Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2016:2016:8356294.
doi: 10.1155/2016/8356294. Epub 2016 Mar 16.

Multiscale CNNs for Brain Tumor Segmentation and Diagnosis

Liya Zhao  1 , Kebin Jia  1
Affiliations

Multiscale CNNs for Brain Tumor Segmentation and Diagnosis

Liya Zhao et al. Comput Math Methods Med. 2016.

Abstract

Early brain tumor detection and diagnosis are critical to clinics. Thus segmentation of focused tumor area needs to be accurate, efficient, and robust. In this paper, we propose an automatic brain tumor segmentation method based on Convolutional Neural Networks (CNNs). Traditional CNNs focus only on local features and ignore global region features, which are both important for pixel classification and recognition. Besides, brain tumor can appear in any place of the brain and be any size and shape in patients. We design a three-stream framework named as multiscale CNNs which could automatically detect the optimum top-three scales of the image sizes and combine information from different scales of the regions around that pixel. Datasets provided by Multimodal Brain Tumor Image Segmentation Benchmark (BRATS) organized by MICCAI 2013 are utilized for both training and testing. The designed multiscale CNNs framework also combines multimodal features from T1, T1-enhanced, T2, and FLAIR MRI images. By comparison with traditional CNNs and the best two methods in BRATS 2012 and 2013, our framework shows advances in brain tumor segmentation accuracy and robustness.

PubMed Disclaimer

Figures

Figure 1
Figure 1
The architecture of traditional ImageNet CNNs.
Figure 2
Figure 2
The architecture of traditional MNIST CNNs.
Figure 3
Figure 3
The workflow of automatic selection of proper image patch size.
Figure 4
Figure 4
The architecture of multiscale three-layer neural network.
Figure 5
Figure 5
The three types of data and the manually generated results.
Figure 6
Figure 6
The comparison of different layers of traditional CNNs.
Figure 7
Figure 7
The comparison of different patch size of traditional CNNs.
Figure 8
Figure 8
The accuracy results of methods integrating different patch sizes.
Figure 9
Figure 9
The comparison with other methods.
See this image and copyright information in PMC

References

    1. Menze B. H., Jakab A., Bauer S., et al. The multimodal brain tumor image segmentation benchmark (BRATS) IEEE Transactions on Medical Imaging. 2014;34(10):1993–2024. doi: 10.1109/TMI.2014.2377694. - DOI - PMC - PubMed
    1. Havaei M., Davy A., Warde-Farley D., et al. Brain tumor segmentation with deep neural networks. http://arxiv.org/abs/1505.03540v1. - PubMed
    1. http://cancer.org.
    1. Jiang J., Wu Y., Huang M. Y., Yang W., Chen W., Feng Q. 3D brain tumor segmentation in multimodal MR images based on learning population- and patient-specific feature sets. Computerized Medical Imaging and Graphics. 2013;37(7):512–521. - PubMed
    1. Gordillo N., Montseny E., Sobrevilla P. State of the art survey on MRI brain tumor segmentation. Magnetic Resonance Imaging. 2013;31(8):1426–1438. doi: 10.1016/j.mri.2013.05.002. - DOI - PubMed

Publication types

MeSH terms