2D U‐Net Framework for Multiclass Automated Segmentation of Healthy Structures, Ischemic Stroke, and Glioblastoma in the Rat Brain
- 1Neuroimaging and Biotechnology Laboratory (NOBEL), Clinical Neurosciences Research Laboratory (LINC) Health Research Institute of Santiago de Compostela (IDIS) Santiago de Compostela Spain
- 2Department of Functional Biology (Physiology), University of Oviedo Oviedo Spain
- 3Neuroradiology Group Health Research Institute of Santiago de Compostela (IDIS) Santiago de Compostela Spain
Abstract
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 T2‐weighted imaging (T2‐wi) and T2‐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 T2‐relaxation maps, and (iii) glioblastoma tumors using T2‐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 (ρ=0.82−0.99) 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.
Related
- COVLIAS 3.5: integration of attention-based segmentation technique with fuzzy dilated convolutional neural networks for improved classification of chest X-ray scans for multiclass pneumonia diagnosis
- Automated thigh muscle segmentation using foundation models: validation and impact of segmentation quality
- CT Ventricular Volume Reference Percentile Curves Using Automated nnU-Net Segmentation: A Study with LLM-Facilitated Cohort Selection
- Automated corneal endothelial cell detection and segmentation in specular microscopy images using Attention and Faster U-Net
- Automated Segmentation of Mass Lesions in Mammography: Mask R-CNN Versus Classical Methods, Fuzzy Clustering and MedSAM
BibTeX
@article{ortegaespina2026unet,
title = {{2D U‐Net Framework for Multiclass Automated Segmentation of Healthy Structures, Ischemic Stroke, and Glioblastoma in the Rat Brain}},
author = {Sara Ortega‐Espina and Sabela Fernández‐Rodicio and María Luz Alonso‐Alonso and Pelayo García‐Acevedo and Ana Sampedro‐Viana and Marcos Bazarra‐Barreiros and María Teresa Abengoza‐Bello and Antonio J. Mosqueira and Jacobo Porto‐Álvarez and Javier Martínez Fernández and Crhistian‐Mario Oblitas and José Castillo and Pablo Hervella and Ramón Iglesias‐Rey},
journal = {Nmr in Biomedicine},
year = {2026},
month = nov,
doi = {10.1002/nbm.70432},
url = {https://doi.org/10.1002/nbm.70432},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - 2D U‐Net Framework for Multiclass Automated Segmentation of Healthy Structures, Ischemic Stroke, and Glioblastoma in the Rat Brain
AU - Sara Ortega‐Espina
AU - Sabela Fernández‐Rodicio
AU - María Luz Alonso‐Alonso
AU - Pelayo García‐Acevedo
AU - Ana Sampedro‐Viana
AU - Marcos Bazarra‐Barreiros
AU - María Teresa Abengoza‐Bello
AU - Antonio J. Mosqueira
AU - Jacobo Porto‐Álvarez
AU - Javier Martínez Fernández
AU - Crhistian‐Mario Oblitas
AU - José Castillo
AU - Pablo Hervella
AU - Ramón Iglesias‐Rey
PY - 2026
DA - 2026/11/04
JO - Nmr in Biomedicine
DO - 10.1002/nbm.70432
UR - https://doi.org/10.1002/nbm.70432
AB - 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 T2‐weighted imaging (T2‐wi) and T2‐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 T2‐relaxation maps, and (iii) glioblastoma tumors using T2‐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 (ρ=0.82−0.99) 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.
SN - 0952-3480
SN - 1099-1492
ER -