Segmentation of Gliomas Based on a Double-Pathway Residual Convolution Neural Network Using Multi-Modality Information

2020 ◽  
Vol 10 (11) ◽  
pp. 2784-2794
Author(s):  
Mingyuan Pan ◽  
Yonghong Shi ◽  
Zhijian Song
Keyword(s):  
Neural Network ◽  
Surgical Navigation ◽  
Conditional Random Field ◽  
Automatic Segmentation ◽  
Convolution Neural Network ◽  
Radiotherapy Planning ◽  
Tumor Segmentation ◽  
Mr Images ◽  
Brain Tumor Segmentation ◽  
Processing Step

The automatic segmentation of brain tumors in magnetic resonance (MR) images is very important in the diagnosis, radiotherapy planning, surgical navigation and several other clinical processes. As the location, size, shape, boundary of gliomas are heterogeneous, segmenting gliomas and intratumoral structures is very difficult. Besides, the multi-center issue makes it more challenging that multimodal brain gliomas images (such as T1, T2, fluid-attenuated inversion recovery (FLAIR), and T1c images) are from different radiation centers. This paper presents a multimodal, multi-scale, double-pathway, 3D residual convolution neural network (CNN) for automatic gliomas segmentation. In the pre-processing step, a robust gray-level normalization method is proposed to solve the multi-center problem, that the intensity range from deferent centers varies a lot. Then, a doublepathway 3D architecture based on DeepMedic toolkit is trained using multi-modality information to fuse the local and context features. In the post-processing step, a fully connected conditional random field (CRF) is built to improve the performance, filling and connecting the isolated segmentations and holes. Experiments on the Multimodal Brain Tumor Segmentation (BRATS) 2017 and 2019 dataset showed that this methods can delineate the whole tumor with a Dice coefficient, a sensitivity and a positive predictive value (PPV) of 0.88, 0.89 and 0.88, respectively. As for the segmentation of the tumor core and the enhancing area, the sensitivity reached 0.80. The results indicated that this method can segment gliomas and intratumoral structures from multimodal MR images accurately, and it possesses a clinical practice value.

scholarly journals Segmentation of Brain Gliomas Based on a Multi-modal Multi-scale Double-pathway 3D Residual CNN

2020 ◽  
Author(s):  
Mingyuan Pan ◽  
Yonghong Shi ◽  
Zhijian Song
Keyword(s):  
Conditional Random Field ◽  
Radiotherapy Planning ◽  
Tumor Segmentation ◽  
Mr Images ◽  
Brain Tumor Segmentation ◽  
Pathway Network ◽  
Multi Scale ◽  
Fluid Attenuated Inversion Recovery ◽  
Fully Connected ◽  
Brain Gliomas

Abstract Background: The automated segmentation of brain gliomas regions in magnetic resonance (MR) images plays an important role in the early diagnosis, intraoperative navigation, radiotherapy planning and prognosis of brain tumors. It is very challenging to segment gliomas and intratumoral structures since the location, size, shape, edema range and boundary of gliomas are heterogeneous, and multimodal brain gliomas images (such as T1, T2, fluid-attenuated inversion recovery (FLAIR), and T1c images) are collected from multiple radiation centers. Methods: This paper presents a multimodal, multi-scale, double-pathway, 3D residual convolution neural network (CNN) for automatic gliomas segmentation. First, a robust gray-level normalization method is proposed to solve the multicenter problem, such as very different intensity ranges due to different imaging protocols. Second, a multi-scale, double-pathway network based on DeepMedic toolkit is trained with different combinations of multimodal MR images for gliomas segmentation. Finally, a fully connected conditional random field (CRF) is used as a post-processing strategy to optimize the segmentation results for addressing the isolated segmentations and holes. Results: Experiments on the Multimodal Brain Tumor Segmentation (BraTS) 2017 and 2019 challenge data show that our methods achieve a good performance in delineating the whole tumor with a Dice coefficient, a sensitivity and a positive predictive value (PPV) of 0.88, 0.89 and 0.88, respectively. Regarding the segmentation of the tumor core and the enhancing area, the sensitivity reached 0.80. Conclusions: Experiments show that our method can accurately segment gliomas and intratumoral structures from multimodal MR images, and it is of great significance to clinical neurosurgery.

