 in pancreatic cystic neoplasm based on radiomics was verified.
