Authors: Sushruta Mishra; Hrudaya Kumar Tripathy; Brojo Kishore Mishra
Addresses: C.V. Raman College of Engineering, Bidyanagar, Mahura, Janla, Khordha, Odisha 752054, India ' School of Computer Engineering, KIIT University, Bhubaneswar, India ' Department of IT, C.V. Raman College of Engineering, Bhubaneswar, India
Abstract: Tumour prediction and classification is regarded as a complex task that needs attention. Moreover, medical experts lack expertise in this section. Hence, an intelligent clinical system model is the time of the hour. Recently, biologically motivated techniques are emerging to be an efficient computing method to solve imprecise and complex problems. Nature forms an immense source of motivation in finding solutions to sophisticated problems IT sector since it is highly robust and dynamic. The result obtained is highly optimised and balanced solution. This is the basic idea of such nature motivated techniques. In our research, we have analysed and implemented some important bio-inspired optimisation techniques to categorise different kinds of tumour. Multilayer perceptron is the classifier used in the process. We have later evaluated our results with some critical metrics like RMSE, Kappa coefficient, accuracy and many others to determine the effectiveness of our system model developed. It is observed that using bio-inspired computation approach enhances the efficiency of tumour classification. The results are depicted in this paper.
Keywords: bio-inspired computation; PSO search; genetic search; GS; evolutionary search; Kappa coefficient; classification accuracy.
International Journal of Computer Aided Engineering and Technology, 2018 Vol.10 No.3, pp.244 - 256
Received: 18 Apr 2016
Accepted: 11 Aug 2016
Published online: 09 Mar 2018 *