Deep semi-supervised learning for brain tumor classification
Autor: | Jie Yang, Irene Yu-Hua Gu, Asgeir Store Jakola, Chenjie Ge |
---|---|
Jazyk: | angličtina |
Rok vydání: | 2020 |
Předmět: |
lcsh:Medical technology
Databases Factual Computer science Brain tumor Semi-supervised learning Overfitting 030218 nuclear medicine & medical imaging 03 medical and health sciences 0302 clinical medicine Neuroimaging Glioma Molecular-based brain tumor classification medicine Humans Radiology Nuclear Medicine and imaging Grading (tumors) Brain Neoplasms business.industry Deep learning Pattern recognition medicine.disease Isocitrate Dehydrogenase Grading Technical Advance lcsh:R855-855.5 030220 oncology & carcinogenesis Mutation Radiographic Image Interpretation Computer-Assisted Neural Networks Computer Supervised Machine Learning Artificial intelligence Neoplasm Grading business Classifier (UML) MRI |
Zdroj: | BMC Medical Imaging, Vol 20, Iss 1, Pp 1-11 (2020) BMC Medical Imaging |
ISSN: | 1471-2342 |
Popis: | Background This paper addresses issues of brain tumor, glioma, classification from four modalities of Magnetic Resonance Image (MRI) scans (i.e., T1 weighted MRI, T1 weighted MRI with contrast-enhanced, T2 weighted MRI and FLAIR). Currently, many available glioma datasets often contain some unlabeled brain scans, and many datasets are moderate in size. Methods We propose to exploit deep semi-supervised learning to make full use of the unlabeled data. Deep CNN features were incorporated into a new graph-based semi-supervised learning framework for learning the labels of the unlabeled data, where a new 3D-2D consistent constraint is added to make consistent classifications for the 2D slices from the same 3D brain scan. A deep-learning classifier is then trained to classify different glioma types using both labeled and unlabeled data with estimated labels. To alleviate the overfitting caused by moderate-size datasets, synthetic MRIs generated by Generative Adversarial Networks (GANs) are added in the training of CNNs. Results The proposed scheme has been tested on two glioma datasets, TCGA dataset for IDH-mutation prediction (molecular-based glioma subtype classification) and MICCAI dataset for glioma grading. Our results have shown good performance (with test accuracies 86.53% on TCGA dataset and 90.70% on MICCAI dataset). Conclusions The proposed scheme is effective for glioma IDH-mutation prediction and glioma grading, and its performance is comparable to the state-of-the-art. |
Databáze: | OpenAIRE |
Externí odkaz: |