Abstract
Differentiating primary central nervous system lymphoma (PCNSL) from glioblastoma is crucial for treatment planning, as these conditions require different therapeutic approaches. However, distinguishing between them using conventional MRI remains challenging due to overlapping imaging characteristics. In this work, we propose a high-throughput SIFT feature-based approach for automated differentiation of PCNSL from glioblastoma.
Our method leverages Scale-Invariant Feature Transform (SIFT) features to capture distinctive imaging patterns that may be subtle to human observers. The framework consists of three key components: (1) a high-throughput SIFT feature extraction module that identifies distinctive patterns, (2) a feature selection mechanism that identifies the most discriminative features, and (3) a classification module that provides automated differentiation.
We evaluate our approach on a comprehensive dataset of PCNSL and glioblastoma cases with confirmed diagnoses. Experimental results demonstrate that our SIFT-based method significantly improves differentiation accuracy compared to traditional approaches. The method shows robust performance across different imaging protocols and patient populations.
The proposed framework represents a significant advancement in brain tumor classification, providing more accurate and reliable differentiation that could improve clinical decision-making and patient outcomes.
BibTeX
@article{Wu2018PrimaryCN,
title={Primary central nervous system lymphoma and glioblastoma image differentiation based on sparse representation system},
author={Guoqing Wu and Zeju Li and Yuanyuan Wang and Jinhua Yu and Yinsheng Chen and Zhongping Chen},
journal={Journal of Biomedical Engineering},
year={2018},
doi={10.7507/1001-5515.201705061}
}