EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Automated segmentation of multiple sclerosis lesions, paramagnetic rims, and central vein sign on MRI provides reliable diagnostic biomarkers

Fengling Hu iD, Zheng Ren, Luyun Chen, Alessandra M Valcarcel, Jordan Dworkin, Brian Renner, Lynn Daboul, Carly M O'Donnell, Elizabeth D Verter, Abigail R Manning, Kelly A Clark, Eunchan Bae und 20 weitere

Imaging Neuroscience · 10.10.2025 · 3

DOI: 10.1162/imag.a.932 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Imaging Neuroscience, online erschienen am 10.10.2025. Verfasst von Fengling Hu und 31 weiteren Autor:innen. Beteiligte Einrichtung: Penn Statistics in Imaging and Visualization Endeavor (PennSIVE), Department of Biostatistics, Epidemiology, and Informatics, Perelman School of Medi…. Die Arbeit wurde bisher 6-mal zitiert (Crossref) und nennt 37 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Institute of Neurological Disorders and Stroke, NIH Medical Scientist Training Program, NIH Medical Research Scholars Program.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Multiple sclerosis (MS) is characterized by central nervous system lesions detectable via MRI. Existing diagnostic criteria incorporate presence of white matter lesions, but specificity can be improved using MS-specific imaging biomarkers, including paramagnetic rim lesions (PRLs) and central vein sign (CVS). However, manual segmentation of lesions, PRLs, and CVS is time-consuming and subjective. We propose a fully-automated joint segmentation method called Automated Lesion, PRL, and CVS Analysis (ALPaCA). We trained ALPaCA using subject-level cross-validation on 47 adults with MS and 50 adults with radiological MS mimics. ALPaCA uses a voxel-wise lesion segmentation method to propose a large set of lesion candidates. Lesion candidates are input into a multi-contrast, multi-label 3D convolutional neural network as 3D patches to produce lesion, PRL, and CVS predictions. When multiple lesions exist within a patch, an attention mechanism identifies which lesion candidate to classify. At the lesion level, ALPaCA achieves cross-validation areas under the receiver operating characteristic curve (AUROCs) of 0.95, 0.91, and 0.87 for lesion, PRL, and CVS classification, outperforming previous methods (all p < 0.001). Correlations between subject-level ALPaCA lesion and PRL scores with manual counts are higher than those of previous methods (p < 0.001; p = 0.03). Subject-level ALPaCA PRL and CVS scores are highly associated with MS in logistic regressions, when controlling for age and sex (p < 0.001). ALPaCA allows for fully-automated simultaneous segmentation of MS lesions, PRLs, and CVS using clinically-feasible scans. These segmentations outperform existing methods at the lesion and subject level.

Schlagwörter

Stichwörter der Autor:innen

central vein signdeep learninglesion segmentationmultiple sclerosisparamagnetic rim lesions

Autor:innen und Institutionen

  • Penn Statistics in Imaging and Visualization Endeavor (PennSIVE), Department of Biostatistics, Epidemiology, and Informatics, Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA, United States.
  • Genentech Inc. A Member of the Roche Group, South San Francisco, CA, United States.
  • Department of Neurology, Cedars-Sinai Medical Center, Los Angeles, CA, United States.
  • Biomedical Imaging Research Institute, Cedars-Sinai Medical Center, Los Angeles, CA, United States.
  • Translational Neuroradiology Section, National Institute of Neurological Disorders and Stroke, National Institute of Health, Bethesda, MD, United States.
  • Department of Neurology, Brigham and Women's Hospital, Boston, MA, United States.

ORCID-Profile: Fengling Hu

Förderung

  • National Institute of Neurological Disorders and Stroke
  • NIH Medical Scientist Training Program
  • NIH Medical Research Scholars Program
  • Conrad N. Hilton Foundation
  • Fondazione Cariplo
  • Roche Foundation
  • National MS Society
  • International Progressive MS Alliance

Zitieren

APA

Hu, F., Ren, Z., Chen, L., Valcarcel, A. M., Dworkin, J., Renner, B., Daboul, L., O'Donnell, C. M., Verter, E. D., Manning, A. R., Clark, K. A., Bae, E., Chen, C., Lou, C., Satterthwaite, T. D., Shou, H., Bilello, M., Nakamura, K., Bar-Or, A., …, & Shinohara, R. T. (2025). Automated segmentation of multiple sclerosis lesions, paramagnetic rims, and central vein sign on MRI provides reliable diagnostic biomarkers. Imaging Neuroscience, 3. https://doi.org/10.1162/imag.a.932

Vancouver

Hu F, Ren Z, Chen L, Valcarcel AM, Dworkin J, Renner B, et al. Automated segmentation of multiple sclerosis lesions, paramagnetic rims, and central vein sign on MRI provides reliable diagnostic biomarkers. Imaging Neuroscience. 2025;3. doi:10.1162/imag.a.932

BibTeX
@article{Hu2025,
  author = {Hu, Fengling and Ren, Zheng and Chen, Luyun and Valcarcel, Alessandra M and Dworkin, Jordan and Renner, Brian and Daboul, Lynn and O'Donnell, Carly M and Verter, Elizabeth D and Manning, Abigail R and Clark, Kelly A and Bae, Eunchan and Chen, Christina and Lou, Carolyn and Satterthwaite, Theodore D and Shou, Haochang and Bilello, Michel and Nakamura, Kunio and Bar-Or, Amit and Calabresi, Peter A and Freeman, Leorah and Henry, Roland G and Longbrake, Erin E and Oh, Jiwon and Schindler, Matthew K and Absinta, Martina and Solomon, Andrew J and Sicotte, Nancy L and Ontaneda, Daniel and Reich, Daniel S and Sati, Pascal and Shinohara, Russell T},
  title = {Automated segmentation of multiple sclerosis lesions, paramagnetic rims, and central vein sign on MRI provides reliable diagnostic biomarkers},
  journal = {Imaging Neuroscience},
  year = {2025},
  volume = {3},
  doi = {10.1162/imag.a.932}
}
Kostenlos starten →