Font Size: a A A

Differentiation Of Benign And Malignant Musculoskeletal Lesions Using Texture Analysis Of Conventional MRI:a Preliminary Study

Posted on:2020-08-05Degree:MasterType:Thesis
Country:ChinaCandidate:W X YuFull Text:PDF
GTID:2404330578978395Subject:Imaging and nuclear medicine
Abstract/Summary:PDF Full Text Request
PurposeTo investigate the diagnostic value of texture analysis derived from conventional MR imaging in differentiating benign and malignant musculoskeletal lesions.Materials and methodsTwenty-four patients with malignant musculoskeletal lesion and 22 patients with benign musculoskeletal lesion were retrospectively analyzed in our study.All patients underwent conventional MR imaging including T1WII STIR,and contrast-enhanced T1WI before surgery.Texture features were calculated from manually drawn ROIs by using MaZda software(version 4.6).The feature selection methods included Fishers coefficient,mutual information(MI),classification error probability combined with average correlation coefficients(POE+ACC)and the combination of the above three methods(MPF).These methods were used to identify the most significant texture features in discriminating benign musculoskeletal lesion from malignant musculoskeletal lesion.The statistical methods including raw data analysis(RDA),principal component analysis(PCA),linear discriminant analysis(LDA)and nonlinear discriminant analysis(NDA)were used to distinguish malignant musculoskeletal lesion from benign musculoskeletal lesion.The results were shown by misclassification rate.ResultsIn the three kinds of sequences,the texture features for differentiating malignant musculoskeletal lesion and benign musculoskeletal lesion were mainly from STIR which had the lowest misclassification rate 2.17%(1/46).The misclassification rate of the combination of the three methods(2.17%to21.74%for MPF)was lower than that of any other kind of method.In the statistical methods,NDA(2.17%to 13.04%)had lower misclassification rate than LDA(6.52%to 21.74%)?RDA(17.39%to 34.78%)and PCA(15.22%to 34.78%),MPF combined with NDA yields the lowest misclassification rate(2.17%).ConclusionTexture analysis of conventional MR imaging can provide reliably objective basis for differentiating benign from malignant musculoskeletal lesions.
Keywords/Search Tags:Magnetic resonance imaging, Musculoskeletal diseases, Texture analysis
PDF Full Text Request
Related items