Title :
Segmentation fusion of image with meningoma
Author :
Tong, Longzheng ; Zhou, Qiming ; Li, Yongzhong ; Wang, Wei ; Ye, Derong ; Ma, Binrong
Author_Institution :
Capital Univ. of Med. Sci., China
Abstract :
Possessing both merits and demerits respectively, CT and MRI test are the two common radiological means to diagnose meningoma. With high-density resolution, CT detects even tiny density differentia of normal tissue and tumor tissue in the brain, but can not distinguish every soft tissue as expected. MRI scans all direction of the tissues freely, and can clearly differentiate soft tissues such as muscle, fascia, fat, gray matter, white matter, and abnormal lesion, but can not accurately identify the calcified part of tumor and the boundary between tumor and cranium. In recent years, researchers around world try to register and fuse images of CT and MRI in order to achieve a new image with more information for diagnosis and differentiating diagnosis. Based on CT and MRI images of meningoma, the normal parts and tumor parts were registered, segmented and processed with weighted fusion. Pseudo color was used on the image to provide more easily recognized information in new one for early clinical diagnosis of the meningoma
Keywords :
biological tissues; biomedical MRI; brain; cancer; computerised tomography; image registration; image segmentation; patient diagnosis; sensor fusion; CT images; CT test; MRI images; MRI test; abnormal lesion; brain; clinical diagnosis; differentiating diagnosis; fascia; fat; gray matter; high-density resolution; image registration; image segmentation; meningoma diagnosis; muscle; normal tissue; pseudo color; segmentation image fusion; soft tissue; tumor tissue; weighted fusion; white matter; Biological tissues; Computed tomography; Cranium; Fascia; Image segmentation; Lesions; Magnetic resonance imaging; Muscles; Neoplasms; Testing;
Conference_Titel :
Signal Processing Proceedings, 2000. WCCC-ICSP 2000. 5th International Conference on
Conference_Location :
Beijing
Print_ISBN :
0-7803-5747-7
DOI :
10.1109/ICOSP.2000.891735