EUVIMEDEUVIMED entdecken
Open Access

Ultra-High-Resolution Photon-Counting-Detector CT with a Dedicated Denoising Convolutional Neural Network for Enhanced Temporal Bone Imaging

Shaojie Chang, John C Benson iD, John I Lane iD, Michael R Bruesewitz iD, Joseph R Swicklik, Jamison E Thorne iD, Emily K Koons, Matthew L Carlson iD, Cynthia H McCollough iD, Shuai Leng iD

American Journal of Neuroradiology · 2025

DOI: 10.3174/ajnr.a8572 ↗

Kurz gefasst

Fachartikel in der Zeitschrift American Journal of Neuroradiology, 2025. Verfasst von Shaojie Chang und 9 weiteren Autor:innen. Beteiligte Einrichtung: From the Depsartment of Radiology (S.C., J.C.B., J.I.L., M.R.B., J.R.S., J.E.T., E.K.K., M.L.C., C.H.M., S.L.), Mayo Clinic, Rochester, Minnesota. Hauptthemen laut MEDLINE: Neural Networks, Computer, Temporal Bone, Tomography, X-Ray Computed. Die Arbeit wurde bisher 9-mal zitiert (Crossref) und nennt 21 Literaturangaben. Der Volltext ist frei in PubMed Central verfügbar. Gefördert von: NIBIB NIH HHS, National Institute of Biomedical Imaging and Bioengineering.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

BACKGROUND AND PURPOSE: Ultra-high-resolution (UHR) photon-counting-detector (PCD) CT improves image resolution but increases noise, necessitating the use of smoother reconstruction kernels that reduce resolution below the 0.125-mm maximum spatial resolution. A denoising convolutional neural network (CNN) was developed to reduce noise in images reconstructed with the available sharpest reconstruction kernel while preserving resolution for enhanced temporal bone visualization to address this issue. MATERIALS AND METHODS: With institutional review board approval, the CNN was trained on 6 patient cases of clinical temporal bone imaging (1885 images) and tested on 20 independent cases using a dual-source PCD-CT (NAEOTOM Alpha). Images were reconstructed using quantum iterative reconstruction at strength 3 (QIR3) with both a clinical routine kernel (Hr84) and the sharpest available head kernel (Hr96). The CNN was applied to images reconstructed with Hr96 and QIR1 kernel. For each case, three series of images (Hr84-QIR3, Hr96-QIR3, and Hr96-CNN) were randomized for review by 2 neuroradiologists assessing the overall quality and delineating the modiolus, stapes footplate, and incudomallear joint. RESULTS: The CNN reduced noise by 80% compared with Hr96-QIR3 and by 50% relative to Hr84-QIR3, while maintaining high resolution. Compared with the conventional method at the same kernel (Hr96-QIR3), Hr96-CNN significantly decreased image noise (from 204.63 to 47.35 HU) and improved its structural similarity index (from 0.72 to 0.99). Hr96-CNN images ranked higher than Hr84-QIR3 and Hr96-QIR3 in overall quality (P < .001). Readers preferred Hr96-CNN for all 3 structures. CONCLUSIONS: The proposed CNN significantly reduced image noise in UHR PCD-CT, enabling the use of the sharpest kernel. This combination greatly enhanced diagnostic image quality and anatomic visualization.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Neural Networks, ComputerTemporal BoneTomography, X-Ray ComputedHumansConvolutional Neural NetworksPhotonsRadiographic Image Interpretation, Computer-AssistedSignal-To-Noise Ratio

Autor:innen und Institutionen

  • From the Depsartment of Radiology (S.C., J.C.B., J.I.L., M.R.B., J.R.S., J.E.T., E.K.K., M.L.C., C.H.M., S.L.), Mayo Clinic, Rochester, Minnesota.
  • Department of Otolaryngology-Head and Neck Surgery (M.L.C.), Mayo Clinic, Rochester, Minnesota.
  • From the Depsartment of Radiology (S.C., J.C.B., J.I.L., M.R.B., J.R.S., J.E.T., E.K.K., M.L.C., C.H.M., S.L.), Mayo Clinic, Rochester, Minnesota Leng.Shuai@mayo.edu.

ORCID-Profile: John C Benson, John I Lane, Michael R Bruesewitz, Jamison E Thorne, Matthew L Carlson, Cynthia H McCollough, Shuai Leng

Förderung

  • NIBIB NIH HHS
  • National Institute of Biomedical Imaging and Bioengineering

Zitieren

APA

Chang, S., Benson, J. C., Lane, J. I., Bruesewitz, M. R., Swicklik, J. R., Thorne, J. E., Koons, E. K., Carlson, M. L., McCollough, C. H., & Leng, S. (2025). Ultra-High-Resolution Photon-Counting-Detector CT with a Dedicated Denoising Convolutional Neural Network for Enhanced Temporal Bone Imaging. American Journal of Neuroradiology. https://doi.org/10.3174/ajnr.a8572

Vancouver

Chang S, Benson JC, Lane JI, Bruesewitz MR, Swicklik JR, Thorne JE, et al. Ultra-High-Resolution Photon-Counting-Detector CT with a Dedicated Denoising Convolutional Neural Network for Enhanced Temporal Bone Imaging. American Journal of Neuroradiology. 2025. doi:10.3174/ajnr.a8572

BibTeX
@article{Chang2025,
  author = {Chang, Shaojie and Benson, John C and Lane, John I and Bruesewitz, Michael R and Swicklik, Joseph R and Thorne, Jamison E and Koons, Emily K and Carlson, Matthew L and McCollough, Cynthia H and Leng, Shuai},
  title = {Ultra-High-Resolution Photon-Counting-Detector CT with a Dedicated Denoising Convolutional Neural Network for Enhanced Temporal Bone Imaging},
  journal = {American Journal of Neuroradiology},
  year = {2025},
  doi = {10.3174/ajnr.a8572}
}
Kostenlos starten →