Related Experiment Video
Updated: Jul 11, 2026
Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images
Published on: April 13, 2013
Isointense Infant Brain Segmentation by Stacked Kernel Canonical Correlation Analysis
Li Wang1, Feng Shi1, Yaozong Gao2
1IDEA Lab, Department of Radiology and BRIC, University of North Carolina at Chapel Hill, NC, USA.
Insights
This study introduces a new method for segmenting infant brain MR images, overcoming low contrast issues around 6 months old. The technique uses 12-month-old brain scans to guide the segmentation of younger brains, improving accuracy.
Area of Science:
- Medical Imaging
- Neuroscience
- Computer Vision
Background:
- Infant brain MRI segmentation is difficult due to rapid maturation and myelination.
- Isointense tissue contrast (white and gray matter) around 6 months poses significant challenges for automated segmentation.
Purpose of the Study:
- To develop a novel method for segmenting isointense infant brain MR images.
- To address the challenge of extremely low tissue contrast in 6-month-old infant brains.
Main Methods:
- Proposed a segmentation method based on stacked kernel canonical correlation analysis (KCCA).
- Utilized 12-month-old brain images with high contrast to guide segmentation of 6-month-old images.
- Employed stacked KCCA for optimized common feature representation and sparse patch-based multi-atlas labeling.
Main Results:
- The novel KCCA-based method demonstrated significantly improved performance compared to state-of-the-art techniques.
- Evaluated on 20 isointense brain images using leave-one-out cross-validation.
Conclusions:
- Stacked kernel canonical correlation analysis offers a robust solution for segmenting challenging isointense infant brain MR images.
- The method effectively leverages longitudinal data (12-month-old scans) to enhance segmentation accuracy in younger subjects.
Abstract:
Segmentation of isointense infant brain (at ~6-months-old) MR images is challenging due to the ongoing maturation and myelination process in the first year of life. In particular, signal contrast between white and gray matters inverses around 6 months of age, where brain tissues appear isointense and hence exhibit extremely low tissue contrast, thus posing significant challenges for automated segmentation. In this paper, we propose a novel segmentation method to address the above-mentioned challenges based on stacked kernel canonical correlation analysis (KCCA). Our main idea is to utilize the 12-month-old brain image with high tissue contrast to guide the segmentation of 6-month-old brain images with extremely low contrast. Specifically, we use KCCA to learn the common feature representations for both 6-month-old and the subsequent 12-month-old brain images of same subjects to make their features comparable in the common space. Note that the longitudinal 12-month-old brain images are not required in the testing stage, and they are required only in the KCCA based training stage to provide a set of longitudinal 6- and 12-month-old image pairs for training. Moreover, for optimizing the common feature representations, we propose a stacked KCCA mapping, instead of using only the conventional one-step of KCCA mapping. In this way, we can better use the 12-month-old brain images as multiple atlases to guide the segmentation of isointense brain images. Specifically, sparse patch-based multi-atlas labeling is used to propagate tissue labels in the (12-month-old) atlases and segment isointense brain images by measuring patch similarity between testing and atlas images with their learned common features. The proposed method was evaluated on 20 isointense brain images via leave-one-out cross-validation, showing much better performance than the state-of-the-art methods.
