科研之友
微信
新浪微博
Facebook
分享链接
摘要
Glioblastoma is the most lethal kind of brain tumor; however, patient prognoses vary significantly. Patient age and tumor location are two factors that heavily affect prognoses. Until now, few studies have focused on the relationship between age and tumor location. In our research, the relationship between tumor location and patient age was analyzed based on magnetic resonance (MR) images. This paper presents a complete process from preprocessing to statistical analysis in order to investigate the anatomic distribution of glioblastoma based on age group classifications. An improved image segmentation method based on two developed techniques was proposed to effectively segment the 3D MR images. After MR image processing, the glioblastorna of patients from different age groups were respectively registered to the standard atlas. The brain atlas segments the brain into the frontal, temporal, parietal, occipital, insular lobes, and central regions. The normalized tumor regions of different age groups were then overlapped. In the lesion overlap maps, the color range indicated the occurrence rate of lesions, which revealed obvious differences between patients of different age groups. The experiment showed that the glioblastoma location statistically associated with patient age. Patients younger than 45 were more likely to experience glioblastoma in the right side of frontal lobe, and patients over 60 were more likely to experience glioblastoma in the temporal lobe. This finding indicates possible different glioblastoma growth patterns for patients in different age groups.
- 出版日期2017-2
- 单位复旦大学
