Ultrasound probe localization by tracking skin features

Ultrasound probe localization with respect to the human body is essential for freehand three-dimensional ultrasound (3D US), image-guided surgery, and longitudinal studies. Existing methods for probe localization, however, typically involve bulky and expensive equipment, and suffer from patient motion artifacts. This thesis presents a highly cost-effective and miniature-mobile system for ultrasound probe localization in six degrees of freedom that is robust to rigid patient motion. In this system, along with each acquisition of an ultrasound image, skin features in the scan region are recorded by a lightweight camera rigidly mounted to the probe. Through visual simultaneous localization and mapping (visual SLAM), a skin map is built based on skin features and the probe poses are estimated. Each pose estimate is refined in a Bayesian probabilistic framework that incorporates visual SLAM, ultrasound images, and a prior motion model. Extraction of human skin features and their distinctiveness in the context of probe relocalization were extensively evaluated. The system performance for free-hand 3D US was validated on three body parts: lower leg, abdomen, and neck. The motion errors were quantified, and the volume reconstructions were validated through comparison with ultrasound images. The reconstructed tissue structures were shown to be consistent with observations in ultrasound imaging, which suggests the system's potential in improving clinical workflows. In conjunction with this localization system, an intuitive interface was developed to provide real-time visual guidance for ultrasound probe realignment, which allows repeatable image acquisition in localized therapies and longitudinal studies. Through in-vivo experiments, it was shown that this system significantly improves spatial consistency of tissue structures in repeated ultrasound scans.

Department dc:contributor.department

Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science.

Grantor dc:publisher

Massachusetts Institute of Technology

Year dc:date.issued

2014

Author dc:creator

• Sun, Shih-Yu

Advisor dc:contributor.advisor

• Brian W. Anthony and Charles G. Sodini.

• Electrical Engineering and Computer Science.

Statement dc:rights

• M.I.T. theses are protected by copyright. They may be viewed from this source for any purpose, but reproduction or distribution in any format is prohibited without written permission. See provided URL for inquiries about permission.

Licence dc:rights.uri

• http://dspace.mit.edu/handle/1721.1/7582

Language dc:language.iso

eng

Handle dc:identifier.uri

http://hdl.handle.net/1721.1/93069

OAI identifier oai:identifier

oai:dspace.mit.edu:1721.1/93069