Impact of a silver filter with deep learning reconstruction on non-contrast MDCT of adrenal glands in terms of image quality, density values and radiation dose.
- 1Department of Radiology B, University Hospital of Strasbourg - New Civil Hospital, 1, place de l' hôpital BP 426, Strasbourg Cedex, 67091, France. [email protected].
- 2Department of Radiology B, University Hospital of Strasbourg - New Civil Hospital, 1, place de l' hôpital BP 426, Strasbourg Cedex, 67091, France.
- 3Medical Physics Department, University Hospital of Strasbourg - New Civil Hospital, 1, place de l' hôpital BP 426, Strasbourg Cedex, 67091, France.
Abstract
Background
To investigate whether the use of a silver beam filtration with deep learning reconstruction (DLR) impact the normal adrenal gland on unenhanced MDCT in terms of density, image quality and radiation dose.Methods
This unicentric retrospective study evaluated two unenhanced MDCT acquisitions of 123 patients performed with a conventional filtration (Group A) and with a silver filtration (Group B) on a 320-detector row CT scanner with DLR. A total of 246 series, corresponding to 492 adrenal glands, were analyzed. We recorded the HU values of the normal adrenal gland, subcutaneous fat and extra corporeal air as well as the radiation dose and assessed image quality.Results
The subjective image quality was excellent, with no significant difference between the two acquisitions with a mean score on the Likert scale of 4.6 ± 0.3 and 4.1 ± 0.6 for groups A and B, respectively (p = 0.09). Interobserver agreement for image quality was good and quite similar for both groups (group B: k= 0.86; group A: k= 0.90). The mean density of adrenal gland was 15.65 ± 5.5 HU with conventional filtration (Group A) and 13.05 ± 4.5 HU for silver filtration (Group B). For the right adrenal gland, the median values were 15.8 HU and 12.9 HU for group A and B, respectively (p=0.01). For the left adrenal gland, the median values were 15.5 HU and 13.2 HU for group A and B, respectively (p=0.03). A subgroup of 135 adrenal glands (including both right and left side) with a density value ≤ 15HU using conventional filtration (Subgroup A) showed a statistically significant lower median value with silver filtration (Subgroup B). Inside this ≤ 15 HU subgroup, 59 adrenal glands had a density value equal or lower than 10 HU of which 48 belonged to subgroup B. A reduction in HU values of 18% was observed with the silver filter in the overall cohort (492 adrenal glands), and 26% in the subgroup ≤ 15 HU. Similarly, the mean score for image noise was 7.6 ± 1.5 for conventional filter and 8.3 ± 0.9 for silver filtration, with no statistically significant difference (p = 0.08). Radiation dose was significantly reduced by 35% in examinations performed with silver filtration when compared to conventional filter.Conclusion
The spontaneous density of normal adrenal gland is lower than previously reported. Implementing silver filtration with DLR in CT acquisition achieves a significant radiation dose reduction while preserving image quality, along with a decrease in adrenal gland HU values, often below 10 HU. This must be taken into account as this is the threshold of the key reference for adenoma.Trial registration
This study is part of a routine clinical practice. This unicentric retrospective study was approved by our local institutional review board, and written informed patient consent was not required. The patients were orally informed that their second CT examination will be performed using the newest technology available on the CT machine in our department.
Related
- AI-accelerated compressed sensing and deep learning reconstruction for colon MRI: prospective validation of image quality and preliminary assessment of staging accuracy versus contrast-enhanced CT
- Post-implantation 3-T MRI in patients with an epicranial stimulation device: image quality and deep learning-based electrode reconstruction
- Deep Learning Reconstruction versus Hybrid Iterative Reconstruction for Spectral CT: A Multi-Institutional Image Quality Assessment
- Evaluation of accelerated whole-body diffusion weighted imaging with deep learning reconstruction in patients with metastatic prostate cancer: assessment of image quality and ADC estimates.
