Journal article
Noise-Based Image Harmonization Significantly Increases Repeatability and Reproducibility of Radiomics Features in PET Images: A Phantom Study
Tomography (Ann Arbor), Vol.8(2), pp.1113-1128
04/13/2022
DOI: 10.3390/tomography8020091
PMID: 35448725
Abstract
For multicenter clinical studies, characterizing the robustness of image-derived radiomics features is essential. Features calculated on PET images have been shown to be very sensitive to image noise. The purpose of this work was to investigate the efficacy of a relatively simple harmonization strategy on feature robustness and agreement. A purpose-built texture pattern phantom was scanned on 10 different PET scanners in 7 institutions with various different image acquisition and reconstruction protocols. An image harmonization technique based on equalizing a contrast-to-noise ratio was employed to generate a "harmonized" alongside a "standard" dataset for a reproducibility study. In addition, a repeatability study was performed with images from a single PET scanner of variable image noise, varying the binning time of the reconstruction. Feature agreement was measured using the intraclass correlation coefficient (ICC). In the repeatability study, 81/93 features had a lower ICC on the images with the highest image noise as compared to the images with the lowest image noise. Using the harmonized dataset significantly improved the feature agreement for five of the six investigated feature classes over the standard dataset. For three feature classes, high feature agreement corresponded with higher sensitivity to the different patterns, suggesting a way to select suitable features for predictive models.
Details
- Title: Subtitle
- Noise-Based Image Harmonization Significantly Increases Repeatability and Reproducibility of Radiomics Features in PET Images: A Phantom Study
- Creators
- Harald Keller - Techna Institute, University Health Network, Toronto, ON M5G 2C4, CanadaTina Shek - Techna Institute, University Health Network, Toronto, ON M5G 2C4, CanadaBrandon Driscoll - Techna Institute, University Health Network, Toronto, ON M5G 2C4, CanadaYiwen Xu - University of TorontoBrian Nghiem - Techna Institute, University Health Network, Toronto, ON M5G 2C4, CanadaSadek Nehmeh - Department of Radiology, Weill Cornell Medical College, New York, NY 10021, USAMilan Grkovski - Memorial Sloan Kettering Cancer CenterCharles Ross Schmidtlein - Memorial Sloan Kettering Cancer CenterMikalai Budzevich - Moffitt Cancer CenterYoganand Balagurunathan - Moffitt Cancer CenterJohn J Sunderland - University of Iowa, RadiologyReinhard R Beichel - University of Iowa, Electrical and Computer EngineeringCarlos Uribe - University of British ColumbiaTing-Yim Lee - Lawson Health Research InstituteFiona Li - Western UniversityDavid A Jaffray - Techna Institute, University Health Network, Toronto, ON M5G 2C4, CanadaIvan Yeung - Techna Institute, University Health Network, Toronto, ON M5G 2C4, Canada
- Resource Type
- Journal article
- Publication Details
- Tomography (Ann Arbor), Vol.8(2), pp.1113-1128
- Publisher
- MDPI
- DOI
- 10.3390/tomography8020091
- PMID
- 35448725
- ISSN
- 2379-1381
- eISSN
- 2379-139X
- Grant note
- OQI- 137992 / CIHR P30 CA008748 / NCI NIH HHS U01 CA140206 / NCI NIH HHS P30 CA086862 / NCI NIH HHS U01CA140206 / NIH HHS
- Language
- English
- Date published
- 04/13/2022
- Academic Unit
- Radiology; Electrical and Computer Engineering; Physics and Astronomy; Radiation Oncology
- Record Identifier
- 9984244595902771
Metrics
21 Record Views