Pathological brain detection based on AlexNet and transfer learning

• Published in: Journal of computational science Vol. 30; pp. 41 - 47
• Main Authors: Lu, Siyuan, Lu, Zhihai, Zhang, Yu-Dong
• Format: Journal Article
• Language: English
• Published: Elsevier B.V 01.01.2019
• Subjects: AlexNet; Computer aided diagnosis; Deep learning; Magnetic resonance image; Transfer learning; Deep learning; AlexNet; Computer aided diagnosis; Transfer learning; Magnetic resonance image
• ISSN: 1877-7503; 1877-7511
• DOI: 10.1016/j.jocs.2018.11.008

•We proposed a novel pathological brain detection based on computer vision and deep learning.•AlexNet was employed and served as the feature extractor.•Transfer learning was used to train a part of the network and it converged faster.•The proposed method achieved 100% classification accuracy which outperformed state-of-the-arts. The aim of this study is to automatically detect pathological brain in magnetic resonance images (MRI) based on deep learning structure and transfer learning. Deep learning is now the hottest topic both in academia and industry. However, the volume of brain MRI datasets are usually too small to train the entire deep learning structure. The training can be easily trapped into overfitting. Therefore, we introduced transfer learning to train the deep neural network. Firstly, we obtained the pre-trained AlexNet structure. Then, we replaced parameters of the last three layers with random weights and the rest parameters served as the initial values. Finally, we trained the modified model with our MRI dataset. Experiment results suggested that our method achieved accuracy of 100.00%, which outperformed state-of-the-art approaches.