Authors: Babita Pandey; R.B. Mishra
Addresses: School of Computing and Information Technology, Lovely Professional University, Jalandhar-Delhi G.T. Road (NH-1), Phagwara, Punjab, 144402, India ' Department of Computer Engineering, IIT-BHU, 221005, Varanasi, India
Abstract: In this paper, we have developed a intelligent model that performs two level diagnosis of neuromuscular diseases (NMDs) by using muscular, cognitive, psychological, internal lingual and EMG parameters. At first level, the NMDs are: muscular dystrophy, polymyositis, endocrine myopathy, metabolic myopathy, neuropathy, poliomyelitis and myasthenia gravis. These diseases are further classified in more specific diseases at second level. Data mining methods are used to obtain the important parameters/symptoms. A hierarchal rule base model is integrated with case base reasoning for calculating cumulative confidence factor (CCF) and confidence index. In addition to this a Bayesian network (BN) model is also developed for calculating the probability of occurrence of diseases and to make a comparative view of the results obtained by CCF and BN approaches.
Keywords: neuromuscular diseases; NMDs; cumulative confidence factor; CCF; confidence index; Bayesian networks; intelligent modelling; medical diagnosis; data mining; disease occurrences.
International Journal of Knowledge Engineering and Soft Data Paradigms, 2014 Vol.4 No.3, pp.199 - 226
Published online: 07 Aug 2014 *Full-text access for editors Access for subscribers Purchase this article Comment on this article