Related Experiment Video
Updated: Dec 3, 2025
08:49
Whole-Brain Single-Cell Imaging and Analysis of Intact Neonatal Mouse Brains Using MRI, Tissue Clearing, and Light-Sheet Microscopy
Published on: August 1, 2022
4.0K
Automatic Skull Stripping of Rat and Mouse Brain MRI Data Using U-Net
Li-Ming Hsu1,2,3,4, Shuai Wang2,4, Paridhi Ranadive1
1Center for Animal Magnetic Resonance Imaging, The University of North Carolina at Chapel Hill, Chapel Hill, NC, United States.
Frontiers in Neuroscience
|October 29, 2020
Summary
This study introduces a U-Net deep learning framework for automated rodent brain magnetic resonance imaging (MRI) skull stripping. The method accurately identifies brain boundaries, overcoming challenges in rodent neuroimaging.
Area of Science:
- Neuroimaging
- Medical Image Analysis
- Computational Neuroscience
Background:
- Accurate skull stripping is crucial for rodent brain MRI pre-processing.
- Manual skull stripping is operator-dependent and time-consuming.
- Rodent skull stripping is challenging due to anatomical and imaging differences compared to humans.
Purpose of the Study:
- To develop and validate a deep learning framework (U-Net) for automated rodent brain MRI skull stripping.
- To improve accuracy and reduce operator dependence in rodent neuroimaging.
Main Methods:
- A U-Net deep learning model was implemented for automatic brain boundary identification.
- The model was trained and validated using both in-house and public rodent MRI datasets.
- Performance was benchmarked against state-of-the-art methods.
Main Results:
- The U-Net framework demonstrated robust performance across various MRI protocols in rats and mice.
- Achieved superior averaged Dice similarity coefficients compared to existing methods.
- Significantly outperformed current state-of-the-art approaches (p < 0.05).
Conclusions:
- The proposed U-Net method offers an automated, accurate, and robust solution for rodent brain MRI skull stripping.
- This approach eliminates operator dependence and enhances efficiency in neuroimaging pipelines.
- The framework shows significant potential for advancing rodent neuroimaging research.
