2D U‐Net Framework for Multiclass Automated Segmentation of Healthy Structures, Ischemic Stroke, and Glioblastoma in the Rat Brain
Open Access
- 4 November 2026
- journal article
- Published by Wiley in NMR in Biomedicine
- Vol. 39 (12)
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 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 .Keywords
Funding Information
- Axencia Galega de Innovación (IN607A2022‐03)
- Axencia Galega de Innovación (IN607A2026‐06)
- Axencia Galega de Innovación (IN606B‐2024.1)
- Instituto de Salud Carlos III (ISCIII/PI21/01256/)
- Instituto de Salud Carlos III (PI24/00813)
- Instituto de Salud Carlos III (DTS23/00103)
- Instituto de Salud Carlos III (AC21_2/00014)
- Instituto de Salud Carlos III (CP22/00061)
- Instituto de Salud Carlos III (FI22/00200)
- Instituto de Salud Carlos III (CM23/00173)
- Instituto de Salud Carlos III (JR25/00032)
- Ministerio de Ciencia e Innovación (PDC2021‐121455‐I00)
- Ministerio de Ciencia e Innovación (PID2022‐141258OB‐I00)
- Ministerio de Ciencia e Innovación (PRTR‐C17.I1)
This publication has 40 references indexed in Scilit:
- DiSCNet: Directional Split Convolution for compute-efficient brain tumor diagnosisComputational Biology and Chemistry, 2026
- Adaptive and efficient deep learning model for automated ischemic stroke lesion segmentationBiomedical Signal Processing and Control, 2026
- Automatic Cerebral Hemisphere Segmentation in Rat MRI with Ischemic Lesions via Attention-based Convolutional Neural NetworksNeuroinformatics, 2022
- Stereotaxic Implantation of F98 Cells in Fischer Rats: A Syngeneic Model to Investigate Photodynamic Therapy Response in Glioma2022
- Multi-level Kronecker Convolutional Neural Network (ML-KCNN) for Glioma Segmentation from Multi-modal MRI Volumetric DataJournal of Digital Imaging, 2021
- A Review of Translational Magnetic Resonance Imaging in Human and Rodent Experimental Models of Small Vessel DiseaseTranslational Stroke Research, 2020
- Automatic Brain Extraction for Rodent MRI ImagesNeuroinformatics, 2020
- A web‐based brain metastases segmentation and labeling platform for stereotactic radiosurgeryMedical Physics, 2020
- Automated brain extraction of multisequence MRI using artificial neural networksHuman Brain Mapping, 2019
- Deep convolutional neural networks for brain image analysis on magnetic resonance imaging: a reviewArtificial Intelligence in Medicine, 2019