Next Article in Journal
Diffusion-Weighted and Dynamic Contrast-Enhanced MRI Derived Imaging Metrics for Stereotactic Body Radiotherapy of Pancreatic Ductal Adenocarcinoma: Preliminary Findings
Previous Article in Journal
4D-CT Attenuation Correction in Respiratory-Gated PET for Hypoxia Imaging: Is It Really Beneficial?
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Tomography
Volume 6
Issue 2
10.18383/j.tom.2020.00003
tomography-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles thumb_up
...
Endorse textsms
...
Comment
Tomography is published by MDPI from Volume 7 Issue 1 (2021). Previous articles were published by another publisher in Open Access under a CC-BY (or CC-BY-NC-ND) licence, and they are hosted by MDPI on mdpi.com as a courtesy and upon agreement with Grapho, LLC.
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Deep Feature Stability Analysis Using CT Images of a Physical Phantom across Scanner Manufacturers, Cartridges, Pixel Sizes, and Slice Thickness

by
Rahul Paul
1,
Mohammed Shafiq-Ul Hassan
2,
Eduardo G. Moros
3,4,
Robert J. Gillies
3,
Lawrence O. Hall
1 and
Dmitry B. Goldgof
1,*
1
Department of Computer Science and Engineering, University of South Florida, Tampa, FL, USA
2
Department of Therapeutic Radiology, Yale School of Medicine, Yale University, New Haven, CT, USA
3
Department of Cancer Physiology, H. L. Moffitt Cancer Center & Research Institute, Tampa, FL, USA
4
Department of Radiation Oncology, H. L. Moffitt Cancer Center & Research Institute, Tampa, FL, USA
*
Author to whom correspondence should be addressed.
Tomography 2020, 6(2), 250-260; https://doi.org/10.18383/j.tom.2020.00003
Submission received: 10 March 2020 / Revised: 8 April 2020 / Accepted: 7 May 2020 / Published: 1 June 2020
Versions Notes

Abstract

Image acquisition parameters for computed tomography scans such as slice thickness and field of view may vary depending on tumor size and site. Recent studies have shown that some radiomics features were dependent on voxel size (= pixel size × slice thickness), and with proper normalization, this voxel size dependency could be reduced. Deep features from a convolutional neural network (CNN) have shown great promise in characterizing cancers. However, how do these deep features vary with changes in imaging acquisition parameters? To analyze the variability of deep features, a physical radiomics phantom with 10 different material cartridges was scanned on 8 different scanners. We assessed scans from 3 different cartridges (rubber, dense cork, and normal cork). Deep features from the penultimate layer of the CNN before (pre-rectified linear unit) and after (post-rectified linear unit) applying the rectified linear unit activation function were extracted from a pre-trained CNN using transfer learning. We studied both the interscanner and intrascanner dependency of deep features and also the deep features' dependency over the 3 cartridges. We found some deep features were dependent on pixel size and that, with appropriate normalization, this dependency could be reduced. False discovery rate was applied for multiple comparisons, to mitigate potentially optimistic results. We also used stable deep features for prognostic analysis on 1 non–small cell lung cancer data set.
Keywords: phantom; convolutional neural network; transfer learning; radiomics; deep feature; NSCLC phantom; convolutional neural network; transfer learning; radiomics; deep feature; NSCLC

Share and Cite

MDPI and ACS Style

Paul, R.; Hassan, M.S.-U.; Moros, E.G.; Gillies, R.J.; Hall, L.O.; Goldgof, D.B. Deep Feature Stability Analysis Using CT Images of a Physical Phantom across Scanner Manufacturers, Cartridges, Pixel Sizes, and Slice Thickness. Tomography 2020, 6, 250-260. https://doi.org/10.18383/j.tom.2020.00003

AMA Style

Paul R, Hassan MS-U, Moros EG, Gillies RJ, Hall LO, Goldgof DB. Deep Feature Stability Analysis Using CT Images of a Physical Phantom across Scanner Manufacturers, Cartridges, Pixel Sizes, and Slice Thickness. Tomography. 2020; 6(2):250-260. https://doi.org/10.18383/j.tom.2020.00003

Chicago/Turabian Style

Paul, Rahul, Mohammed Shafiq-Ul Hassan, Eduardo G. Moros, Robert J. Gillies, Lawrence O. Hall, and Dmitry B. Goldgof. 2020. "Deep Feature Stability Analysis Using CT Images of a Physical Phantom across Scanner Manufacturers, Cartridges, Pixel Sizes, and Slice Thickness" Tomography 6, no. 2: 250-260. https://doi.org/10.18383/j.tom.2020.00003

APA Style

Paul, R., Hassan, M. S. -U., Moros, E. G., Gillies, R. J., Hall, L. O., & Goldgof, D. B. (2020). Deep Feature Stability Analysis Using CT Images of a Physical Phantom across Scanner Manufacturers, Cartridges, Pixel Sizes, and Slice Thickness. Tomography, 6(2), 250-260. https://doi.org/10.18383/j.tom.2020.00003

Article Metrics

Back to TopTop