Related Experiment Video
Updated: Jul 30, 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
3.6K
Automated identification of piglet brain tissue from MRI images using Region-based Convolutional Neural Networks.
Kayla L Stanke1, Ryan J Larsen1, Laurie Rund1
1Department of Animal Sciences, University of Illinois Urbana-Champaign, Champaign, Illinois, United States of America.
Plos One
|May 11, 2023
Summary
Automated piglet brain extraction using Mask R-CNN significantly reduces analysis time and improves uniformity. This deep learning method offers a viable alternative to manual skull stripping in neuroimaging research.
Area of Science:
- Neuroimaging
- Developmental Biology
- Machine Learning
Background:
- Magnetic resonance imaging (MRI) is crucial for studying piglet brain development.
- Manual brain extraction (skull stripping) is time-consuming and prone to inter-rater variability.
- Automated methods are needed to enhance efficiency and consistency in neuroimaging analysis.
Purpose of the Study:
- To demonstrate the efficacy of Mask R-CNN, a deep learning model, for automated piglet brain extraction.
- To validate the accuracy and reliability of the automated skull stripping method.
- To provide a faster and more consistent approach for piglet brain imaging analysis.
Main Methods:
- Utilized Mask R-CNN, a Region-based Convolutional Neural Network, for automated brain extraction.
- Employed Nested Cross-Validation for rigorous model validation.
- Trained and validated the model on imaging data from 32 piglets across six datasets.
Main Results:
- Achieved high accuracy in automated brain extractions, confirmed by visual inspection.
- Reported Dice coefficients ranging from 0.95 to 0.97, indicating excellent overlap.
- Obtained Hausdorff Distance values between 4.1 and 8.3 voxels, demonstrating precise boundary detection.
Conclusions:
- Mask R-CNN provides a highly accurate and reliable method for automated piglet brain extraction (skull stripping).
- This deep learning approach significantly improves efficiency and consistency compared to manual methods.
- The validated R-CNN model is a valuable tool for advancing piglet neuroimaging research and developmental studies.
