Meeting Banner
Abstract #5512

Support Vector Machine based Differentiation between Vasogenic Edema and Non-enhancing Tumor in High-Grade Glioma Patients using Pre and Post Surgery MRI Images

Anirban Sengupta1, Anup Singh1, Sumeet Agarwal2, Pradeep Kumar Gupta3, and Rakesh Kumar Gupta3

1Centre for Biomedical Engineering, IIT Delhi, New Delhi, India, 2Electrical Engineering, IIT Delhi, New Delhi, India, 3Radiology, Fortis Memorial Research Institute, New Delhi, India

Differentiation of non-enhancing tumor from surrounding vasogenic edema is critical for planning tumor surgery as well as radiation therapy. Most studies suggested that histology results should be taken as ground truth instead of radiologist’s decision for validating results. This study is an attempt to differentiate vasogenic-edema from non-enhancing tumor based upon pre and post-surgery MRI images using a SVM classifier. DCE-MRI obtained perfusion parameters were used for classification. A misclassification error of 2.4 % was obtained for differentiating between non-enhancing tumor and edema using a SVM classifier followed by smoothing in post-processing step.

This abstract and the presentation materials are available to members only; a login is required.

Join Here