Impact of non-contrast-enhanced imaging input sequences on the generation of virtual contrast-enhanced breast MRI scans using neural network
European Radiology · 25.10.2024 · 35 (5) · 2603-2616
DOI: 10.1007/s00330-024-11142-3 ↗Kurz gefasst
Fachartikel in der Zeitschrift European Radiology, online erschienen am 25.10.2024. Verfasst von Andrzej Liebert und 13 weiteren Autor:innen. Beteiligte Einrichtung: Institute of Radiology, Universitätsklinikum Erlangen, Friedrich-Alexander-Universität Erlangen-Nürnberg (FAU), Erlangen, Germany. andrzej.liebert@uk…. Hauptthemen laut MEDLINE: Breast Neoplasms, Neural Networks, Computer, Magnetic Resonance Imaging, Image Interpretation, Computer-Assisted. Die Arbeit wurde bisher 11-mal zitiert (Crossref) und nennt 35 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Bayerisches Staatsministerium für Wissenschaft und Kunst.
Abstract (Originalsprache)
Objective
To investigate how different combinations of T1-weighted (T1w), T2-weighted (T2w), and diffusion-weighted imaging (DWI) impact the performance of virtual contrast-enhanced (vCE) breast MRI. Materials and methods The IRB-approved, retrospective study included 1064 multiparametric breast MRI scans (age: 52 ± 12 years) obtained from 2017 to 2020 (single site, two 3-T MRI). Eleven independent neural networks were trained to derive vCE images from varying input combinations of T1w, T2w, and multi-b-value DWI sequences (b-value = 50–1500 s/mm 2 ). Three readers evaluated the vCE images with regard to qualitative scores of diagnostic image quality, image sharpness, satisfaction with contrast/signal-to-noise ratio, and lesion/non-mass enhancement conspicuity. Quantitative metrics (SSIM, PSNR, NRMSE, and median symmetrical accuracy) were analyzed and statistically compared between the input combinations for the full breast volume and both enhancing and non-enhancing target findings.
Results
The independent test set consisted of 187 cases. The quantitative metrics significantly improved in target findings when multi-b-value DWI sequences were included during vCE training ( p < 0.05). Non-significant effects ( p > 0.05) were observed for the quantitative metrics on the full breast volume when comparing input combinations including T1w. Using T1w and DWI acquisitions during vCE training is necessary to achieve high satisfaction with contrast/SNR and good conspicuity of the enhancing findings. The input combination of T1w, T2w, and DWI sequences with three b-values showed the best qualitative performance. Conclusion vCE breast MRI performance is significantly influenced by input sequences. Quantitative metrics and visual quality of vCE images significantly benefit when multi b-value DWI is added to morphologic T1w-/T2w sequences as input for model training. Key Points Question How do different MRI sequences impact the performance of virtual contrast-enhanced (vCE) breast MRI?
Findings
The input combination of T1-weighted, T2-weighted, and diffusion-weighted imaging sequences with three b-values showed the best qualitative performance. Clinical relevance While in the future neural networks providing virtual contrast-enhanced images might further improve accessibility to breast MRI, the significant influence of input data needs to be considered during translational research. Graphical Abstract
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Substanzen
Autor:innen und Institutionen
- Institute of Radiology, Universitätsklinikum Erlangen, Friedrich-Alexander-Universität Erlangen-Nürnberg (FAU), Erlangen, Germany. andrzej.liebert@uk-erlangen.de.
- Institute of Radiology, Universitätsklinikum Erlangen, Friedrich-Alexander-Universität Erlangen-Nürnberg (FAU), Erlangen, Germany.
- Lehrstuhl für Medizinische Informatik, Friedrich-Alexander-Universität Erlangen-Nürnberg (FAU), Erlangen, Germany.
- Institute of Pathology, Universitätsklinikum Erlangen, Erlangen, Comprehensive Cancer Center Erlangen-EMN, Friedrich-Alexander-Universität Erlangen-Nürnberg (FAU), Erlangen, Germany.
- Department of Gynecology and Obstetrics, Erlangen University Hospital, Comprehensive Cancer Center Erlangen-EMN, Friedrich-Alexander-Universität Erlangen-Nürnberg (FAU), Erlangen, Germany.
- Medizinische Fakultät, Friedrich-Alexander-Universität Erlangen-Nürnberg (FAU), Erlangen, Germany.
Förderung
- Bayerisches Staatsministerium für Wissenschaft und Kunst
Zitieren
APA
Liebert, A., Schreiter, H., Kapsner, L. A., Eberle, J., Ehring, C. M., Hadler, D., Brock, L., Erber, R., Emons, J., Laun, F. B., Uder, M., Wenkel, E., Ohlmeyer, S., & Bickelhaupt, S. (2024). Impact of non-contrast-enhanced imaging input sequences on the generation of virtual contrast-enhanced breast MRI scans using neural network. European Radiology, 35(5), 2603-2616. https://doi.org/10.1007/s00330-024-11142-3
Vancouver
Liebert A, Schreiter H, Kapsner LA, Eberle J, Ehring CM, Hadler D, et al. Impact of non-contrast-enhanced imaging input sequences on the generation of virtual contrast-enhanced breast MRI scans using neural network. European Radiology. 2024;35(5):2603-2616. doi:10.1007/s00330-024-11142-3
BibTeX
@article{Liebert2024,
author = {Liebert, Andrzej and Schreiter, Hannes and Kapsner, Lorenz A and Eberle, Jessica and Ehring, Chris M and Hadler, Dominique and Brock, Luise and Erber, Ramona and Emons, Julius and Laun, Frederik B and Uder, Michael and Wenkel, Evelyn and Ohlmeyer, Sabine and Bickelhaupt, Sebastian},
title = {Impact of non-contrast-enhanced imaging input sequences on the generation of virtual contrast-enhanced breast MRI scans using neural network},
journal = {European Radiology},
year = {2024},
volume = {35},
number = {5},
pages = {2603--2616},
doi = {10.1007/s00330-024-11142-3}
}
EUVIMED