A Convolutional Neural Network Approach to Brain Tumor Segmentation

Havaei, Mohammad; Dutil, Francis; Pal, Chris; Larochelle, Hugo; Jodoin, Pierre-Marc

doi:10.1007/978-3-319-30858-6_17

Cited by 75 publications

(49 citation statements)

References 2 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A deep learning model of CNNs usually has millions or even billions of parameters. To train the deep CNNs with sufficient training samples, image patch-based techniques are adopted (Zikic et al, 2014 ; Davy et al, 2014 ; Urban et al, 2014 ; Dvorak and Menze, 2015 ; Havaei et al, 2015, 2017; Pereira et al, 2015 ; Kamnitsas et al, 2017 ; Pereira et al, 2016 ; Zhang et al, 2015 ; Moeskops et al, 2016 ; de Brebisson and Montana, 2015). With the image patch based representation, the image segmentation problem can be solved as a classification problem of image patches.…”

Section: Methodsmentioning

confidence: 99%

“…The proposed method is able to segment brain images slice-by-slice, which is much faster than the image patch based segmentation methods. Our method could achieve competitive segmentation performance based on 3 MR imaging modalities (Flair, T1c, T2), rather than 4 modalities (Flair, T1, T1c, T2) (Menze et al, 2015 ; Goetz et al, 2014 ; Reza and Iftekharuddin, 2014 ; Kleesiek et al, 2014 ; Meier et al, 2014 ; Zikic et al, 2014 ; Davy et al, 2014 ; Urban et al, 2014 ; Dvorak and Menze, 2015 ; Havaei et al, 2015 ; Pereira et al, 2015 ; Agn et al, 2015 ; Vaidhya et al, 2015 ; Kamnitsas et al, 2017 ; Yi et al, 2016 ; Havaei et al, 2017 ; Pereira et al, 2016), which could help reduce the cost of data acquisition and storage. We have evaluated our method based on imaging data provided by the Multimodal Brain Tumor Image Segmentation Challenge (BRATS) 2013, the BRATS 2015, and the BRATS 2016.…”

Section: Introductionmentioning

confidence: 97%

“…Particularly, Convolutional Neural Networks (CNNs) were adopted for brain tumor image segmentation in the Multimodal Brain Tumor Image Segmentation Challenge (BRATS) 2014 (Zikic et al, 2014 ; Davy et al, 2014 ; Urban et al, 2014). More deep learning based brain tumor segmentation methods were presented in the BRATS 2015 and different deep learning models were adopted, including CNNs (Dvorak and Menze, 2015 ; Havaei et al, 2015 ; Pereira et al, 2015), convolutional restricted Boltzman machines (Agn et al, 2015), and Stacked Denoising Autoencoders (Vaidhya et al, 2015). …”

Section: Introductionmentioning

confidence: 99%

“…Particularly, both 3D-CNNs (Urban et al, 2014 ; Kamnitsas et al, 2017 ; Yi et al, 2016) and 2D-CNNs (Zikic et al, 2014 ; Davy et al, 2014 ; Dvorak and Menze, 2015 ; Havaei et al, 2015 ; Pereira et al, 2015 ; Havaei et al, 2017 ; Pereira et al, 2016) models were adopted to build tumor segmentation methods. Although 3D-CNNs can potentially take full advantage of 3D information of the MRI data, the network size and computational cost are increased too.…”

Section: Introductionmentioning

confidence: 99%

“…Most of the above CNN brain tumor segmentation methods assumed that each voxel’s label is independent, and they didn’t take the appearance and spatial consistency into consideration. To take the local dependencies of labels into account, Havaei et al constructed a cascaded architecture by taking the pixel-wise probability segmentation results obtained by CNNs trained at early stages as additional input to their following CNNs (Havaei et al, 2015, 2017). To take into consideration appearance and spatial consistency of the segmentation results, Markov Random Fields (MRFs), particularly Conditional Random Fields (CRFs), have been integrated with deep learning techniques in image segmentation studies, either used as a post-process step of CNNs (Kamnitsas et al, 2017 ; Chen et al, 2014) or formulated as neural networks (Zheng et al, 2015 ; Liu et al, 2015).…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

