Robust bone surface and acoustic shadow segmentation from ultrasound for computer assisted orthopedic surgery
Citation & Export
Hide
Simple citation
Alsinan, Ahmed.
Robust bone surface and acoustic shadow segmentation from ultrasound for computer assisted orthopedic surgery. Retrieved from
https://doi.org/doi:10.7282/t3-69d5-ne37
Export
Description
TitleRobust bone surface and acoustic shadow segmentation from ultrasound for computer assisted orthopedic surgery
Date Created2021
Other Date2021-01 (degree)
Extent1 online resource (xiii, 103 pages)
DescriptionOrthopedic surgeries have been a prominent procedure in treating interminable pain and disabilities, due to musculoskeletal diseases, e.g. osteoarthritis, spinal conditions, osteoporosis, and low-energy fractures. Imaging has been an integral component of surgical and non-surgical orthopedic procedures such as total knee replacement (TKR), intramedullary nail locking for femoral shaft fractures, and pedicle screw insertion for spinal fusion surgery. Current practice during these procedures relies on intra-procedure 2D fluoroscopy as the main imaging modality for localization and visualization of bones, fractures, implants, and surgical tool positions. However, with such projection imaging, surgeons and clinicians typically face considerable difficulties in accurately localizing bone fragments in 3D space and assessing the adequacy and accuracy of the procedure. This problem has been overcome with 3D fluoroscopy units, however, they are twice as expensive and not widely available as standard 2D units. Additionally, fluoroscopy involves significant ionizing radiation exposure, which should be kept at minimal in order to avoid potential long-term complications. In order to overcome some of these limitations and provide a safe alternative, 2D/3D ultrasound (US) has emerged as a safe alternative while remaining relatively cheap and widely available. US image data, however, is typically characterized by high levels of speckle noise, reverberation, anisotropy and signal dropout which introduce significant difficulties during interpretation of captured data. Limited field-of-view and being a user dependent imaging modality cause additional difficulties during data collection since a single-degree deviation angle by the operator can reduce the signal strength by $50\%$. In order to overcome these difficulties automatic bone segmentation and registration methods have been developed.
The goal of this research is to develop robust, accurate, real-time and automatic image segmentation and localization methods for bone structures in US guided interventional orthopedic procedures. A multimodal convolutional neural network(CNN)-based technique is developed for segmenting bone surfaces from in vivo US scans, in which fusion of feature maps and multimodal images are incorporated to abate sensitivity to variations that are caused by imaging artifacts and low intensity bone boundaries. A block-based CNN for segmentation of bone surfaces from in vivo US scans is also proposed. We utilize fusion of feature maps and employ multi-modal images to abate sensitivity to variations caused by imaging artifacts and low intensity bone boundaries. We also propose a conditional Generative Adversarial Network (cGAN)-based method for accurate real-time segmentation of bone shadow regions from in vivo US scans. Finally, a novel GAN architecture designed to perform accurate, robust and real-time segmentation of bone shadow images from in vivo US data, is proposed. We show how the segmented bone shadow regions can be used as an additional proxy to improve bone surface segmentation results of a multi-feature guided CNN architecture. Extensive validation studies were performed to address the engineering challenges found in real clinical situations. Validating the proposed methods with clinical studies will in turn help in the future design, development, and evaluation of a 3D US based CAOS system which could improve performance by providing better assessment and placement and results in reduction of operations time. This can decrease the cost and improve efficiency by replacing fluoroscopy at key points in the diagnosis and treatment.
NotePh.D.
NoteIncludes bibliographical references
Genretheses, ETD doctoral
LanguageEnglish
CollectionSchool of Graduate Studies Electronic Theses and Dissertations
Organization NameRutgers, The State University of New Jersey
RightsThe author owns the copyright to this work.