On the nature of data collection for soft-tissue image-to-physical organ registration: a noise characterization study

Jarrod A. Collins; Jon S. Heiselman; Jared A. Weis; Logan W. Clements; Amber L. Simpson; William R. Jarnagin; Michael I. Miga

doi:10.1117/12.2255844

3 March 2017 On the nature of data collection for soft-tissue image-to-physical organ registration: a noise characterization study

Jarrod A. Collins, Jon S. Heiselman, Jared A. Weis, Logan W. Clements, Amber L. Simpson, William R. Jarnagin, Michael I. Miga

Author Affiliations +

Proceedings Volume 10135, Medical Imaging 2017: Image-Guided Procedures, Robotic Interventions, and Modeling; 101351Y (2017) https://doi.org/10.1117/12.2255844
Event: SPIE Medical Imaging, 2017, Orlando, Florida, United States

Abstract

In image-guided liver surgery (IGLS), sparse representations of the anterior organ surface may be collected intraoperatively to drive image-to-physical space registration. Soft tissue deformation represents a significant source of error for IGLS techniques. This work investigates the impact of surface data quality on current surface based IGLS registration methods. In this work, we characterize the robustness of our IGLS registration methods to noise in organ surface digitization. We study this within a novel human-to-phantom data framework that allows a rapid evaluation of clinically realistic data and noise patterns on a fully characterized hepatic deformation phantom. Additionally, we implement a surface data resampling strategy that is designed to decrease the impact of differences in surface acquisition. For this analysis, n=5 cases of clinical intraoperative data consisting of organ surface and salient feature digitizations from open liver resection were collected and analyzed within our human-to-phantom validation framework. As expected, results indicate that increasing levels of noise in surface acquisition cause registration fidelity to deteriorate. With respect to rigid registration using the raw and resampled data at clinically realistic levels of noise (i.e. a magnitude of 1.5 mm), resampling improved TRE by 21%. In terms of nonrigid registration, registrations using resampled data outperformed the raw data result by 14% at clinically realistic levels and were less susceptible to noise across the range of noise investigated. These results demonstrate the types of analyses our novel human-to-phantom validation framework can provide and indicate the considerable benefits of resampling strategies.

Citation Download Citation

Jarrod A. Collins, Jon S. Heiselman, Jared A. Weis, Logan W. Clements, Amber L. Simpson, William R. Jarnagin, and Michael I. Miga "On the nature of data collection for soft-tissue image-to-physical organ registration: a noise characterization study", Proc. SPIE 10135, Medical Imaging 2017: Image-Guided Procedures, Robotic Interventions, and Modeling, 101351Y (3 March 2017); https://doi.org/10.1117/12.2255844

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available

Members: $17.00

Non-members: $21.00 ADD TO CART

PROCEEDINGS
12 PAGES

DOWNLOAD PAPER SAVE TO MY LIBRARY

GET CITATION

RIGHTS & PERMISSIONS

Get copyright permission Get copyright permission on Copyright Marketplace

KEYWORDS

Rigid registration

Image registration

Data modeling

Liver

Surgery

Data acquisition

Image-guided intervention

Show All Keywords

Keywords/Phrases

Search In:

Publication Years