Wang, Jihang
(2017)
Navigating in Patient Space Using Camera Pose Estimation Relative to the External Anatomy.
Doctoral Dissertation, University of Pittsburgh.
(Unpublished)
Abstract
Ultrasound probe localization is essential for volumetric imaging with a 2D ultrasound probe, and for establishing a recorded anatomical context for ultrasound-guided surgery and for longitudinal studies. The existing techniques for probe localization, however, require external tracking devices, making them inconvenient for clinical use. In addition, the probe pose is typically measured with respect to a fixed coordinate system independent of the patient’s anatomy, making it difficult to correlate ultrasound studies across time.
This dissertation concerns the development and evaluation of a novel self-contained ultrasound probe tracking system, which navigates the probe in patient space using camera pose estimation relative to the anatomical context. As the probe moves in patient space, a video camera on the probe is used to automatically identify natural skin features and subdermal cues, and match them with a pre-acquiring high-resolution 3D surface map that serves as an atlas of the anatomy. We have addressed the problem of distinguishing rotation from translation by including an inertial navigation system (INS) to accurately measure rotation. Experiments on both a phantom containing an image of human skin (palm) as well as actual human upper extremity (fingers, palm, and wrist) validate the effectiveness of our approach.
We have also developed a real-time 3D interactive visualization system that superimposes the ultrasound data within the anatomical context of the exterior of the patient, to permit accurate anatomic localization of ultrasound data. The combination of the proposed tracking approach and the visualization system may have broad implications for ultrasound imaging, permitting the compilation of volumetric ultrasound data as the 2D probe is moved, as well as comparison of real-time ultrasound scans registered with previous scans from the same anatomical location. In a broader sense, tools that self-locate by viewing the patient’s exterior could have broad beneficial impact on clinical medicine.
Share
| Citation/Export: |
|
| Social Networking: |
|
Details
| Item Type: |
University of Pittsburgh ETD
|
| Status: |
Unpublished |
| Creators/Authors: |
|
| ETD Committee: |
|
| Date: |
1 February 2017 |
| Date Type: |
Publication |
| Defense Date: |
29 November 2016 |
| Approval Date: |
1 February 2017 |
| Submission Date: |
15 November 2016 |
| Access Restriction: |
No restriction; Release the ETD for access worldwide immediately. |
| Number of Pages: |
125 |
| Institution: |
University of Pittsburgh |
| Schools and Programs: |
Swanson School of Engineering > Bioengineering |
| Degree: |
PhD - Doctor of Philosophy |
| Thesis Type: |
Doctoral Dissertation |
| Refereed: |
Yes |
| Uncontrolled Keywords: |
ultrasound, tracking, anatomical coordinates, computer vision, guidance, ProbeSight. |
| Date Deposited: |
01 Feb 2017 19:05 |
| Last Modified: |
02 Feb 2017 06:15 |
| URI: |
http://d-scholarship.pitt.edu/id/eprint/30308 |
Metrics
Monthly Views for the past 3 years
Plum Analytics
Actions (login required)
 |
View Item |
![[feed]](/images/feed-icon-32x32.png){kind=icon}