Wireless video capsule endoscopy (VCE) plays an increasingly important role in assisting clinical diagnoses of gastrointestinal (GI) diseases. It provides a non-invasive way to examine the entire small intestine, where other conventional endoscopic instruments can barely reach. Existing examination systems for the VCE cannot track the location of a endoscopic capsule, which prevents the physician from identifying the exact location of the diseases. During the eight hour examination time, the video capsule continuously keeps taking images at a frame rate up to six frame per sec, so it is possible to extract the motion information from the content of the image sequence. Many attempts have been made to develop computer vision algorithms to detect the motion of the capsule based on the small changes in the consecutive video frames and then trace the location of the capsule. However, validation of those algorithms has become a challenging topic because conducting experiments on the human body is extremely difficult due to individual differences and legal issues. In this thesis, two validation approaches for motion tracking of the VCE are presented in detail respectively. One approach is to build a physical testbed with a plastic pipe and an endoscopy camera; the other is to build a virtual testbed by creating a three-dimensional virtual small intestine model and simulating the motion of the capsule. Based on the virtual testbed, a physiological factor, intestinal contraction, has been studied in terms of its influence on visual based localization algorithm and a geometric model for measuring the amount of contraction is proposed and validated via the virtual testbed. Empirical results have made contributions in support of the performance evaluation of other research on the visual based localization algorithm of VCE.

This thesis mainly focuses on proposing a 4D (three spatial dimensions plus time) tissue-volume preserving non-rigid image registration algorithm for pulmonary 4D computed tomography (4DCT) data sets to provide relevant information for radiation therapy and to estimate pulmonary ventilation. The sum of squared tissue volume difference (SSTVD) similarity cost takes into account the CT intensity changes of spatially corresponding voxels, which is caused by variations of the fraction of tissue within voxels throughout the respiratory cycle. The proposed 4D SSTVD registration scheme considers the entire dynamic 4D data set simultaneously, using both spatial and temporal information. We employed a uniform 4D cubic B-spline parametrization of the transform and a temporally extended linear elasticity regularization of deformation field to ensure temporal smoothness and thus biological plausibility of estimated deformation. A multi-resolution multi-grid registration framework was used with a limited-memory Broyden Fletcher Goldfarb Shanno (LBFGS) optimizer for rapid convergence rate, robustness against local minima and limited memory consumption. The algorithm was prototyped in Matlab and then fully implemented in C++ in Elastix package based on the Insight Segmentation and Registration Toolkit (ITK). We conducted experiments on 2D+t synthetic images to demonstrate the effectiveness of the proposed method. The 4D SSTVD algorithm was also tested on clinical pulmonary 4DCT data sets in comparison with existing 3D pairwise SSTVD algorithm and 4D sum of squared difference (SSD) algorithm. The mean landmark error and mean landmark irregularity were calculated based on manually annotated landmarks on publicly available 4DCT data sets to evaluate the accuracy and temporal smoothness of the registration results. A 4D landmarking software tool was also designed and implemented in Java as an ImageJ plug-in to help facilitate the landmark labeling process in 4DCT data sets.

A significant barrier to applying the techniques of machine learning to the domain of object boundary detection is the need to obtain a large database of correctly labeled examples. Inspired by developmental psychology, this paper proposes that boundary detection can be learned from the output of a motion tracking algorithm that separates moving objects from their static surroundings. Motion segmentation solves the database problem by providing cheap, unlimited, labeled training data. A probabilistic model of the textural and shape properties of object boundaries can be trained from this data and then used to efficiently detect boundaries in novel images via loopy belief propagation. / Singapore-MIT Alliance (SMA)

The loss of a hand or arm is a devastating life event that results in many months of healing and challenging rehabilitation. Technology has allowed the development of an electronic replacement for a lost limb but similar advancements in therapy have not occurred. The situation is made more challenging because people with amputations often do not live near specialized rehabilitation centres. As a result, delays in therapy can worsen common complications like nerve pain and joint stiffness. For children born without a limb, poor compliance with the use of their prosthesis leads to delays in therapy and may affect their development. In many parts of the world, amputation rehabilitation does not exist. Fortunately, we live in an age where advances in technology and engineering can help solve these problems. Virtual reality creates a simulated world or environment through computer animation much like what is seen in modern video games. An experienced team of rehabilitation doctors, therapists, engineers and computer scientists are required to realize a system such as this. A person with an amputation will be taught to control objects in the virtual world by wearing a modified electronic prosthesis. Using computers, it will be possible to analyze his or her movements within the virtual world and improve the wearer's skills. The goals of this system include making the system portable and internet compatible so that people living in remote areas can also receive therapy. The novel approach of using virtual reality to rehabilitate people with upper limb amputations will help them return to normal activities by providing modern and appropriate rehabilitation, reducing medical complications, improving motivation (via gaming modules), advancing health care technology and reducing health care costs. The use of virtual reality technology in the field of amputee rehabilitation is in its earliest stages of development world wide. A virtual environment (VE) will facilitate the early rehabilitation of a patient before they are clinically ready to be fitted with an actual prosthesis. In order to create a successful virtual reality rehabilitation system such as this, an accurate method of tracking the arm in real-time is necessary. A linear displacement sensor and a microelectromechanical system (MEMS) inertial measurement unit (IMU) were used to create a device for capturing the motion of a user's movement with the intent that the data provided by the device be used along with a VE as a virtual rehabilitation tool for new upper extremity amputation patients. This thesis focuses on the design and testing of this motion capture device in order to determine the suitability of current commercially available sensing components as used in this system. Success will be defined by the delivery of accurate position and orientation data from the device so that that data can be used in a virtual environment. Test results show that with current MEMS sensors, the error introduced by double integrating acceleration data is too significant to make an IMU an acceptable choice for position tracking. However, the device designed here has proven to be an excellent cable emulator, and would be well suited if used as an orientation tracker.

