EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

2D U‐Net Framework for Multiclass Automated Segmentation of Healthy Structures, Ischemic Stroke, and Glioblastoma in the Rat Brain

Sara Ortega-Espina iD, Sabela Fernández-Rodicio iD, María Luz Alonso-Alonso iD, Pelayo García-Acevedo, Ana Sampedro-Viana, Marcos Bazarra-Barreiros iD, María Teresa Abengoza-Bello, Antonio J Mosqueira, Jacobo Porto-Álvarez iD, Javier Martínez Fernández, Crhistian-Mario Oblitas iD, José Castillo und 2 weitere

NMR in Biomedicine · 2026-12 · 39 (12)

DOI: 10.1002/nbm.70432 ↗

Kurz gefasst

Fachartikel in der Zeitschrift NMR in Biomedicine, online erschienen am 04.11.2026. Verfasst von Sara Ortega-Espina und 13 weiteren Autor:innen. Beteiligte Einrichtung: Neuroimaging and Biotechnology Laboratory (NOBEL), Clinical Neurosciences Research Laboratory (LINC), Health Research Institute of Santiago de Compos…. Hauptthemen laut MEDLINE: Glioblastoma, Brain, Ischemic Stroke, Brain Neoplasms. Die Arbeit nennt 46 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Xunta de Galicia (Axencia Galega de Innovación), Instituto de Salud Carlos III (ISCIII), Spanish Ministry of Science and Innovation.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Rodents are essential for preclinical brain disease research, including stroke, tumors, and neurodegenerative disorders. Magnetic resonance imaging (MRI) enables longitudinal brain tissue visualization, but accurate segmentation is crucial for meaningful analysis. This study introduces a multitask 2D U‐Net, a convolutional neural network designed for automated segmentation and quantification of brain regions, ischemic stroke lesions, and tumors in rats using T 2 ‐weighted imaging (T 2 ‐wi) and T 2 ‐relaxation maps. The model was evaluated on the following three MRI datasets: (i) segmentation of multiple brain regions in healthy rats, (ii) ischemic stroke lesions using T 2 ‐relaxation maps, and (iii) glioblastoma tumors using T 2 ‐wi. Performance was assessed using the Dice coefficient (DC), Hausdorff distance (HD), sensitivity (SEN), and volume ratio (VR), comparing network outputs against semiautomatic ground truths and existing models. A total of 1949 images from 111 rats were analyzed, expanded to 5847 through data augmentation. After correcting an animal‐level train/test partitioning that prevented data leakage between subjects, the model achieved robust accuracy across all tasks (DC = 0.80–0.96, HD = 0.83–2.16 mm, SEN = 0.75–0.96, VR = 0.89–1.10), showing a strong correlation () with manual annotations. This framework improves segmentation accuracy, robustness, and applicability across different MRI settings and conditions, addressing issues like interrater variability and time inefficiency when analyzing rodent brain MRI data. Datasets are publicly available at https://github.com/NOBEL‐MRI/Rat‐Datasets .

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

GlioblastomaBrainIschemic StrokeBrain NeoplasmsImage Processing, Computer-AssistedNeural Networks, ComputerAnimalsRatsMagnetic Resonance ImagingConvolutional Neural NetworksAutomationMale

Stichwörter der Autor:innen

2D U‐NetAIMRIartificial intelligencedeep learninggliobastomasegmentationstroke

Autor:innen und Institutionen

  • Neuroimaging and Biotechnology Laboratory (NOBEL), Clinical Neurosciences Research Laboratory (LINC), Health Research Institute of Santiago de Compostela (IDIS), Santiago de Compostela, Spain.
  • Department of Functional Biology (Physiology), University of Oviedo, Oviedo, Spain.
  • Neuroradiology Group, Health Research Institute of Santiago de Compostela (IDIS), Santiago de Compostela, Spain.

ORCID-Profile: Sara Ortega-Espina, Sabela Fernández-Rodicio, María Luz Alonso-Alonso, Marcos Bazarra-Barreiros, Jacobo Porto-Álvarez, Crhistian-Mario Oblitas, Pablo Hervella, Ramón Iglesias-Rey

Förderung

  • Xunta de Galicia (Axencia Galega de Innovación)
  • Instituto de Salud Carlos III (ISCIII)
  • Spanish Ministry of Science and Innovation
  • Axencia Galega de Innovación
  • Ministerio de Ciencia e Innovación

Zitieren

APA

Ortega-Espina, S., Fernández-Rodicio, S., Alonso-Alonso, M. L., García-Acevedo, P., Sampedro-Viana, A., Bazarra-Barreiros, M., Abengoza-Bello, M. T., Mosqueira, A. J., Porto-Álvarez, J., Fernández, J. M., Oblitas, C. M., Castillo, J., Hervella, P., & Iglesias-Rey, R. (2026). 2D U‐Net Framework for Multiclass Automated Segmentation of Healthy Structures, Ischemic Stroke, and Glioblastoma in the Rat Brain. NMR in Biomedicine, 39(12). https://doi.org/10.1002/nbm.70432

Vancouver

Ortega-Espina S, Fernández-Rodicio S, Alonso-Alonso ML, García-Acevedo P, Sampedro-Viana A, Bazarra-Barreiros M, et al. 2D U‐Net Framework for Multiclass Automated Segmentation of Healthy Structures, Ischemic Stroke, and Glioblastoma in the Rat Brain. NMR in Biomedicine. 2026;39(12). doi:10.1002/nbm.70432

BibTeX
@article{OrtegaEspina2026,
  author = {Ortega-Espina, Sara and Fernández-Rodicio, Sabela and Alonso-Alonso, María Luz and García-Acevedo, Pelayo and Sampedro-Viana, Ana and Bazarra-Barreiros, Marcos and Abengoza-Bello, María Teresa and Mosqueira, Antonio J and Porto-Álvarez, Jacobo and Fernández, Javier Martínez and Oblitas, Crhistian-Mario and Castillo, José and Hervella, Pablo and Iglesias-Rey, Ramón},
  title = {2D U‐Net Framework for Multiclass Automated Segmentation of Healthy Structures, Ischemic Stroke, and Glioblastoma in the Rat Brain},
  journal = {NMR in Biomedicine},
  year = {2026},
  volume = {39},
  number = {12},
  doi = {10.1002/nbm.70432}
}
Kostenlos starten →