A deep learning model integrating FCNNs and CRFs for brain tumor segmentation

Zhao

Song

et al. 2018

Medical Image Analysis

672

336

View full text Add to dashboard Cite

Accurate and reliable brain tumor segmentation is a critical component in cancer diagnosis, treatment planning, and treatment outcome evaluation. Build upon successful deep learning techniques, a novel brain tumor segmentation method is developed by integrating fully convolutional neural networks (FCNNs) and Conditional Random Fields (CRFs) in a unified framework to obtain segmentation results with appearance and spatial consistency. We train a deep learning based segmentation model using 2D image patches and image slices in following steps: 1) training FCNNs using image patches; 2) training CRFs as Recurrent Neural Networks (CRF-RNN) using image slices with parameters of FCNNs fixed; and 3) fine-tuning the FCNNs and the CRF-RNN using image slices. Particularly, we train 3 segmentation models using 2D image patches and slices obtained in axial, coronal and sagittal views respectively, and combine them to segment brain tumors using a voting based fusion strategy. Our method could segment brain images slice-by-slice, much faster than those based on image patches. We have evaluated our method based on imaging data provided by the Multimodal Brain Tumor Image Segmentation Challenge (BRATS) 2013, BRATS 2015 and BRATS 2016. The experimental results have demonstrated that our method could build a segmentation model with Flair, T1c, and T2 scans and achieve competitive performance as those built with Flair, T1, T1c, and T2 scans.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 97%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

A deep learning model integrating FCNNs and CRFs for brain tumor segmentation

Zhao

Song

et al. 2018

Medical Image Analysis

672

336

View full text Add to dashboard Cite

show abstract

Brain tumor segmentation using holistically nested neural networks in MRI images

et al. 2017

View full text Add to dashboard Cite

Purpose Gliomas are rapidly progressive, neurologically devastating, largely fatal brain tumors. Magnetic Resonance Imaging (MRI) is a widely used technique employed in the diagnosis and management of gliomas in clinical practice. MRI is also the standard imaging modality used to delineate the brain tumor target as part of treatment planning for the administration of radiation therapy. Despite more than 20 years of research and development, computational brain tumor segmentation in MRI images remains a challenging task. We are presenting a novel method of automatic image segmentation based on holistically-nested neural networks that could be employed for brain tumor segmentation of MRI images. Methods Two preprocessing techniques were applied to MRI images. The N4ITK method was employed for correction of bias field distortion. A novel landmark-based intensity normalization method was developed so that tissue types have a similar intensity scale in images of different subjects for the same MRI protocol. The holistically-nested neural networks (HNN), which extend from the convolutional neural networks (CNN) with a deep supervision through an additional weighted-fusion output layer, was trained to learn the multi-scale and multi-level hierarchical appearance representation of the brain tumor in MRI images and was subsequently applied to produce a prediction map of the brain tumor on test images. Finally, the brain tumor was obtained through an optimum thresholding on the prediction map. Results The proposed method was evaluated on both the Multimodal Brain Tumor Image Segmentation (BRATS) Benchmark 2013 training data sets, and clinical data from our institute. A dice similarity coefficient (DSC) and sensitivity of 0.78 and 0.81 were achieved on 20 BRATS 2013 training data sets with High Grade Gliomas (HGG), based on a two-fold cross-validation. The HNN model built on the BRATS 2013 training data was applied to 10 clinical data sets with HGG from a locally developed database. DSC and sensitivity of 0.83 and 0.85 were achieved. A quantitative comparison indicated that the proposed method outperforms the popular fully convolutional network (FCN) method. In terms of efficiency, the proposed method took around 10 hours for training with 50,000 iterations, and approximately 30 seconds for testing of a typical MRI image in the BRATS 2013 data set with a size of 160×216×176, using a DELL PRECISION workstation T7400, with an NVIDIA Tesla K20c GPU. Conclusions An effective brain tumor segmentation method for MRI images based on a HNN has been developed. The high level of accuracy and efficiency make this method practical in brain tumor segmentation. It may play a crucial role in both brain tumor diagnostic analysis and in the treatment planning of radiation therapy.

show abstract