- 35 views
- 52 downloads
Robotic Arm Assisted Multiple Apical-View 3D Fusion of Echocardiography for Enhanced Right and Left Ventricular Assessment and Measurement
-
- Author / Creator
- Alquwaynim, Khalid Waleed
-
Introduction: The RV's complex anatomy presents unique challenges for echocardiographic evaluation, often resulting in suboptimal image quality and diagnostic accuracy. LV has a regular, symmetrical shape compared to the RV, which simplifies the quantifications of its volumes. Traditional three-dimensional echocardiography (3DE) encounters challenges such as limited field of view (FOV) and poor signal-to-noise ratio (SNR), which restrict comprehensive assessment. To overcome this, multi-view 3-D echocardiography (M3DFE) has emerged as a solution where multiple 3DE datasets from various acoustic windows are fused, with precise spatial and temporal alignment. Robotic technology has been extensively used in different healthcare fields lately. However, researchers have recently suggested integrating a robot to assist in scanning and tracking positions in 3D echocardiography on humans. This study explores the integration of a robotic arm to enhance 3DE by providing accurate 3D transducer tracking, aiming to fuse different apical views more effectively and improve RV and LV assessment quality. The proposed approach believed to be the first of its kind, tests the hypothesis that robotic-assisted data fusion can lead to superior RV and LV function and structure evaluations, tested on both volunteers and a select patient group.
Methods: This study used datasets from the Mazankowski Alberta Heart Institute, including 20 echocardiographic scans (15 healthy volunteers and five patients), approved by the University of Alberta's Health Research Ethics Board. For volunteer scans, a UR10e robotic arm with a Philips EPIQ 7C scanner ensured precision. The protocol involved multi-beat acquisition across apical and parasternal 3D views to create detailed sub-volumes. The same process was followed for patient scans, excluding the robotic arm. Data were converted to NRRD using 3D Slicer for post-processing alignment. Image fusion was done via averaging (Python) and wavelet methods (3D Slicer). SNR, CNR, and RV/LV volume measurements were analyzed using TomTec and 3D Slicer for fused images. A user study evaluated the fusion's qualitative effectiveness.
Results: The study showed improved image quality with higher SNR and CNR using wavelet fusion and averaging techniques. Robotic arm echocardiography outperformed manual methods in SNR and CNR, highlighting its precision and repeatability. While the sample size was small and differences were not statistically significant, trends were positive. Fused data slightly improved RV and LV volume measurements, and inter-rater assessments confirmed better clarity and easier segmentation in post-fusion images, particularly with the wavelet method. These results suggest that robotic assistance and fusion methods can enhance echocardiography for better cardiac diagnostics and patient care.
Conclusion: This study utilized data from a robotic arm to track the transducers and align the 3D echocardiography images on the volunteers and compares its performance with patient scans acquired without robot arm assistance. An apical-to-apical registration was performed, followed by single-view fusion using averaging and wavelet-based methods. The resulting enhancement in the images suggests improved image quality. While there were noticeable improvements in RV and LV quantifications in the small sample size, they were not statistically significant. Further studies with larger sample sizes and utilizing machine learning techniques are recommended to provide further improvement in image quality.
-
- Subjects / Keywords
-
- Graduation date
- Fall 2024
-
- Type of Item
- Thesis
-
- Degree
- Master of Science
-
- License
- This thesis is made available by the University of Alberta Library with permission of the copyright owner solely for non-commercial purposes. This thesis, or any portion thereof, may not otherwise be copied or reproduced without the written consent of the copyright owner, except to the extent permitted by Canadian copyright law.