半自動分析磁振造影頸動脈斑塊成份

Semi-automatic analysis of carotid plaque composition in magnetic resonance imaging

陳愛琪(Ai-Chi Chen)

指導教授 ： 徐良育

中原大學/工學院/生物醫學工程研究所/碩士(2015年)

https://doi.org/10.6840/cycu201500960

摘要

頸動脈粥狀硬化導致斑塊破裂和血栓形成是中風主要的發病機制與危險因素，而斑塊是否破裂取決於斑塊的成份。由於斑塊從形成到破裂包含許多過程，例如:炎症，免疫，代謝，凝血等。因此如果只是單純呈現動脈管腔的狹窄程度而無法得知斑塊穩定性，已不能滿足臨床的需要。所以需要對斑塊的形態和功能進行綜合評估。
由於粥狀硬化導致的斑塊成份、特性不同，因而MRI檢查可利用不同權種特性產生不同對比的影像強度。本篇論文提出使用MRI檢查的T1W、T2W、3D TOF與T1W+C 四種權重影像評估斑塊的種類。由於MRI檢查為多序列檢查因此影像張數甚多，使用人工閱片方式既費時又容易因人為因素導致有所誤差。因此我們將四種權重影像套入所發展的程式分析軟體，針對感興趣區域半自動化圈選管腔、管壁面積並區分各種斑塊種類，既省時又能以客觀方式評估判斷斑塊種類與成份。研究結果顯示，使用本分析軟體可以節省34%的圈選時間。同時對出血性、鈣化、脂質與纖維化斑塊的分辨準確率可以達到83.3%、100%，100%和100%。結果也證明利用本研究的成果可以成功的診斷斑塊的成份，提供成為臨床醫師在影像判讀與疾病治療時的重要指標。

關鍵字

粥狀硬化 ； 斑塊 ； 半自動圈選

並列摘要

Atherosclerotic that lead to plaque rupture and thrombosis is the main mechanism and major risk factor of stroke. However, whether plaque will rupture or not depends on the composition of the plaque. There are multiple pathologic stages from plaque formation to rupture including inflammation, autoimmune response, metabolism and thrombosis. Thus, simply showing the extent of carotid artery stenosis without the knowledge of plaque stability can no longer satisfy the clinical needs. Hence it is necessary to evaluate the morphology together with the functionality in order to understand the possibility of plaque rupture.
Since the plaque caused by atherosclerosis results in difference in plaque characteristics and components using difference weighted image techniques in MRI study can generate different signal intensity . In this study, four different weighted images, T1 weighted, T2 weighted, 3 dimensional time of flight (3D TOF) and contrast-enhanced study were used to evaluate the plaque types. In one single MRI study, there are multiple sequences and each sequence has multiple slices. It is time consuming and error-prone to have a radiologist to interpret the MRI images manually. Therefore, we propose a new software program to evaluate these four sequences. It segments the region of interest in the carotid artery, semi-automatically. In the same time, it assesses the carotid artery diameter and plaque area to evaluate plaque type and composition. The results indicate that using this software can reduce 34% of processing time. At the same time, the classification accuracies for hemorrhage, calcification, lipid and fibrous plaques are 83.3%, 100%, 100% and 100%, respectively. It demonstrates that using this new method, it can assist the radiologist and clinician in image interpretation and clinical decision-making in managing carotid artery atherosclerosis.

並列關鍵字

Atherosclerotic ； plaque ； semi-automatic segmentation

參考文獻

[4] 游. 張. 王雲銘, "磁振造影對比劑之發展簡介與研究趨勢," ed, 2009, pp. 165-177.

[1] C. R. Conti, "Updated pathophysiologic concepts in unstable coronary artery disease," Am Heart J, vol. 141, pp. S12-4, Feb 2001.

[2] J. S. Forrester, "Prevention of plaque rupture: a new paradigm of therapy," Ann Intern Med, vol. 137, pp. 823-33, Nov 19 2002.

[3] N. Takaya, C. Yuan, B. Chu, T. Saam, H. Underhill, J. Cai, et al., "Association between carotid plaque characteristics and subsequent ischemic cerebrovascular events: a prospective assessment with MRI--initial results," Stroke, vol. 37, pp. 818-23, Mar 2006.

[5] M. Cinat, C. T. Lane, H. Pham, A. Lee, S. E. Wilson, and I. Gordon, "Helical CT angiography in the preoperative evaluation of carotid artery stenosis," Journal of vascular surgery : official publication, the Society for Vascular Surgery [and] International Society for Cardiovascular Surgery, North American Chapter, vol. 28, pp. 290-300, 1998.