Echocardiography plays an important role in the assessment of cardiovascular diseases. The lack of ionizing radiation and portability make it one of the safest imaging modalities. Although two-dimensional echocardiography is widely used to obtain the motion of the heart structures in real-time, three-dimensional (3D) echocardiography allows for scanning of the heart in 3D with unlimited postprocessing geometries compared to 2D. However, the feasibility is limited because of speckle noise, poor quality, limited field of view and missing anatomical structures. The entire heart cannot be imaged in a single 3D echocardiography scan in most cases, and further improvements are needed to solve the problem. This study proposes a point-based rigid registration followed by B-spline non-rigid registration to align 4D echocardiography images obtained from different sonographic windows. The approach was tested on scans obtained from three volunteer participants. The accuracy of registration was visually and quantitatively assessed by delineating the left ventricle in each scan and computing the Dice score overlap metric and the Hausdorff distance mutual proximity measure between the first scan and the rest. The overall results indicate that the proposed registration approach improves the alignment of the images compared to the original scans.

Performing surgery on fast-moving heart structures while the heart is freely beating is next to impossible. Nevertheless, the ability to do this would greatly benefit patients. By controlling a teleoperated robot to continuously follow the heart's motion, the heart can be made to appear stationary. The surgeon will then be able to operate on a seemingly stationary heart when in reality it is freely beating. The heart's motion is measured from ultrasound images and thus involves a non-negligible delay due to image acquisition and processing, estimated to be 150 ms that, if not compensated for, can cause the teleoperated robot's end-effector (i.e., the surgical tool) to collide with and puncture the heart. This research proposes the use of a Smith predictor to compensate for this time delay in calculating the reference position for the teleoperated robot. The results suggest that heart motion tracking is improved as the introduction of the Smith predictor significantly decreases the mean absolute error, which is the error in making the distance between the robot's end-effector and the heart follow the surgeon's motion, and the mean integrated square error.

Surgery on a freely beating-heart is extremely difficult as the surgeon must perform the procedure while following the heart's fast motion. However, by controlling a teleoperated robot to continuously follow the heart's motion, the surgeon can operate on a seemingly stationary heart. The heart's motion is calculated from ultrasound images and thus involves a non-negligible delay estimated to be 100 ms that, if not compensated for, can cause the robot end-effector (i.e., the surgical tool) to collide with and puncture the heart. This research proposes the use of a Smith predictor to compensate for this time delay. The results suggest that this improves heart motion tracking as the mean absolute error, the difference between the surgeon's motion and the distance between the heart and surgical tool, and the mean integrated square error decreased.

Real-time 3-D echocardiography opens up the possibility of interactive, fast 3-D analysis of cardiac anatomy and function. However, at the present time its quantitative power cannot be fully exploited due to the limited quality of the images. In this paper, we present an algorithm to register apical and parasternal echocardiographic datasets that uses a new similarity measure, based on local orientation and phase differences. By using phase and orientation to guide registration, the effect of artifacts intrinsic to ultrasound images is minimized. The presented method is fully automatic except for initialization. The accuracy of the method was validated qualitatively, resulting in 85% of the cardiac segments estimated having a registration error smaller than 2 mm, and no segments with an error larger than 5 mm. Robustness with respect to landmark initialization was validated quantitatively, with average errors smaller than 0.2 mm and 0.5 degrees for initialization landmarks rotations of up to 15 degrees and translations of up to 10 mm.