摘要
In this study, a cascade algorithm combined Kohonen feature map with FCM was applied to segment the MR brain images. The method we proposed was proved has better performance than FCM method usually used in image segmentation researches. Because the algorithm is unsupervised, it can reduce the errors caused by intraobserver and interobserver estimation. In this paper, we also utilized the MR images acquired by PAIR protocol to verify the result of the image segmentation.
原文 | ???core.languages.en_GB??? |
---|---|
頁(從 - 到) | 1079-1080 |
頁數 | 2 |
期刊 | Annual International Conference of the IEEE Engineering in Medicine and Biology - Proceedings |
卷 | 3 |
出版狀態 | 已出版 - 1996 |
事件 | Proceedings of the 1996 18th Annual International Conference of the IEEE Engineering in Medicine and Biology Society. Part 4 (of 5) - Amsterdam, Neth 持續時間: 31 10月 1996 → 3 11月 1996 |