Related Experiment Video
Updated: Aug 7, 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
3.7K
An automatic and accurate deep learning-based neuroimaging pipeline for the neonatal brain
Dan Dan Shen1, Shan Lei Bao2, Yan Wang1
1Department of Medical Imaging, Affiliated Hospital and Medical School of Nantong University, NO.20 Xisi Road, Nantong, Jiangsu, 226001, People's Republic of China.
Pediatric Radiology
|March 8, 2023
Summary
This study introduces an automated deep learning pipeline for precise neonatal brain segmentation and analysis using MRI. The developed system demonstrates high accuracy and reliability for both normal and abnormal brain development studies.
Area of Science:
- Neuroimaging
- Medical Image Analysis
- Artificial Intelligence in Medicine
Background:
- Accurate segmentation of neonatal brain structures is vital for understanding development and diagnosing disorders.
- A need exists for automated, end-to-end pipelines for neonatal brain imaging analysis.
Purpose of the Study:
- To develop and validate a deep learning-based pipeline for segmenting and analyzing neonatal brain structural MRI.
- To assess the pipeline's accuracy, robustness, and generalizability across different MRI data.
Main Methods:
- A deep learning architecture was developed for segmenting neonatal brains into 9 tissues and 87 structures.
- Validation involved two cohorts (n=582 and n=37) using metrics like Dice Similarity Score (DSC) and Hausdorff distance (H95).
- Regional volume and cortical surface analysis were performed using FSL, with Intraclass Correlation Coefficient (ICC) assessing reliability.
Main Results:
- The pipeline achieved excellent segmentation performance (DSC: 0.96, H95: 0.99 mm) for thin-slice MRI.
- Good agreement with ground truth was observed for regional volume (ICC > 0.80) and cortical surface analysis.
- Comparable results were obtained for thick-slice MRI (DSC: 0.92, H95: 3.00 mm).
Conclusions:
- An automatic, accurate, and reliable pipeline for neonatal brain segmentation and analysis from structural MRI has been developed.
- The pipeline demonstrated excellent reproducibility and stability across different MRI resolutions.
- This tool facilitates advanced research into neonatal brain development and disorders.
