Perceptual thresholds for differences in CT noise texture

Luuk J Oostveen; Kirsten Boedeker; Daniel Shin; Craig K Abbey; Ioannis Sechopoulos

doi:10.1117/1.JMI.11.3.035501

Perceptual thresholds for differences in CT noise texture

J Med Imaging (Bellingham). 2024 May;11(3):035501. doi: 10.1117/1.JMI.11.3.035501. Epub 2024 May 9.

Authors

Luuk J Oostveen¹, Kirsten Boedeker², Daniel Shin², Craig K Abbey³, Ioannis Sechopoulos^{1

4

5}

Affiliations

¹ Radboud University Medical Center, Nijmegen, The Netherlands.
² Canon Medical Systems Corporation, Los Angeles, California, United States.
³ University of California, Santa Barbara, Santa Barbara, California, United States.
⁴ University of Twente, Enschede, The Netherlands.
⁵ Dutch Expert Centre for Screening, Nijmegen, The Netherlands.

PMID: 38737494
PMCID: PMC11086665 (available on 2025-05-09)
DOI: 10.1117/1.JMI.11.3.035501

Abstract

Purpose: The average ( $f_{av}$ ) or peak ( $f_{peak}$ ) noise power spectrum (NPS) frequency is often used as a one-parameter descriptor of the CT noise texture. Our study develops a more complete two-parameter model of the CT NPS and investigates the sensitivity of human observers to changes in it.

Approach: A model of CT NPS was created based on its $f_{peak}$ and a half-Gaussian fit ( $σ$ ) to the downslope. Two-alternative forced-choice staircase studies were used to determine perceptual thresholds for noise texture, defined as parameter differences with a predetermined level of discrimination performance (80% correct). Five imaging scientist observers performed the forced-choice studies for eight directions in the $f_{peak} / σ$ -space, for two reference NPSs (corresponding to body and lung kernels). The experiment was repeated with 32 radiologists, each evaluating a single direction in the $f_{peak} / σ$ -space. NPS differences were quantified by the noise texture contrast ( $C_{texture}$ ), the integral of the absolute NPS difference.

Results: The two-parameter NPS model was found to be a good representation of various clinical CT reconstructions. Perception thresholds for $f_{peak}$ alone are $0.2 lp / cm$ for body and $0.4 lp / cm$ for lung NPSs. For $σ$ , these values are 0.15 and $2 lp / cm$ , respectively. Thresholds change if the other parameter also changes. Different NPSs with the same $f_{peak}$ or $f_{av}$ can be discriminated. Nonradiologist observers did not need more $C_{texture}$ than radiologists.

Conclusions: $f_{peak}$ or $f_{av}$ is insufficient to describe noise texture completely. The discrimination of noise texture changes depending on its frequency content. Radiologists do not discriminate noise texture changes better than nonradiologists.

Keywords: computed tomography; noise power spectrum; noise texture; perception.