Deep Convolutional Neural Networks for Multi-Modality Isointense Infant Brain Image Segmentation

By Wenlu Zhang, Rongjian Li, Houtao Deng, Li Wang, Weili Lin, Shuiwang Ji, Dinggang Shen. NeuroImage, 2015

800 citations recorded by the source corpus.

The segmentation of infant brain tissue images into white matter (WM), gray matter (GM), and cerebrospinal fluid (CSF) plays an important role in studying early brain development in health and disease. In the isointense stage (approximately 6–8 months of age), WM and GM exhibit similar levels of intensity in both T1 and T2 MR images, making the tissue segmentation very challenging. Only a small number of existing methods have been designed for tissue segmentation in this isointense stage; however, they only used a single T1 or T2 images, or the combination of T1 and T2 images. In this paper, we propose to use deep convolutional neural networks (CNNs) for segmenting isointense stage brain tissues using multi-modality MR images. CNNs are a type of deep models in which trainable filters and local neighborhood pooling operations are applied alternatingly on the raw input images, resulting in a hierarchy of increasingly complex features. Specifically, we used multimodality information from T1, T2, and fractional anisotropy (FA) images as inputs and then generated the segmentation maps as outputs. The multiple intermediate layers applied convolution, pooling, normalization, and other operations to capture the highly nonlinear mappings between inputs and outputs. We compared the performance of our approach with that of the commonly used segmentation methods on a set of manually segmented isointense stage brain images. Results showed that our proposed model significantly outperformed prior methods on infant brain tissue segmentation. In addition, our results indicated that integration of multi-modality images led to significant performance improvement.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/1e17cc9d319f5dc886bf1382acaf3318de9b2e7f

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Computer Vision and Pattern Recognition

Citations in the corpus

Deep convolutional neural networks for brain image analysis on magnetic resonance imaging: a review (2017, methodology): “The purpose of post-processing could be either to create hard segmentation labels [23] or to remove false positives [16, 26, 22, 24, 122, 130] from segmentation results, using different algorithms such as connected component analysis and Conditional Random Fields (CRF).”
Improving automated multiple sclerosis lesion segmentation with a cascaded 3D convolutional neural network approach (2017, background): “CNNs have also gained popularity in brain imaging, specially in tissue segmentation (Zhang et al., 2015; Moeskops et al., 2016a) and brain tumor segmentation (Kamnitsas et al.”
Deep Learning in Medical Image Analysis (2017, background): “Particularly, great improvements in computer vision inspired its use to medical image analysis such as image segmentation (22, 23), image registration (24), image fusion (25), image annotation (26), computer-aided diagnosis and prognosis (27, 28, 29), lesion/landmark detection (30, 31, 32), and…”
Automatic Segmentation of MR Brain Images With a Convolutional Neural Network (2016, methodology): “[32] presented a method using CNNs for the segmentation of three tissue types: white matter (WM), gray matter (GM), and cerebrospinal fluid (CSF), in MR brain images of 6–8 months old infants, which have low contrast between WM and GM.”
CT Super-Resolution GAN Constrained by the Identical, Residual, and Cycle Learning Ensemble (GAN-CIRCLE) (2018, background): “Convolutional Neural Networks (CNNs) learn hierarchical features and representations, leadning to a large number of CNN-based SR models for natural images [32]–[39].”
How much data is needed to train a medical image deep learning system to achieve necessary high accuracy (2015, background): “Several approaches to this question have been introduced and explored in different applications (Figueroa et al. (2012), Beleites et al. (2013), Dobbin & Simon (2007), Dobbin et al. (2008)).”
PSNet: prostate segmentation on MRI based on a convolutional neural network (2018, background): “Zhang et al. 10 proposed to use deep CNNs for segmenting iso-intense stage brain tissues using multimodality MRI.”
Deep Learning for Brain MRI Segmentation: State of the Art and Future Directions (2017, methodology): “Zhang et al. [27] presented a 2D (input patch size 13 × 13 pixels) patch-wise CNN approach to segment WM, GM, and CSF from multimodal (i.e., T1, T2, and fractional anisotropy) MR images of infants.”
Breast Cancer Multi-classification from Histopathological Images with Structured Deep Learning Model (2017, methodology): “Furthermore, although CNNs inspired by Kunihiko Fukushima 26, 27 , has been used for medical image analysis, e.g., image segmentation 28, 29 image fusion and registration 30–32 , but there still exists a lot of room for improvement of medical data in comparison with the computer vision domain 7,…”
Data Augmentation Using Learned Transformations for One-Shot Medical Image Segmentation (2019, methodology): “To mitigate the need for large labeled training datasets, these methods often use data augmentation along with hand-engineered pre-processing steps and architectures [2, 40, 53, 57, 63, 65, 82].”
Modified U-Net (mU-Net) with Incorporation of Object-Dependent High Level Features for Improved Liver and Liver-Tumor Segmentation in CT Images (2019, methodology): “emerged and have been applied for liver [11]–[14], head and neck [15], [16], prostate [17], and brain [18], [19]”
Deep Learning for Health Informatics (2017, methodology): “CNNs have also been proposed for the segmentation of isointense stage brain tissues [131] and brain extraction from multimodality MR im-”
A deep learning model integrating FCNNs and CRFs for brain tumor segmentation (2017, methodology): “To train the deep CNNs with sufficient training samples, image patch-based techniques are adopted [20-25, 28, 30, 31, 38-40].”