Enlarged Training Dataset by Pairwise GANs for Molecular-Based Brain Tumor Classification
Journal article, 2020
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.
molecular-based brain tumor subtype classification
glioma
MRI
deep learning
generative adversarial networks
multi-modal
data augmentation
Author
Chenjie Ge
Chalmers, Electrical Engineering, Signal Processing and Biomedical Engineering
Irene Yu-Hua Gu
Chalmers, Electrical Engineering
Asgeir Store Jakola
University of Gothenburg
Jie Yang
Shanghai Jiao Tong University
IEEE Access
2169-3536 (ISSN) 21693536 (eISSN)
Vol. 8 1 22560-22570 8970509Subject Categories
Other Computer and Information Science
Electrical Engineering, Electronic Engineering, Information Engineering
Computer Vision and Robotics (Autonomous Systems)
Medical Image Processing
Areas of Advance
Health Engineering
DOI
10.1109/ACCESS.2020.2969805