Document Details

Document Type : Thesis 
Document Title :
Iterative method for Lung Tumor Localization in 4D CT
الطريقة التكرارية لتحديد موقع ورم رئوي في جهاز التصوير المقطعي رباعي الأبعاد
 
Subject : Faculty of Engineering - Department of Nuclear Engineering 
Document Language : Arabic 
Abstract : In the last decade, there were immense advancements in the medical imaging modalities. These advancements can scan a whole volume of the lung organ in high resolution images within a short time. According to this performance, the physicians can clearly describe and photograph the complicated anatomical and pathological structures of lung. Therefore, these advancements give large opportunities for all types of lung cancer treatment available and will increase the survival rate. However, lung cancer is still one of the major causes of death in 2012 with around 19% of all the cancer patients. Several factors may affect survival rate. One of the serious effects is the breathing process, which can affect the accuracy of diagnosis and lung tumor treatment plan. Therefore, this proposal focuses on dominating this effect. In radiotherapy, to dominate this effect, additional margin is added to the clinical target volume to cover the motion of the target volume as a result of breathing, and it should be limited to protect the normal tissue. In diagnosis lung, the respiratory motion cause blurring in the diagnostic images. This proposal develops a semi automated algorithm to localize the 3D lung tumor positions across all respiratory data during respiratory motion. The algorithm can be divided into two stages. First, a lung tumor segmentation for the first phase of the 4D computed tomography (CT). Lung tumor segmentation is performed using an active contours method. Then, localize the tumor 3D position across all next phases using a 12 degrees of freedom of an affine transformation. Two data set where used in this study, a compute simulate for 4D CT using extended cardiac-torso (XCAT) phantom and 4D CT clinical data sets. The result and error calculation is presented as root mean square error (RMSE). The average error in data sets is 0.935 mm ± 0.36. Finally, evaluation and quantitative comparison for the results with a state-of-the-art registration algorithm has introduced. The results obtained from the proposed localization algorithm show a promising outcome to localize a lung tumor in 4D CT data. 
Supervisor : Dr. Mohammed S. Al-Johani 
Thesis Type : Master Thesis 
Publishing Year : 1435 AH
2014 AD 
Co-Supervisor : Dr: Majdi R. Al-Nowami 
Added Date : Wednesday, September 24, 2014 

Researchers

Researcher Name (Arabic)Researcher Name (English)Researcher TypeDr GradeEmail
محمد إسماعيل السليمانيAl-Sulimane, Mohammad EsmaelResearcherMaster 

Files

File NameTypeDescription
 37364.pdf pdf 

Back To Researches Page