This thesis focuses on the analysis of the cardiac electrical and kinematic function for heart failure patients. An expected outcome is a set of computational tools that may help a clinician in understanding, diagnosing and treating patients suffering from cardiac motion asynchrony, a specific aspect of heart failure. Understanding the inverse electro-kinematic coupling relationship is the main task of this study. With this knowledge, the widely available cardiac image sequences acquired non-invasively at clinics could be used to estimate the cardiac electrophysiology (EP) without having to perform the invasive cardiac EP mapping procedures. To this end, we use real clinical cardiac sequence and a cardiac electromechanical model to create controlled synthetic sequence so as to produce a training set in an attempt to learn the cardiac electro-kinematic relationship. Creating patient-specific database of synthetic sequences allows us to study this relationship using a machine learning approach. A first contribution of this work is a non-linear registration method applied and evaluated on cardiac sequences to estimate the cardiac motion. Second, a new approach in the generation of the synthetic but virtually realistic cardiac sequence which combines a biophysical model and clinical images is developed. Finally, we present the cardiac electrophysiological activation time estimation from medical images using a patient-specific database of synthetic image sequences.

A mathematical model for a Two-Span Web Roller machine is defined in order to facilitate motion planning, motion tracking and state observer design for tracking web tension and web velocity. Differential Flatness is utilized to create reference trajectories that are tracked with a high convergence rate. Flatness also allows for nominal input torque generation without integration. Constraints on the inputs are satisfied through the motion planning phase. A partial state feedback linearization is performed and an exponential tracking dynamic feedback controller is defined. An exponential Kalman-related tension observer is also defined with semi-optimal gain formulation. The observer takes advantage of the bilinearity of the dynamics up to additive output nonlinearity. The closed-loop system is simulated in MatLab with comparisons to reference trajectories previously employed in literature. The importance of proper motion planning is demonstrated by producing excellent performance compared with existing tracking and tension observing methods.

Motion controlling is a rapidly improving area of game technologies. In the last few years, motion sensing devices for video games such as Nintendo Wii, Microsoft Kinect for Xbox 360 and Sony PlayStation Move have gained popularity among players with many compatible motion controlled games. Microsoft Kinect for Xbox 360 provides a controller free interaction system in which the player controls games by using only body movements. Although Kinect provides a natural way of interaction, rolling action of body joints are not recognized within the standard motion sensing scope of the tool. Aim of this thesis is to provide an improved gameplay system with an increased degree of freedom by introducing rolling axis of movement using Microsoft Kinect for Xbox 360 for motion sensing. This improved gameplay system provides the players a more natural and accurate way of motion controlled interaction, eliminating unnatural gestures that are needed to be memorized to compensate for lacking of the roll movement recognition.

Recent years have witnessed great improvements in ways of game controlling yielding to higher level of interaction. Release of motion controller devices radically changed the conventional ways of interaction that have been used for controlling games so far, also giving developers the opportunity of exploring various new possible ways of interaction. One of these off the shelf tools, Microsoft Kinect for Xbox 360, recognizes motions of the players as game controlling inputs. Although touchless interaction is perceived to be attractive, games that mimic real life activities such as table tennis, sword fighting, baseball and golf may benefit from the player&rsquo / s holding a tangible object to get more involved into game, sensing the actions deeply. In this thesis, a tangible gameplay interaction method that senses whether or not the player holds an object in the hand / if so, detects its dimensions and incorporates the hand-held object into gameplay by projecting motions of the player accordingly, is developed using Microsoft Kinect for Xbox 360. Developed algorithm is implemented on an experimental game and a user study is performed which revealed that an improved gameplay with more natural and accurate motion controlling yielding to new possible actions is achieved with the developed system.

Synaisthe is an intermedia dance performance that premiered in March 2011 at the University Co-op Presents the Cohen New Works Festival, at the University of Texas at Austin. The context of Synaisthe is two-fold. In one aspect, the work lives in the world of theatrical design, emerging technologies, storytelling practices, and performance theory. In its subject matter, Synaisthe is in conversation with cognitive science, neurology, human perception, and the exploration of individual experience. This event, driven by a technology centered devising process, included; dance, a live band, an infra-red motion tracking floor projection system, manipulation of media through midi-instrument control, a walkman converted into a wearable sonic-fabric costume reading device, and an audience inclusive dance party. The creation of Synaisthe came about as a result of three major aims. A design and technology centered devising process, immersive interaction, and exploration of non-traditional collaborative structures. I endeavor to create a performance as means of researching the influence of media on performance practice. / text

A mathematical model for a Two-Span Web Roller machine is defined in order to facilitate motion planning, motion tracking and state observer design for tracking web tension and web velocity. Differential Flatness is utilized to create reference trajectories that are tracked with a high convergence rate. Flatness also allows for nominal input torque generation without integration. Constraints on the inputs are satisfied through the motion planning phase. A partial state feedback linearization is performed and an exponential tracking dynamic feedback controller is defined. An exponential Kalman-related tension observer is also defined with semi-optimal gain formulation. The observer takes advantage of the bilinearity of the dynamics up to additive output nonlinearity. The closed-loop system is simulated in MatLab with comparisons to reference trajectories previously employed in literature. The importance of proper motion planning is demonstrated by producing excellent performance compared with existing tracking and tension observing methods.