With the development of MR techniques, the T2WI, DWI and delayed T1-enhanced imaging techniques have applied as the necessary and general clinical scanning sequences for the bone tumors. And the tumors would present a specific intensity distribution for a specific modality of these high resolution images. Therefore, this work investigated and evaluated the role of textures extracted from T2WI, DWI and delayed T1-enhanced imaging in characterizing the subtypes of bone tumors, and furtherly evaluate the ability of these textures to differentiate benign and malignant tumorsby using support vector machine classifier1 (SVM), which might be helpful for clinical diagnosis and studies.Results showed that the texture parameters extracted from these images have the ability to distinguish benign from malignant bone tumors based on SVM.
This abstract and the presentation materials are available to members only; a login is required.