EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

DeepFLAIR: A neural network approach to mitigate signal and contrast loss in temporal lobes at 7 Tesla FLAIR images

Daniel Uher, Gerhard S Drenthen, Benedikt A Poser, Paul A M Hofman, Louis G Wagner, Rick H G J van Lanen, Christianne M Hoeberigs, Albert J Colon, Olaf E M G Schijns, Jacobus F A Jansen, Walter H Backes

Magnetic Resonance Imaging · 2024

DOI: 10.1016/j.mri.2024.04.013 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Magnetic Resonance Imaging, 2024. Verfasst von Daniel Uher und 10 weiteren Autor:innen. Beteiligte Einrichtung: Department of Radiology and Nuclear Medicine, Maastricht University Medical Centre, Maastricht, the Netherlands; Mental Health and Neuroscience Insti…. Hauptthemen laut MEDLINE: Temporal Lobe, Magnetic Resonance Imaging, Signal-To-Noise Ratio, Image Processing, Computer-Assisted. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 38 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Epilepsiefonds.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

BACKGROUND AND PURPOSE: Higher magnetic field strength introduces stronger magnetic field inhomogeneities in the brain, especially within temporal lobes, leading to image artifacts. Particularly, T2-weighted fluid-attenuated inversion recovery (FLAIR) images can be affected by these artifacts. Here, we aimed to improve the FLAIR image quality in temporal lobe regions through image processing of multiple contrast images via machine learning using a neural network. METHODS: Thirteen drug-resistant MR-negative epilepsy patients (age 29.2 ± 9.4y, 5 females) were scanned on a 7 T MRI scanner. Magnetization-prepared (MP2RAGE) and saturation-prepared with 2 rapid gradient echoes, multi-echo gradient echo with four echo times, and the FLAIR sequence were acquired. A voxel-wise neural network was trained on extratemporal-lobe voxels from the acquired structural scans to generate a new FLAIR-like image (i.e., deepFLAIR) with reduced temporal lobe inhomogeneities. The deepFLAIR was evaluated in temporal lobes through signal-to-noise (SNR), contrast-to-noise (CNR) ratio, the sharpness of the gray-white matter boundary and joint-histogram analysis. Saliency mapping demonstrated the importance of each input image per voxel. RESULTS: SNR and CNR in both gray and white matter were significantly increased (p < 0.05) in the deepFLAIR's temporal ROIs, compared to the FLAIR. The gray-white matter boundary sharpness was either preserved or improved in 10/13 right-sided temporal regions and was found significantly increased in the ROIs. Multiple image contrasts were influential for the deepFLAIR reconstruction with the MP2RAGE second inversion image being the most important. CONCLUSIONS: The deepFLAIR network showed promise to restore the FLAIR signal and reduce contrast attenuation in temporal lobe areas. This may yield a valuable tool, especially when artifact-free FLAIR images are not available.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Temporal LobeMagnetic Resonance ImagingSignal-To-Noise RatioImage Processing, Computer-AssistedNeural Networks, ComputerArtifactsHumansFemaleAdultMaleYoung AdultWhite Matter

Stichwörter der Autor:innen

Artifact removalFLAIRMachine learningUltra-high field MRI

Autor:innen und Institutionen

  • Department of Radiology and Nuclear Medicine, Maastricht University Medical Centre, Maastricht, the Netherlands; Mental Health and Neuroscience Institute (MHeNs), Maastricht University, Maastricht, the Netherlands; Depa…
  • Department of Radiology and Nuclear Medicine, Maastricht University Medical Centre, Maastricht, the Netherlands; Mental Health and Neuroscience Institute (MHeNs), Maastricht University, Maastricht, the Netherlands.
  • Faculty of Psychology and Neuroscience (FPN), Maastricht University, the Netherlands.
  • Department of Radiology and Nuclear Medicine, Maastricht University Medical Centre, Maastricht, the Netherlands; Academic Center for Epileptology, Kempenhaeghe and Maastricht University Medical Centre, Heeze/Maastricht,…
  • Academic Center for Epileptology, Kempenhaeghe and Maastricht University Medical Centre, Heeze/Maastricht, the Netherlands.
  • Mental Health and Neuroscience Institute (MHeNs), Maastricht University, Maastricht, the Netherlands; Department of Neurosurgery, Maastricht University Medical Center, Maastricht, the Netherlands.

Förderung

  • Epilepsiefonds

Zitieren

APA

Uher, D., Drenthen, G. S., Poser, B. A., Hofman, P. A. M., Wagner, L. G., van Lanen, R. H. G. J., Hoeberigs, C. M., Colon, A. J., Schijns, O. E. M. G., Jansen, J. F. A., & Backes, W. H. (2024). DeepFLAIR: A neural network approach to mitigate signal and contrast loss in temporal lobes at 7 Tesla FLAIR images. Magnetic Resonance Imaging. https://doi.org/10.1016/j.mri.2024.04.013

Vancouver

Uher D, Drenthen GS, Poser BA, Hofman PAM, Wagner LG, van Lanen RHGJ, et al. DeepFLAIR: A neural network approach to mitigate signal and contrast loss in temporal lobes at 7 Tesla FLAIR images. Magnetic Resonance Imaging. 2024. doi:10.1016/j.mri.2024.04.013

BibTeX
@article{Uher2024,
  author = {Uher, Daniel and Drenthen, Gerhard S and Poser, Benedikt A and Hofman, Paul A M and Wagner, Louis G and van Lanen, Rick H G J and Hoeberigs, Christianne M and Colon, Albert J and Schijns, Olaf E M G and Jansen, Jacobus F A and Backes, Walter H},
  title = {DeepFLAIR: A neural network approach to mitigate signal and contrast loss in temporal lobes at 7 Tesla FLAIR images},
  journal = {Magnetic Resonance Imaging},
  year = {2024},
  doi = {10.1016/j.mri.2024.04.013}
}
Kostenlos starten →