Three-Dimensional Parallel Convolution Neural Network Brain Tumor Segmentation Based on Dilated Convolution

2020 ◽  
Vol 57 (14) ◽  
pp. 141009
Author(s):  
冯博文 Feng Bowen ◽  
吕晓琪 Lü Xiaoqi ◽  
谷宇 Gu Yu ◽  
李菁 Li Qing ◽  
刘阳 Liu Yang
Keyword(s):  
Neural Network ◽  
Brain Tumor ◽  
Three Dimensional ◽  
Convolution Neural Network ◽  
Tumor Segmentation ◽  
Brain Tumor Segmentation ◽  
Dilated Convolution

Brain Tumor Detection Using Multipath Convolution Neural Network (CNN)

2020 ◽  
Vol 10 (4) ◽  
pp. 43-53
Author(s):  
Mukesh Kumar Chandrakar ◽  
Anup Mishra
Keyword(s):  
Neural Network ◽  
Brain Tumor ◽  
Performance Metrics ◽  
Research Problem ◽  
Tumor Detection ◽  
Convolution Neural Network ◽  
Detection Methods ◽  
Tumor Segmentation ◽  
Brain Tumor Segmentation ◽  
Medical Image Diagnosis

Brain tumor segmentation is an emerging application of automated medical image diagnosis. Robust approach of brain tumor segmentation and detection is a research problem, and the performance metrics of the existing tumor detection methods are not appropriately known. Deep neural network using convolution neural network (CNN) is being researched in this direction, but no general architecture is found that can be used as robust method for brain tumor detection. The authors have proposed a multipath CNN architecture for brain tumor segmentation and detection, which provides improved results as compared to existing methods. The proposed work has been tested for datasets BRATS2013, BRTAS2015, and BRATS2017 with significant improvement in dice index and timing values by utilizing the capability of multipath CNN architecture, which combines both local and global paths.

scholarly journals Deep Weber Dominant Local Order Based Feature Generator and Improved Convolution Neural Network for Brain Tumor Segmentation in MR Images

2020 ◽  
Vol 9 (3) ◽  
pp. 3150-3157
Keyword(s):  
Neural Network ◽  
Feature Extraction ◽  
Brain Tumor ◽  
Median Filter ◽  
Histogram Equalization ◽  
Convolution Neural Network ◽  
Tumor Segmentation ◽  
Local Order ◽  
Dice Similarity Coefficient ◽  
Brain Tumor Segmentation

This paper introduces a scheme for retrieving deep features to carry out the procedure of recognising brain tumors from MR image. Initially, the MR brain image is denoised through the Modified Decision Based Unsymmetric Trimmed Median Filter (MDBUTMF) after that the contrast of the image is improved through Contrast Limited Adaptive Histogram Equalization (CLAHE). Once the pre-processing task is completed, the next phase is to extract the feature. In order to acquire the features of pre-processed images, this article offers a feature extraction technique named Deep Weber Dominant Local Order Based Feature Generator (DWDLOBFG). Once the deep features are retrieved, the next stage is to separate the brain tumor. Improved Convolution Neural Network (ICNN) is used to achieve this procedure. To explore the efficiency of deep feature extraction and in-depth machine learning methods, four performance indicators were used: Sensitivity (SEN), Jaccard Index (JI), Dice Similarity Coefficient (DSC) and Positive Predictive Value (PPV). The investigational outputs illustrated that the DWDLOBFG and ICNN achieve best outputs than existing techniques.

scholarly journals Segmentation of Brain Tumors Using Three-Dimensional Convolutional Neural Network on MRI Images 3D MedImg-CNN

