Image quality evaluation in a modern PET system: impact of new reconstructions methods and a radiomics approach
Keywords: 
Materias Investigacion::Ciencias de la Salud::Radiología
Issue Date: 
2019
Publisher: 
Springer Science and Business Media LLC
ISSN: 
2045-2322
Note: 
This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.
Citation: 
Reynés-Llompart, G. (Gabriel); Sabaté-Llobera, A. (Aida); Llinares-Tello, E. (Elena); et al. "Image quality evaluation in a modern PET system: impact of new reconstructions methods and a radiomics approach". Scientific Reports. 9 (10640), 2019, 1 - 9
Abstract
The present work investigates the influence of different biological and physical parameters on image quality (IQ) perception of the abdominal area in a modern PET scanner, using new reconstruction algorithms and testing the utility of a radiomics approach. Scans of 112 patients were retrospectively included. Images were reconstructed using both OSEM + PSF and BSRM methods, and IQ of the abdominal region was subjectively evaluated. First, 22 IQ related parameters were obtained (including count rate and biological or mixed parameters) and compared to the subjective IQ scores by means of correlations and logistic regression. Second, an additional set of radiomics features was extracted, and a model was constructed by means of an elastic-net regression. For the OSEM + PSF and especially for the BSRM reconstructions, IQ parameters presented only at best moderated correlations with the subjective IQ. None of the studied parameters presented a good predictive power for IQ, while a simple radiomics model increased the performance of the IQ prediction. These results suggest the necessity of changing the standard parameters to evaluate IQ, particularly when a BSRM algorithm is involved. Furthermore, it seems that a simple radiomics model can outperform the use of any single parameter to assess IQ.

Files in This Item:
Thumbnail
File
s41598-019-46937-8.pdf
Description
Size
1.42 MB
Format
Adobe PDF


Items in Dadun are protected by copyright, with all rights reserved, unless otherwise indicated.