Abstract
This paper addresses issues of brain tumor subtype classification using Magnetic Resonance Images (MRIs) from different scanner modalities like T1 weighted, T1 weighted with contrast-enhanced, T2 weighted and FLAIR images. Currently most available glioma datasets are relatively moderate in size, and often accompanied with incomplete MRIs in different modalities. To tackle the commonly encountered problems of insufficiently large brain tumor datasets and incomplete modality of image for deep learning, we propose to add augmented brain MR images to enlarge the training dataset by employing a pairwise Generative Adversarial Network (GAN) model. The pairwise GAN is able to generate synthetic MRIs across different modalities. To achieve the patient-level diagnostic result, we propose a post-processing strategy to combine the slice-level glioma subtype classification results by majority voting. A two-stage course-to-fine training strategy is proposed to learn the glioma feature using GAN-augmented MRIs followed by real MRIs. To evaluate the effectiveness of the proposed scheme, experiments have been conducted on a brain tumor dataset for classifying glioma molecular subtypes: isocitrate dehydrogenase 1 (IDH1) mutation and IDH1 wild-type. Our results on the dataset have shown good performance (with test accuracy 88.82%). Comparisons with several state-of-the-art methods are also included.
Highlights
Gliomas are one of the most common tumors originating from the brain [1]
We propose a novel scheme to improve the performance of gliomas characterization and subtype classification based on the real and pairwise Generative Adversarial Network (GAN)-augmented MR images in multi-modality forms
The proposed scheme has been tested using real and pairwise GAN-augmented Magnetic Resonance Images (MRIs) as training data, results obtained on the testing dataset have demonstrated that the scheme is effective and robust
Summary
Gliomas are one of the most common tumors originating from the brain [1]. World Health Organization (WHO) grades gliomas into four classes (grades I-IV) according to their aggressiveness. The pairwise GANs differ from the conventional GANs in terms of its aim and the cost function It is employed mainly for generating synthetic images across different modalities of MRIs (e.g. from FLAIR to T2, or from T2 to T1) in medical images. This leads to a different cost function of the pairwise GAN where tumor areas are enhanced for generating synthetic MRIs. our choice of the generator and discriminator is similar to [21], it is worth noting that the pairwise GAN is used to train two pairs of generators and discriminators simultaneously with tumor mask added as the prior to focus on the tumor regions. Two generators G∗m and G∗n are used for synthesizing MRIs across two modalities
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