2021 ◽  
Vol 15 (4) ◽  
pp. 0-0
Keyword(s):  
Neural Network ◽  
Convolutional Neural Network ◽  
High Performance ◽  
Three Dimensional ◽  
Tumor Segmentation ◽  
Dice Similarity Coefficient ◽  
Segmentation Method ◽  
Mr Images ◽  
Brain Tumor Segmentation ◽  
Fully Automatic

We consider the problem of fully automatic brain tumor segmentation in MR images containing glioblastomas. We propose a three Dimensional Convolutional Neural Network (3D MedImg-CNN) approach which achieves high performance while being extremely efficient, a balance that existing methods have struggled to achieve. Our 3D MedImg-CNN is formed directly on the raw image modalities and thus learn a characteristic representation directly from the data. We propose a new cascaded architecture with two pathways that each model normal details in tumors. Fully exploiting the convolutional nature of our model also allows us to segment a complete cerebral image in one minute. The performance of the proposed 3D MedImg-CNN with CNN segmentation method is computed using dice similarity coefficient (DSC). In experiments on the 2013, 2015 and 2017 BraTS challenges datasets; we unveil that our approach is among the most powerful methods in the literature, while also being very effective.

scholarly journals Automated Brain Tumor Segmentation and Identification using MR Images

2019 ◽  
Vol 8 (12) ◽  
pp. 2647-2651
Keyword(s):  
Brain Tumor ◽  
Human Brain ◽  
Computational Time ◽  
Support Vector ◽  
Tumor Segmentation ◽  
Svm Classifier ◽  
Automatic Identification ◽  
Mr Images ◽  
Brain Tumor Segmentation ◽  
Processing Step

Automatic identification of tumor in human brain is a challenging task due to its varying in size, shape and location. This paper proposes a multi-modality technique for the segmentation of brain tumor its classification to differentiate easily between cancerous and non-cancerous tumor from MR images of the human brain. To achieve this, different segmentation and classification techniques have been applied. The important stages involved in the proposed technique are pre-processing, segmentation and classification stages. The pre-processing step is carried out using wavelet transform, segmentation stage is done by applying modified Chan-Vese model and finally the extracted tumor can be classified as benign or malignant using Support Vector Machine (SVM) classifier. The experimental results on MR images prove that, the proposed method is efficient and robust to noise. Moreover, the comparisons with existing techniques also show that, the proposed method takes less computational time and classify the tumors very accuratel

Segmentation of Brain Tumor Tissues in HGG and LGG MR Images Using 3D U-net Convolutional Neural Network

2018 ◽  
Vol 7 (2) ◽  
pp. 18-30 ◽  
Author(s):  
Poornachandra Sandur ◽  
C. Naveena ◽  
V.N. Manjunath Aradhya ◽  
Nagasundara K. B.
Keyword(s):  
Neural Network ◽  
Brain Tumor ◽  
Brain Tumors ◽  
Convolutional Neural Network ◽  
Radiotherapy Planning ◽  
Tumor Segmentation ◽  
Brain Tumor Segmentation ◽  
Diagnosis And Prognosis ◽  
Tumor Tissues ◽  
Magnetic Resonance Imaging Mri

The quantitative assessment of tumor extent is necessary for surgical planning, as well as monitoring of tumor growth or shrinkage, and radiotherapy planning. For brain tumors, magnetic resonance imaging (MRI) is used as a standard for diagnosis and prognosis. Manually segmenting brain tumors from 3D MRI volumes is tedious and depends on inter and intra observer variability. In the clinical facilities, a reliable fully automatic brain tumor segmentation method is necessary for the accurate delineation of tumor sub regions. This article presents a 3D U-net Convolutional Neural Network for segmentation of a brain tumor. The proposed method achieves a mean dice score of 0.83, a specificity of 0.80 and a sensitivity of 0.81 for segmenting the whole tumor, and for the tumor core region a mean dice score of 0.76, a specificity of 0.79 and a sensitivity of 0.73. For the enhancing region, the mean dice score is 0.68, a specificity of 0.73 and a sensitivity of 0.77. From the experimental analysis, the proposed U-net model achieved considerably good results compared to the other segmentation models.

Sign in / Sign up

Export Citation Format

Share Document