The Value Of Texture Analysis Technology Based On Enhanced CT Images In The Diagnosis Of Pancreatic Cancer And Liver Metastasis

Objective To explore the value of texture analysis based on CT enhanced images in predicting the pathological degree of differentiation of pancreatic carcinoma(PC).MethodsEighty-three patients with PC who went through postoperative pathology diagnose and CT examination were selected at the Attached Provincial Hospital of Anhui Medical University.Among them,34 cases were moderately differentiated,13 cases were poorly differentiated and 36 cases were moderately poorly differentiated.The images in the arterial and venous phases with the lesions at their largest cross section were selected to manually outline the region of interest(ROI)to delineate lesions using open-source software.A total of 396 features were extracted from the ROI using AK software.Spearman correlation analysis and random forest selection by filter(rf SBF)in the caret package of R studio were used to select the discriminating features.ResultsTwelve features are placed in the arterial phases and comprehensive group,six features placed in the venous phases.In the arterial phase,the areas under the ROC curve(AUC)of diagnosing poorly differentiated,moderately differentiated and moderate-poorly differentiated cases,0.80,1,and 0.80 were presented in the training group respectively and0.77,1,and 0.77 were shown in the test group respectively;in the venous phases,the values were 0.81,1,and 0.82 in the training group while 0.74,1,and 0.74 were in the test group.ConclusionTexture analysis based on contrast-enhanced CT images can be used as an adjunct for the preoperative assessment of the pathological degrees of differentiation of PC.ObjectiveTo explore the value of imaging omics features based on CT enhanced images in evaluating liver metastasis of pancreatic cancer.MethodsA retrospective analysis of the data of 84 patients who were pathologically confirmed as pancreatic cancer at the Attached Provincial Hospital of Anhui Medical University.Since the incipient stage of the patients undergoing enhanced CT examination,40 cases of liver metastasis occurred within 6 months,during which 29 cases occurred in 1 month,11 cases had liver metastasis within 1-6 months;46 cases had no liver metastasis in 6 months.The CT images including images of the arterial phase,the venous phase and the three-dimensional range of the lesion is selected to manually outline the region of interest(ROI).Use ITK-SNAP software(www.itksnap.org)to select three-dimensional images of patients’ arterial and venous CT lesions for ROI delineation and segmentation.After delineating the images,the processed images are obtained,and the delineated images are standardized and processed by GE.AK software(GE health care,Analysis Kit,Version:3.2.0.R)extracts relevant features of the processed image.The delineation software that we used is open-source software,and the Mann-Whitney U test is used for the extracted features to screen out the optimal feature subsets in each phase.And based on the obtained lesion-related characteristics to build a model,get the receiver operating characteristic curve(ROC),calculate the corresponding area under the ROC curve(AUC),and use the magnitude of the result value to evaluate the evaluation performance of each phase model.ResultsA total of 10 diagnostic features were included in the arterial phases and the venous phases.Among them,the three features with the most diagnostic performance are the first-order texture analysis 90 quantile AUC value of 0.74(0.64-0.84),the cluster protrusion AUC value of the gray-level co-occurrence matrix is 0.74(0.63-0.85)and the average absolute deviation The AUC value is 0.73(0.62-0.84).The combined serial imaging omics label has better diagnostic efficiency in the training group and the verification group,and the AUC value of the area under the curve is 0.85(0.77-0.93).Imaging omics has a good predictive value in assessing liver metastasis of pancreatic cancer,with an accuracy of84.20% in the training group and 76.80% in the verification group.ConclusionTexture analysis technology based on CT enhanced images has certain value in the prediction and evaluation of liver metastasis of pancreatic cancer.