Pre-Doctoral Research & Experiences

Role of Susceptibility Weighted Imaging (SWI) in Glioma and Lymphoma Detection

Motivation: Susceptibility-weighted-imaging(SWI) is known to improve the diagnostic accuracy of glioma grading. Existing method of intra-tumoral-susceptibility-signal-intensities(ITSS) grading is semi-quantitative, manual-count- dependent and includes hemorrhage as-well-as vasculature. Haemorrhage contributes to false ITSS-count and subsequently to misclassification of tumor-grading.

Work:

1)     Formulated a quantitative-approach that calculates the vasculature-volume within tumors by automatically filtering out the hemorrhage from ITSS using R2*-quantitative-variations and connected-component-analysis based segmentation-algorithm. This work evaluates the efficiency of proposed-ITSS-vasculature-volume (IVV) to differentiate various grades of glioma and compared it with reported semi-quantitative-ITSS-approach.

2)     Quantitative-susceptibility-mapping (QSM) helps to characterize different stages of intracranial-hemorrhages and calcifications. A comparative approach was taken to use QSM for segmentation of ITSS into hemorrhage and vasculature; compute IVV and compare the results with R2*-based-IVV; evaluate its role in glioma grading.

3)     Worked on developing a novel quantitative approach which combines SWI, R2-Star-relaxivity and DCE-MRI (rCBV & Leakage) parameters for segmenting ITSS and its further classification into biological-behavior-based sub-categories: hemorrhage, leaky (agressive) and non-leaky (passive) vessels.

4)     We explored an approach based upon texture feature extraction from segmented SWI lesion, enabled automatic classification of tumors into primary central nervous system lymphoma and grade-IV glioma cases. One of the texturefeature, Contrast, provided highest AUC along with high sensitivity and specificity. This classification might improve diagnosis and grading of tumors.