Diagnosis of cerebral microbleed via VGG and extreme learning machine trained by Gaussian map bat algorithm

• DOI: 10.1007/s12652-020-01789-3
• Publication Date: 2020-02-24T05:16:37Z
• Authors: Siyuan Lu; Kaijian Xia; Shui-Hua Wang
• Keywords: Extreme Learning Machine

Cerebral microbleed (CMB) is a serious public health concern. It associated with dementia, which can be detected brain magnetic resonance image (MRI). CMBs often appear as tiny round dots on MRIs, and they spotted anywhere over brain. Therefore, manual inspection tedious lengthy, the results are short in reproducible. In this paper, novel automatic CMB diagnosis method was proposed based deep learning optimization algorithms, used MRI input output non-CMB. Firstly, sliding window processing employed to generate dataset from MRIs. Then, pre-trained VGG obtain features dataset. Finally, an ELM trained by Gaussian-map bat algorithm (GBA) for identification. Results showed that VGG-ELM-GBA provided better generalization performance than several state-of-the-art approaches.