作者: Yudong Zhang , Shuihua Wang , Zhengchao Dong , Preetha Phillip , Genlin Ji
DOI: 10.2528/PIER15040602
关键词: Cross-validation 、 CAD 、 Wavelet entropy 、 Artificial intelligence 、 Particle swarm optimization 、 Pattern recognition 、 Artificial neural network 、 Machine learning 、 Computer science 、 Offline learning 、 Biogeography-based optimization 、 Magnetic resonance imaging
摘要: Background) We proposed a novel computer-aided diagnosis (CAD) system based on the hybridization of biogeography-based optimization (BBO) and particle swarm (PSO), with goal detecting pathological brains in MRI scanning. (Method) The method used wavelet entropy (WE) to extract features from MR brain images, followed by feed-forward neural network (FNN) training Hybridization BBO PSO (HBP), which combined exploration ability exploitation PSO. (Results) 10 repetition k-fold cross validation result showed that HBP outperformed existing FNN methods WE + HBP-FNN fourteen state-of-the-art CAD systems classification terms accuracy. achieved accuracy 100%, 99.49% over Dataset-66, Dataset-160, Dataset-255, respectively. offline learning cost 208.2510 s for merely 0.053s online prediction. (Conclusion) achieves nearly perfect detection