Radiation therapy is a cancer treatment method that employs high-energy radiation beams to destroy cancer cells by damaging the ability of these cells to reproduce. Thoracic and abdominal tumors may change their positions during respiration by as much as three centimeters during radiation treatment. The prediction of respiratory motion has become an important research area because respiratory motion severely affects precise radiation dose delivery. This study describes recent radiotherapy technologies including tools for measuring target position during radiotherapy and tracking-based delivery systems. In the first part of our study we review three prediction approaches of respiratory motion, i.e., model-based methods, model-free heuristic learning algorithms, and hybrid methods. In the second part of our work we propose respiratory motion estimation with hybrid implementation of extended Kalman filter. The proposed method uses the recurrent neural network as the role of the predictor and the extended Kalman filter as the role of the corrector. In the third part of our work we further extend our research work to present customized prediction of respiratory motion with clustering from multiple patient interactions. For the customized prediction we construct the clustering based on breathing patterns of multiple patients using the feature selection metrics that are composed of a variety of breathing features. In the fourth part of our work we retrospectively categorize breathing data into several classes and propose a new approach to detect irregular breathing patterns using neural networks. We have evaluated the proposed new algorithm by comparing the prediction overshoot and the tracking estimation value. The experimental results of 448 patients’ breathing patterns validated the proposed irregular breathing classifier.

Recent reports suggest that measuring the objective quality is very essential towards the success of colonoscopy. Several quality indicators (i.e. metrics) proposed in recent studies are implemented in software systems that compute real-time quality scores for routine screening colonoscopy. Most quality metrics are derived based on various temporal events occurred during the colonoscopy procedure. The location of the phase boundary between the insertion and the withdrawal phases and the amount of circumferential inspection are two such important temporal events. These two temporal events can be determined by analyzing various camera motions of the colonoscope. This dissertation put forward a novel method to estimate X, Y and Z directional motions of the colonoscope using motion vector templates. Since abnormalities of a WCE or a colonoscopy video can be found in a small number of frames (around 5% out of total frames), it is very helpful if a computer system can decide whether a frame has any mucosal abnormalities. Also, the number of detected abnormal lesions during a procedure is used as a quality indicator. Majority of the existing abnormal detection methods focus on detecting only one type of abnormality or the overall accuracies are somewhat low if the method tries to detect multiple abnormalities. Most abnormalities in endoscopy images have unique textures which are clearly distinguishable from normal textures. In this dissertation a new method is proposed that achieves the objective of detecting multiple abnormalities with a higher accuracy using a multi-texture analysis technique. The multi-texture analysis method is designed by representing WCE and colonoscopy image textures as textons.