Measurement and Three Dimensional Visualization of Stenosed Vessels in Magnetic Resonance Angiography
碩士 === 大葉大學 === 工業工程與科技管理學系 === 94 === Though Magnetic Resonance Imaging contains the three dimensional information、the images are still presented as the two dimensional format. 3D visualization of Magnetic Resonance Angiography (MRA) images can improve the quality of clinical diagnosis. The stenosi...
| Main Authors: | , |
|---|---|
| Other Authors: | |
| Format: | Others |
| Language: | zh-TW |
| Published: |
2006
|
| Online Access: | http://ndltd.ncl.edu.tw/handle/24628843454614505054 |
| Summary: | 碩士 === 大葉大學 === 工業工程與科技管理學系 === 94 === Though Magnetic Resonance Imaging contains the three dimensional information、the images are still presented as the two dimensional format. 3D visualization of Magnetic Resonance Angiography (MRA) images can improve the quality of clinical diagnosis.
The stenosis of brain vessel is the main cause of the stroke in the modern society. In this thesis、the Pulse Coupled Neural Network (PCNN) is applied to segment the vascular tree in MRA images. The sensitivity analysis is conducted to search the threshold’s decline weight (αθ)、experimental results showed that αθ is 1、outperforms the others parameter in terms of segmentation quality. After segmentation、a connected component algorithm is applied to link the tree-like structured vascular system. 3D visualization is used to present the reconstructed output. The algorithm to calculate the slope and a normal along the skeleton of the vessel is applied to measure the width of the vessel、which assists users to quantify the degree of stenosis of vessel in the brain MRA images.
|
|---|