Date on Master's Thesis/Doctoral Dissertation
Electrical and Computer Engineering
Electrical Engineering, PhD
Committee Co-Chair (if applicable)
4D-CT; lung segmentation; adaptive shape prior; radiation induced lung injury; computer aided diagnosis
Medical imaging is an important emerging technology that has been intensively used in the last few decades for disease diagnosis and monitoring as well as for the assessment of treatment effectiveness. Medical images provide a very large amount of valuable information that is too huge to be exploited by radiologists and physicians. Therefore, the design of computer-aided diagnostic (CAD) system, which can be used as an assistive tool for the medical community, is of a great importance. This dissertation deals with the development of a complete CAD system for lung cancer patients, which remains the leading cause of cancer-related death in the USA. In 2014, there were approximately 224,210 new cases of lung cancer and 159,260 related deaths. The process begins with the detection of lung cancer which is detected through the diagnosis of lung nodules (a manifestation of lung cancer). These nodules are approximately spherical regions of primarily high density tissue that are visible in computed tomography (CT) images of the lung. The treatment of these lung cancer nodules is complex, nearly 70% of lung cancer patients require radiation therapy as part of their treatment. Radiation-induced lung injury is a limiting toxicity that may decrease cure rates and increase morbidity and mortality treatment. By finding ways to accurately detect, at early stage, and hence prevent lung injury, it will have significant positive consequences for lung cancer patients. The ultimate goal of this dissertation is to develop a clinically usable CAD system that can improve the sensitivity and specificity of early detection of radiation-induced lung injury based on the hypotheses that radiated lung tissues may get affected and suffer decrease of their functionality as a side effect of radiation therapy treatment. These hypotheses have been validated by demonstrating that automatic segmentation of the lung regions and registration of consecutive respiratory phases to estimate their elasticity, ventilation, and texture features to provide discriminatory descriptors that can be used for early detection of radiation-induced lung injury. The proposed methodologies will lead to novel indexes for distinguishing normal/healthy and injured lung tissues in clinical decision-making. To achieve this goal, a CAD system for accurate detection of radiation-induced lung injury that requires three basic components has been developed. These components are the lung fields segmentation, lung registration, and features extraction and tissue classification. This dissertation starts with an exploration of the available medical imaging modalities to present the importance of medical imaging in today’s clinical applications. Secondly, the methodologies, challenges, and limitations of recent CAD systems for lung cancer detection are covered. This is followed by introducing an accurate segmentation methodology of the lung parenchyma with the focus of pathological lungs to extract the volume of interest (VOI) to be analyzed for potential existence of lung injuries stemmed from the radiation therapy. After the segmentation of the VOI, a lung registration framework is introduced to perform a crucial and important step that ensures the co-alignment of the intra-patient scans. This step eliminates the effects of orientation differences, motion, breathing, heart beats, and differences in scanning parameters to be able to accurately extract the functionality features for the lung fields. The developed registration framework also helps in the evaluation and gated control of the radiotherapy through the motion estimation analysis before and after the therapy dose. Finally, the radiation-induced lung injury is introduced, which combines the previous two medical image processing and analysis steps with the features estimation and classification step. This framework estimates and combines both texture and functional features. The texture features are modeled using the novel 7th-order Markov Gibbs random field (MGRF) model that has the ability to accurately models the texture of healthy and injured lung tissues through simultaneously accounting for both vertical and horizontal relative dependencies between voxel-wise signals. While the functionality features calculations are based on the calculated deformation fields, obtained from the 4D-CT lung registration, that maps lung voxels between successive CT scans in the respiratory cycle. These functionality features describe the ventilation, the air flow rate, of the lung tissues using the Jacobian of the deformation field and the tissues’ elasticity using the strain components calculated from the gradient of the deformation field. Finally, these features are combined in the classification model to detect the injured parts of the lung at an early stage and enables an earlier intervention.
Soliman, Ahmed Soliman Naeem, "Computational methods for the analysis of functional 4D-CT chest images." (2016). Electronic Theses and Dissertations. Paper 2567.