Please use this identifier to cite or link to this item:
|Title:||Segmentation of coronary artery to quantify stenosis from computed tomography angiography data||Authors:||Yu, Zhen||Keywords:||DRNTU::Engineering||Issue Date:||2016||Abstract:||Computed Tomography Angiography (CTA) is a popular diagnostic alternative to Invasive Coronary Angiography (ICA). The CTA medical exam creates the images of both the tissues and blood vessels in the targeted part of the patient's body, which is produced by the combination of an injected contrast medium. With the advancing of Multi-detector Computed Tomography (CT) scanner during the past 10 years, cardiac CT Angiography (cCTA) has been extensively used as a non-invasive diagnostic alternative on CHD. By applying the Quantitative Coronary Angiography (QCA), the severity of CAS is quantified by Diameter Stenosis, which is actually a projection-based 1-D surrogate for stenosis. But, it is rather a kind of estimation, as different rating results may be obtained on the same lesion with different projection angles and different interpreters. In contrast, the Area Stenosis (AS, 1 minus the ratio of actual lumen area to the stenosis-free lumen area) will not vary. AS, together with the length of lesion which produce the Volume Stenosis (VS), would provide more accurate, reproducible and informative indicators of the severity of stenosis for clinical decision making. So, we visualized coronary arteries in cCTA and quantify the severity of CAS in terms of Area Stenosis and Volume Stenosis in an objective manner from the different cCTA data sets in this project. Meanwhile, we also studied and explored how to apply an interactive software module that is designed on a core function called “virtual contrast injection” algorithm, which was introduced to separate arteries and veins from each other, but is now extended for more general CTA/ MRA image processing purposes, in particular for coronary artery CTA segmentation and centerline extraction. Besides applying the interactive software module, we also combined the Gaussian Mixture Model for threshold setting as well as manual editing through ITK-SNAP tool to further refine the coronary artery lumen extraction.||URI:||http://hdl.handle.net/10356/68092||Rights:||Nanyang Technological University||Fulltext Permission:||restricted||Fulltext Availability:||With Fulltext|
|Appears in Collections:||EEE Student Reports (FYP/IA/PA/PI)|
Items in DR-NTU are protected by copyright, with all rights reserved, unless otherwise indicated.