- Deep Learning Reconstruction for Ultra-High-Resolution Photon-Counting CT of the Lung: Image Quality and Texture Characterization
BibTeX
@article{roy2026impact,
title = {{Impact of a silver filter with deep learning reconstruction on non-contrast MDCT of adrenal glands in terms of image quality, density values and radiation dose.}},
author = {Catherine Roy and Quentin le Duff and Luc Mertz and Hugo Jaillet and Aissam Labani and Sébastien Moliére and Mickael Ohana},
journal = {BMC medical imaging},
year = {2026},
month = jul,
doi = {10.1186/s12880-026-02598-4},
url = {https://europepmc.org/article/MED/42894125},
note = {CC BY},
}
RIS
TY - JOUR
TI - Impact of a silver filter with deep learning reconstruction on non-contrast MDCT of adrenal glands in terms of image quality, density values and radiation dose.
AU - Catherine Roy
AU - Quentin le Duff
AU - Luc Mertz
AU - Hugo Jaillet
AU - Aissam Labani
AU - Sébastien Moliére
AU - Mickael Ohana
PY - 2026
DA - 2026/07/01
JO - BMC medical imaging
DO - 10.1186/s12880-026-02598-4
UR - https://europepmc.org/article/MED/42894125
AB - Background To investigate whether the use of a silver beam filtration with deep learning reconstruction (DLR) impact the normal adrenal gland on unenhanced MDCT in terms of density, image quality and radiation dose.Methods This unicentric retrospective study evaluated two unenhanced MDCT acquisitions of 123 patients performed with a conventional filtration (Group A) and with a silver filtration (Group B) on a 320-detector row CT scanner with DLR. A total of 246 series, corresponding to 492 adrenal glands, were analyzed. We recorded the HU values of the normal adrenal gland, subcutaneous fat and extra corporeal air as well as the radiation dose and assessed image quality.Results The subjective image quality was excellent, with no significant difference between the two acquisitions with a mean score on the Likert scale of 4.6 ± 0.3 and 4.1 ± 0.6 for groups A and B, respectively (p = 0.09). Interobserver agreement for image quality was good and quite similar for both groups (group B: k= 0.86; group A: k= 0.90). The mean density of adrenal gland was 15.65 ± 5.5 HU with conventional filtration (Group A) and 13.05 ± 4.5 HU for silver filtration (Group B). For the right adrenal gland, the median values were 15.8 HU and 12.9 HU for group A and B, respectively (p=0.01). For the left adrenal gland, the median values were 15.5 HU and 13.2 HU for group A and B, respectively (p=0.03). A subgroup of 135 adrenal glands (including both right and left side) with a density value ≤ 15HU using conventional filtration (Subgroup A) showed a statistically significant lower median value with silver filtration (Subgroup B). Inside this ≤ 15 HU subgroup, 59 adrenal glands had a density value equal or lower than 10 HU of which 48 belonged to subgroup B. A reduction in HU values of 18% was observed with the silver filter in the overall cohort (492 adrenal glands), and 26% in the subgroup ≤ 15 HU. Similarly, the mean score for image noise was 7.6 ± 1.5 for conventional filter and 8.3 ± 0.9 for silver filtration, with no statistically significant difference (p = 0.08). Radiation dose was significantly reduced by 35% in examinations performed with silver filtration when compared to conventional filter.Conclusion The spontaneous density of normal adrenal gland is lower than previously reported. Implementing silver filtration with DLR in CT acquisition achieves a significant radiation dose reduction while preserving image quality, along with a decrease in adrenal gland HU values, often below 10 HU. This must be taken into account as this is the threshold of the key reference for adenoma.Trial registration This study is part of a routine clinical practice. This unicentric retrospective study was approved by our local institutional review board, and written informed patient consent was not required. The patients were orally informed that their second CT examination will be performed using the newest technology available on the CT machine in our department.
SN - 1471-2342
ER -