Brain Tumor Segmentation Using Convolutional Neural Networks in MRI Images

Thaha, M. Mohammed; Kumar, K. Pradeep Mohan; Murugan, B. S.; Dhanasekeran, S.; Vijayakarthick, P.; Selvi, A. Senthil

doi:10.1007/s10916-019-1416-0

Cited by 150 publications

(58 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Convolutional neural network models can automatically extract hierarchical features from the image data and achieve the end‐to‐end prediction without omissions and tediousness of many manual processes. To date, CNN‐based methods have been used to achieve excellent performance in radiotherapy workflows including automatic segmentation,29‐31 deformable registration,32,33 and synthetic computed tomography (CT) generation from magnetic resonance (MR) images 34…”

Section: Introductionmentioning

confidence: 99%

A method of using deep learning to predict three‐dimensional dose distributions for intensity‐modulated radiotherapy of rectal cancer

Zhou

Zhao

Song

et al. 2020

J Applied Clin Med Phys

View full text Add to dashboard Cite

Purpose To develop and test a three‐dimensional (3D) deep learning model for predicting 3D voxel‐wise dose distributions for intensity‐modulated radiotherapy (IMRT). Methods A total of 122 postoperative rectal cancer cases treated by IMRT were considered in the study, of which 100 cases were randomly selected as the training–validating set and the remaining as the testing set. A 3D deep learning model named 3D U‐Res‐Net_B was constructed to predict 3D dose distributions. Eight types of 3D matrices from CT images, contoured structures, and beam configurations were fed into the independent input channel, respectively, and the 3D matrix of dose distributions was taken as the output to train the 3D model. The obtained 3D model was used to predict new 3D dose distributions. The predicted accuracy was evaluated in two aspects: (a) The dice similarity coefficients (DSCs) of different isodose volumes, the average dose difference of all voxels within the body, and 3%/5 mm global gamma passing rates of organs at risks (OARs) and planned target volume (PTV) were used to address the spatial correspondence between predicted and clinical delivered 3D dose distributions; (b) The dosimetric index (DI) including homogeneity index, conformity index, V50, V45 for PTV and OARs between predicted and clinical truth were statistically analyzed with the paired‐samples t test. The model was also compared with 3D U‐Net and the same architecture model without beam configurations input (named as 3D U‐Res‐Net_O). Results The 3D U‐Res‐Net_B model predicted 3D dose distributions accurately. For the 22 testing cases, the average prediction bias ranged from −1.94% to 1.58%, and the overall mean absolute errors (MAEs) was 3.92 ± 4.16%; there was no statistically significant difference for nearly all DIs. The model had a DSCs value above 0.9 for most isodose volumes, and global 3D gamma passing rates varying from 0.81 to 0.90 for PTV and OARs, clearly outperforming 3D U‐Res‐Net_O and being slightly superior to 3D U‐Net. Conclusions This study developed a more general deep learning model by considering beam configurations input and achieved an accurate 3D voxel‐wise dose prediction for rectal cancer treated by IMRT, a potentially easier clinical implementation for more comprehensive automatic planning.

show abstract

Section: Introductionmentioning

confidence: 99%

A method of using deep learning to predict three‐dimensional dose distributions for intensity‐modulated radiotherapy of rectal cancer

Zhou

Zhao

Song

et al. 2020

J Applied Clin Med Phys

View full text Add to dashboard Cite

show abstract

“…This is both a black box process and a potential error source. Since the errors in the bone model can affect the registration accuracy, future investigation may be required to explore other segmentation techniques, e.g., the automatic segmentation using Convolutional Neural Networks (Thaha et al, 2019;Felfeliyan et. al, 2019).…”

Section: Discussionmentioning

confidence: 99%

An Automatic Icp-Based 2d-3d Registration Method for a High-Speed Biplanar Videoradiography Imaging System

Zhang

Lichti

Küpper

et al. 2020

Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci.

View full text Add to dashboard Cite

Abstract. High-Speed Biplanar Videoradiography (HSBV) is an X-ray based non-invasive imaging system that can be used to derive dynamic bony translations and rotations. The 2D-3D registration process matches a 3D bone model acquired from magnetic resonance imaging (MRI) or computed tomography (CT) scans with the 2D X-ray image pairs. This study focuses on the registration of MRI data as it can acquire detailed soft tissue contrast that cannot be easily discerned in CT scans. A novel 2D-3D registration method is reported in this paper that is suitable for the MRI-based bone models with high precision and high efficiency. In addition, an automatic initialization procedure with 64 starting poses is established to avoid user intervention in the registration. The method has been tested using the HSBV image sequence of a knee joint during walking. Thirty-five consecutive poses from the sequence were tested for the registration, and 50 non-consecutive poses randomly selected from the sequence were tested for the automatic initialization. The registration precision for each axis was 0.49 to 0.54 mm. For the initialization validation test, 48 over 50 frames were successfully initialized and two failed due to portions of the joint falling outside of the field-of-view of the system. The average time for each initialization is only about 6 min. The improved 2D-3D registration will allow determination of precise 3D kinematic parameters with high efficiency. These kinematic parameters can be used to calculate joint cartilage contact mechanics that provide insight into the mechanical processes and mechanisms of joint degeneration or pathology.

show abstract

“…The HGG network consists of 7 convolutional layers. Thaha et al [34] employed 3 × 3 kernels to build a deep architecture of an enhanced CNN model. To improve the performance, they performed intensity normalization and data augmentation in pre-processing the images.…”

Section: Related Workmentioning

confidence: 99%

Image classification-based brain tumour tissue segmentation

Alqazzaz

Sun

Yang

et al. 2020

Multimed Tools Appl

View full text Add to dashboard Cite

Brain tumour tissue segmentation is essential for clinical decision making. While manual segmentation is time consuming, tedious, and subjective, it is very challenging to develop automatic segmentation methods. Deep learning with convolutional neural network (CNN) architecture has consistently outperformed previous methods on such challenging tasks. However, the local dependencies of pixel classes cannot be fully reflected in the CNN models. In contrast, hand-crafted features such as histogram-based texture features provide robust feature descriptors of local pixel dependencies. In this paper, a classification-based method for automatic brain tumour tissue segmentation is proposed using combined CNNbased and hand-crafted features. The CIFAR network is modified to extract CNN-based features, and histogram-based texture features are fused to compensate the limitation in the CIFAR network. These features together with the pixel intensities of the original MRI images are sent to a decision tree for classifying the MRI image voxels into different types of tumour tissues. The method is evaluated on the BraTS 2017 dataset. Experiments show that the proposed method produces promising segmentation results.

show abstract

Brain Tumor Segmentation Using Convolutional Neural Networks in MRI Images

Cited by 150 publications

References 14 publications

A method of using deep learning to predict three‐dimensional dose distributions for intensity‐modulated radiotherapy of rectal cancer

A method of using deep learning to predict three‐dimensional dose distributions for intensity‐modulated radiotherapy of rectal cancer

An Automatic Icp-Based 2d-3d Registration Method for a High-Speed Biplanar Videoradiography Imaging System

Image classification-based brain tumour tissue segmentation

Contact Info

Product

Resources

About