Pathological Brain Detection using Extreme Learning Machine Trained with Improved Whale Optimization Algorithm

Pathological brain detection has made remarkable progress, thus numerous successful automated systems have been conceived. However, the diagnostic accuracy obtained by them is far from perfection. This paper investigates an automated system through MR images that notably enhance the current outcomes. The proposed PBDS employs fast discrete curvelet transform via wrapping (FDCT- WR) strategy to derive significant features from the MR brain images. Then, it adopts a combined approach called PCA + LDA to yield more discriminative and reduced features sets. Lastly, for classification, we introduce a novel improved learning method dubbed as IWOA-ELM which is a combination of two algorithms such as improved whale optimization algorithm (IWOA) and extreme learning machine (ELM). IWOA in the proposed method helps in optimizing the hidden node parameters, while an analytical procedure is adopted for computation of the output weights. To find the global optima using IWOA, we consider both the norm of the output weights and root mean squared error (RMSE). The proposed system is rigorously evaluated on three publicly available datasets and a comparative analysis has been made with the existing schemes. The simulation results demonstrate that the suggested approach outperforms other state-of-the-art approaches. Furthermore, it has also been noticed that IWOA-ELM method yields superior performance than conventional learning algorithms.