Enlarged Training Dataset by Pairwise GANs for Molecular-Based Brain Tumor Classification
| Autor: | Jie Yang, Asgeir Store Jakola, Irene Yu-Hua Gu, Chenjie Ge |
|---|---|
| Jazyk: | angličtina |
| Rok vydání: | 2020 |
| Předmět: |
General Computer Science
Computer science Brain tumor multi-modal Fluid-attenuated inversion recovery Glioma glioma medicine T1 weighted General Materials Science Modality (human–computer interaction) medicine.diagnostic_test business.industry Deep learning General Engineering Magnetic resonance imaging Pattern recognition Molecular-based brain tumor subtype classification medicine.disease Feature (computer vision) Pairwise comparison Artificial intelligence lcsh:Electrical engineering. Electronics. Nuclear engineering generative adversarial networks business lcsh:TK1-9971 MRI data augmentation |
| Zdroj: | IEEE Access, Vol 8, Pp 22560-22570 (2020) |
| ISSN: | 2169-3536 |
| Popis: | 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. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|