Presentation Details
An Automatic Method for Brain Tumors Segmentation Based on Deep Convolutional Neural Network
ID：66
View Protection：ATTENDEE
Updated Time：2021-11-08 09:25:08
Hits：1842
Oral Presentation
Start Time：2021-11-13 16:15（Asia/Shanghai）
Duration：15min
Session：PS1 Plenary Session 1 »
No files
Abstract
[Purpose] Accurate outline of tumor targets is critical to a high quality radiotherapy plan. Manual segmentation is of great workload and has a strong artificial subjectivity. Using deep learning method to assist automatically segmenting of tumor target is the penetration and application of artificial intelligence in medicine.
[Methods] A 6-layer model of deep Convolution Neural Network (CNN) has been constructed by taking advantage of different types of layers for brain tumor segmentation. This model is a 6 layer CNN model (6-CNN) composed of three convolution layers, two pool layers and one full connection layer. To obtain enough samples for 6-CNN model training, a patch-based technology has been adopted. That is to successively extract a local area from the whole image as a patch. And the center pixel value is taken as the pixel value of the whole patch. Similarly, the label of the center pixel is also taken as the label of the whole patch. Thus the 6-CNN model transforms the brain tumor image segmentation into patch classification based on the excellent classification characteristics of deep convolution neural network. The model combines the local features of patch, the information extracted from shallow network and the global features to predict the category label of the central pixel of patch.
[Results] The model is validated on BRATS 2015 dataset and results show that the segmentation accuracy can be up to Dice Similarity Coefficient (DSC) 90%±4%.
[Conclusions] An automatic deep CNN segmentation model for brain tumors has been constructed based on MRI image patches, which is expected to assist or even substitute the manual segmentation of brain tumors.
[Methods] A 6-layer model of deep Convolution Neural Network (CNN) has been constructed by taking advantage of different types of layers for brain tumor segmentation. This model is a 6 layer CNN model (6-CNN) composed of three convolution layers, two pool layers and one full connection layer. To obtain enough samples for 6-CNN model training, a patch-based technology has been adopted. That is to successively extract a local area from the whole image as a patch. And the center pixel value is taken as the pixel value of the whole patch. Similarly, the label of the center pixel is also taken as the label of the whole patch. Thus the 6-CNN model transforms the brain tumor image segmentation into patch classification based on the excellent classification characteristics of deep convolution neural network. The model combines the local features of patch, the information extracted from shallow network and the global features to predict the category label of the central pixel of patch.
[Results] The model is validated on BRATS 2015 dataset and results show that the segmentation accuracy can be up to Dice Similarity Coefficient (DSC) 90%±4%.
[Conclusions] An automatic deep CNN segmentation model for brain tumors has been constructed based on MRI image patches, which is expected to assist or even substitute the manual segmentation of brain tumors.
Keywords
Accurately radiotherapy,Brain Tumor Segmentation,Deep Learning,Convolutional Neural Network
Speaker
辉 林
合肥工业大学Submission Author
All Comments
Important Date
-
Conference Date
Nov 13
2021
toNov 14
2021
-
Sep 30 2021
Contribution Submission Deadline
-
Nov 14 2021
Registration deadline
Sponsored By
Medical Physics Branch of Chinese Society of Biomedical Engineering
IEEE Beijing Section
Life Electronics Society of Chinese Institute of Electronics
IEEE Beijing Section
Life Electronics Society of Chinese Institute of Electronics
Organized By
Anhui Biomedical Engineering Society.
University of Science and Technology of China
University of Science and Technology of China
Contact Information
- Mrs. Yang
- ic******@ustc.edu.cn
- +86********
Previous Conferences
-
2017-11-04 China Beijing,China
2017 International Conference on Medical Imaging Physics and Engineering -
2017-11-03 China 海淀区
2017 IEEE 国际医学影像物理和工程大会 -
2013-10-19 China 沈阳市
2013 IEEE International Conference on Medical Imaging Physics and Engineering (ICMIPE)
Submit Comment