| Author: | Victorova, Maria |
| Title: | Automation of ultrasound assessment of scoliosis with robotic scanning |
| Advisors: | Zheng, Yongping (BME) |
| Degree: | Ph.D. |
| Year: | 2022 |
| Subject: | Scoliosis -- Ultrasonic imaging Diagnostic ultrasonic imaging -- Automation Hong Kong Polytechnic University -- Dissertations |
| Department: | Department of Biomedical Engineering |
| Pages: | xix, 120 pages : color illustrations |
| Language: | English |
| Abstract: | Can scoliosis assessment be performed with the automatic ultrasound approach, and what are the benefits such method potentially has? Those are the questions the current thesis uncover. The Chapter 1 - "Introduction" firstly explains why ultrasound assessment is preferable to be used for scoliosis assessment; it also discusses different applications of robotic ultrasound systems and lastly states the proposed robotic-ultrasound system for scoliosis assessment. The thesis is composed of several main chapters, each focusing on different aspects of the robotic-ultrasound system. Chapters are presented in chronological order in a way that the proposed system enhances throughout the chapters. Each chapter has its own introduction and indicates state-ofÂthe-art research gaps, methods, experiments, results, and discussions. The beginning of the system design is described in Chapter 2 - "Development of Robotic Ultrasound System for Scoliosis Assessment". It presents the hybrid force/position method for robotic control and its evaluation on two phantoms with different stiffness. The primary outcome of the work presented in this chapter is a control-ready robotic-ultrasound system for phantom scanning, which can be controlled through a graphical interface by an operator. The next advancement on the system was an exploration of deep learning computer vision capabilities to process spinal ultrasound images in Chapter 3 "Force-Ultrasound Fusion: Vertebrae Level Classification". This chapter describes the fusion of force information and deep learning processed images information for the human back robotic ultrasound scanning. By fusing this information, the resulting spinal scan can be represented as a sequence of individual vertebrae levels in the lumbar spine region, from L5 to L1. The image processing approaches mentioned above boosted the widening of deep learning applications for robotic ultrasound scanning. Chapter 4 - "Spinal Landmark Localization and Tracking for Scoliosis Assessment" proposes a method of real-time spinous process localization during the scanning. It also shows the robotic control method to center the spinous process in the field of view of the ultrasound probe to follow the spinal curve during the scanning. To make scanning more complaint and less-variable dependant, the spinal region classification was used for parameters adjustments. The method was evaluated both on phantom and human. Last Chapter 5 "Comparison of Automatic and Manual Scanning" shows the results of the massive human scanning with both automatic and manual. The analysis between the two sets is presented. |
| Rights: | All rights reserved |
| Access: | open access |
Copyright Undertaking
As a bona fide Library user, I declare that:
- I will abide by the rules and legal ordinances governing copyright regarding the use of the Database.
- I will use the Database for the purpose of my research or private study only and not for circulation or further reproduction or any other purpose.
- I agree to indemnify and hold the University harmless from and against any loss, damage, cost, liability or expenses arising from copyright infringement or unauthorized usage.
By downloading any item(s) listed above, you acknowledge that you have read and understood the copyright undertaking as stated above, and agree to be bound by all of its terms.
Please use this identifier to cite or link to this item:
https://theses.lib.polyu.edu.hk/handle/200/11664