Yafei Wang1, Yue Zhang1, Lingyi Xu1, Yu Sun1, Lei Xiang2, Meiping Ye2, Suiren Wan1, Bing Zhang2, and Bin Zhu2
1The Laboratory for Medical Electronics, School of Biological Sciences and Medical Engineering, Southeast University, Nanjing, China, People's Republic of, 2Department of Radiology, The Affiliated Drum Tower Hospital of Nanjing University Medical School, Nanjing, China, People's Republic of
Classification
or grading of brain tumor alone would not be enough for clinical use,
therefore we designed a comprehensive multi-layer system combining the two
functions together. Firstly, we designed it as a three-layer system according
to clinic workflow. Then, we extracted new features from multi-modality MRI and
patients’ clinical information, which were easily ignored or difficult found by
eyes. And then we implemented SVM and Tumor Model to classify tumor type and
tumor grade. This study proposed a novel multi-layer system for clinic use by
reducing the diagnosis uncertainty.
