Attention U-Net with Active Contour Based Hybrid Loss for Brain Tumor Segmentation

2021 ◽  
pp. 35-45
Author(s):  
Dang-Tien Nguyen ◽  
Thi-Thao Tran ◽  
Van-Truong Pham
Keyword(s):  
Brain Tumor ◽  
Active Contour ◽  
Tumor Segmentation ◽  
Brain Tumor Segmentation

Hybrid active contour model and deep belief network based approach for brain tumor segmentation and classification

Sensor Review ◽  
2019 ◽  
Vol 39 (4) ◽  
pp. 473-487 ◽  
Author(s):  
Ayalapogu Ratna Raju ◽  
Suresh Pabboju ◽  
Ramisetty Rajeswara Rao
Keyword(s):  
Brain Tumor ◽  
Sensitivity And Specificity ◽  
Active Contour ◽  
Active Contour Model ◽  
Tumor Segmentation ◽  
Brain Tumor Segmentation ◽  
Content Type ◽  
Contour Model ◽  
The Brain ◽  
Mri Image

Purpose Brain tumor segmentation and classification is the interesting area for differentiating the tumorous and the non-tumorous cells in the brain and classifies the tumorous cells for identifying its level. The methods developed so far lack the automatic classification, consuming considerable time for the classification. In this work, a novel brain tumor classification approach, namely, harmony cuckoo search-based deep belief network (HCS-DBN) has been proposed. Here, the images present in the database are segmented based on the newly developed hybrid active contour (HAC) segmentation model, which is the integration of the Bayesian fuzzy clustering (BFC) and the active contour model. The proposed HCS-DBN algorithm is trained with the features obtained from the segmented images. Finally, the classifier provides the information about the tumor class in each slice available in the database. Experimentation of the proposed HAC and the HCS-DBN algorithm is done using the MRI image available in the BRATS database, and results are observed. The simulation results prove that the proposed HAC and the HCS-DBN algorithm have an overall better performance with the values of 0.945, 0.9695 and 0.99348 for accuracy, sensitivity and specificity, respectively. Design/methodology/approach The proposed HAC segmentation approach integrates the properties of the AC model and BFC. Initially, the brain image with different modalities is subjected to segmentation with the BFC and AC models. Then, the Laplacian correction is applied to fuse the segmented outputs from each model. Finally, the proposed HAC segmentation provides the error-free segments of the brain tumor regions prevailing in the MRI image. The next step is to extract the useful features, based on scattering transform, wavelet transform and local Gabor binary pattern, from the segmented brain image. Finally, the extracted features from each segment are provided to the DBN for the training, and the HCS algorithm chooses the optimal weights for DBN training. Findings The experimentation of the proposed HAC with the HCS-DBN algorithm is analyzed with the standard BRATS database, and its performance is evaluated based on metrics such as accuracy, sensitivity and specificity. The simulation results of the proposed HAC with the HCS-DBN algorithm are compared against existing works such as k-NN, NN, multi-SVM and multi-SVNN. The results achieved by the proposed HAC with the HCS-DBN algorithm are eventually higher than the existing works with the values of 0.945, 0.9695 and 0.99348 for accuracy, sensitivity and specificity, respectively. Originality/value This work presents the brain tumor segmentation and the classification scheme by introducing the HAC-based segmentation model. The proposed HAC model combines the BFC and the active contour model through a fusion process, using the Laplacian correction probability for segmenting the slices in the database.

A novel content-based active contour model for brain tumor segmentation

2012 ◽  
Vol 30 (5) ◽  
pp. 694-715 ◽  
Author(s):  
Jainy Sachdeva ◽  
Vinod Kumar ◽  
Indra Gupta ◽  
Niranjan Khandelwal ◽  
Chirag Kamal Ahuja
Keyword(s):  
Brain Tumor ◽  
Active Contour ◽  
Active Contour Model ◽  
Tumor Segmentation ◽  
Brain Tumor Segmentation ◽  
Contour Model

scholarly journals Automatic selection of localized region-based active contour models using image content analysis applied to brain tumor segmentation

2017 ◽  
Vol 91 ◽  
pp. 69-79 ◽  
Author(s):  
Elisee Ilunga–Mbuyamba ◽  
Juan Gabriel Avina–Cervantes ◽  
Jonathan Cepeda–Negrete ◽  
Mario Alberto Ibarra–Manzano ◽  
Claire Chalopin
Keyword(s):  
Content Analysis ◽  
Brain Tumor ◽  
Active Contour ◽  
Tumor Segmentation ◽  
Image Content ◽  
Brain Tumor Segmentation ◽  
Active Contour Models ◽  
Automatic Selection ◽  
Contour Models ◽  
Selection Of

Automatic MRI Brain Tumor Segmentation System Based on Localizing Active Contour Models

2011 ◽  
Vol 219-220 ◽  
pp. 1342-1346 ◽  
Author(s):  
Ying Wang ◽  
Zhi Xian Lin ◽  
Jian Guo Cao ◽  
Mao Qing Li
Keyword(s):  
Brain Tumor ◽  
Active Contour ◽  
Automatic Segmentation ◽  
Brain Mri ◽  
Initial Contour ◽  
Tumor Segmentation ◽  
Brain Tumor Segmentation ◽  
Active Contour Models ◽  
Mri Brain ◽  
Contour Models

In this paper, an automatic segmentation system was developed for MRI brain tumor. Local region-based active contour models were suitable for heterogeneous features of brain MRI image. But the models are sensitive to initial contour, which generally requires manual setting. An automatic MRI brain tumor segmentation system were developed based on localized contour models, which can identify tumor-dominant slice, set initial contour automatically and segment tumor’s contours from all MRI slices autonomously. K-means clustering and grayscale analysis were combined to identify tumor-dominant slice. Multi-threshold algorithm with the aid of erosion and dilation operators was adopted to obtain an initial contour for the tumor-dominant slice. The segmentation contour from the local active contour models was applied as initial contours of two-side neighboring slices. MRI brain tumor data were applied to validate the automatic segmentation system.

Sign in / Sign up

Export Citation Format

Share Document