Skip to main content

Lund University Publications

LUND UNIVERSITY LIBRARIES

How does intelligent noise reduction software influence the image quality in pelvic digital radiography; a phantom study

Hussner, E. D. ; Sundby, S. ; Outzen, C. B. ; Jensen, J. ; Tingberg, A. LU orcid and Precht, H. (2025) In Journal of Medical Imaging and Radiation Sciences 56(2).
Abstract

Introduction: This study aims to evaluate the effects of a novel noise reduction software (INR, Canon Europe, Amsterdam, NL) on image quality (IQ) in Digital Radiography (DR) pelvic phantom images. Methods: In total, 53 pelvic phantom images and 360 technical images of a Contrast Detail Radiography phantom (CDRAD) were collected, including 8 different exposure levels (between 0.8 and 40 mAs at 70 kV) and 6 intensities of INR (ranging from settings 0–10). The pelvic images were evaluated by three reporting radiographers using absolute Visual Grading Analysis (VGA). The CDRAD images were analyzed with a CDRAD computer software. Results: The VGA showed that the images with the INR software had higher IQ than the images with no INR. The... (More)

Introduction: This study aims to evaluate the effects of a novel noise reduction software (INR, Canon Europe, Amsterdam, NL) on image quality (IQ) in Digital Radiography (DR) pelvic phantom images. Methods: In total, 53 pelvic phantom images and 360 technical images of a Contrast Detail Radiography phantom (CDRAD) were collected, including 8 different exposure levels (between 0.8 and 40 mAs at 70 kV) and 6 intensities of INR (ranging from settings 0–10). The pelvic images were evaluated by three reporting radiographers using absolute Visual Grading Analysis (VGA). The CDRAD images were analyzed with a CDRAD computer software. Results: The VGA showed that the images with the INR software had higher IQ than the images with no INR. The observers gave a high VGA score to the images with INR of 3.2 mAs and higher. There was a tendency for an INR level of 5 or 7 to give the highest VGA scores. In addition, the CDRAD study showed a significant improvement in IQ with increasing INR levels at the lower exposure levels. Conclusion: An improvement in overall IQ was seen at lower exposure levels when the INR software was used both for the VGA and the CDRAD study. Clinical tests including patient images need to be performed before implementing INR in practice to verify accurate diagnostic performance.

(Less)
Please use this url to cite or link to this publication:
author
; ; ; ; and
organization
publishing date
type
Contribution to journal
publication status
published
subject
keywords
Artificial intelligence, Digital radiography, Image quality, Noise reduction, Pelvis, Radiation dose
in
Journal of Medical Imaging and Radiation Sciences
volume
56
issue
2
article number
101814
publisher
Elsevier
external identifiers
  • pmid:39644735
  • scopus:85211180108
ISSN
1939-8654
DOI
10.1016/j.jmir.2024.101814
language
English
LU publication?
yes
id
cef5e2f7-b6ec-45e4-95e6-ed0af4619c3a
date added to LUP
2025-02-26 15:49:51
date last changed
2025-02-27 03:00:02
@article{cef5e2f7-b6ec-45e4-95e6-ed0af4619c3a,
  abstract     = {{<p>Introduction: This study aims to evaluate the effects of a novel noise reduction software (INR, Canon Europe, Amsterdam, NL) on image quality (IQ) in Digital Radiography (DR) pelvic phantom images. Methods: In total, 53 pelvic phantom images and 360 technical images of a Contrast Detail Radiography phantom (CDRAD) were collected, including 8 different exposure levels (between 0.8 and 40 mAs at 70 kV) and 6 intensities of INR (ranging from settings 0–10). The pelvic images were evaluated by three reporting radiographers using absolute Visual Grading Analysis (VGA). The CDRAD images were analyzed with a CDRAD computer software. Results: The VGA showed that the images with the INR software had higher IQ than the images with no INR. The observers gave a high VGA score to the images with INR of 3.2 mAs and higher. There was a tendency for an INR level of 5 or 7 to give the highest VGA scores. In addition, the CDRAD study showed a significant improvement in IQ with increasing INR levels at the lower exposure levels. Conclusion: An improvement in overall IQ was seen at lower exposure levels when the INR software was used both for the VGA and the CDRAD study. Clinical tests including patient images need to be performed before implementing INR in practice to verify accurate diagnostic performance.</p>}},
  author       = {{Hussner, E. D. and Sundby, S. and Outzen, C. B. and Jensen, J. and Tingberg, A. and Precht, H.}},
  issn         = {{1939-8654}},
  keywords     = {{Artificial intelligence; Digital radiography; Image quality; Noise reduction; Pelvis; Radiation dose}},
  language     = {{eng}},
  number       = {{2}},
  publisher    = {{Elsevier}},
  series       = {{Journal of Medical Imaging and Radiation Sciences}},
  title        = {{How does intelligent noise reduction software influence the image quality in pelvic digital radiography; a phantom study}},
  url          = {{http://dx.doi.org/10.1016/j.jmir.2024.101814}},
  doi          = {{10.1016/j.jmir.2024.101814}},
  volume       = {{56}},
  year         = {{2025}},
}