Advances in Life Science and Technology www.iiste.org ISSN 2224-7181 (Paper) ISSN 2225-062X (Online) Vol.29, 2015 71 Prediction of Severity of Diabetes Mellitus using Fuzzy Cognitive Maps Nitin Bhatia , Sangeet Kumar DAV College, Jalandhar , Punjab, India *E-mail of the corresponding author: sangeetkumararora@yahoo.com Abstract The objective to develop this research paper is concerned with a system which helps diagnose the severity of diabetes. The disease named diabetes mellitus makes the body unable to handle sugar so it causes thirst, frequency of urination, tiredness and many other symptoms. The diabetes mellitus describes a metabolic disorder characterized by chronic hyperglycemia with disturbances of carbohydrate, fat and protein metabolism resulting from defects in insulin secretion, insulin action, or both. It can be caused by number of factors like pancreatic dysfunction, obesity, hereditary, stress, drugs, alcohol etc. It includes long term damage, dysfunction and failure of various organs. The effects of diabetes mellitus include long term damage and failure of various organs. Diabetes mellitus may present with characteristic symptoms such as thirst, polyuria, blurring of vision, and weight loss. This Paper is implemented on soft computing technique, namely Fuzzy Cognitive Maps (FCM) to find out the presence or absence of diabetes mellitus based on the input of sign/symptoms recorded at three fuzzy levels developed by the domain experts. The large amount of data and information that needs to be handled and integrated requires specific methodologies and tools. The FCM based decision support system was developed with a view to help medical and nursing personnel to assess patient status assist in making a diagnosis. The software tool was tested on 50 cases, showing results with an accuracy of 96%. The analysis of experimental results of different applicants checks the correctness and consistency of decision Support system for correct decision making. Keywords: Fuzzy Logic, FCM, Diabetes Mellitus, Prediction, Symptoms. 1 Introduction FCM have been employed to model knowledge-based systems, the most common type of artificial intelligence in medicine systems in routine clinical use for diagnosing thyroid diagnosis. They contain medical knowledge, usually about a very specifically defined task, and are able to reason with data from individual patients to come up with reasoned conclusions (Papageorgiou et al. 2008b). The computer-based model is used for differential diagnosis of specific language impairment (SLI), a language disorder that, in many cases, cannot be easily diagnosed. This difficulty necessitates the development of a methodology to assist the speech therapist in the diagnostic process. The methodology tool is based on fuzzy cognitive maps and constitutes a qualitative and quantitative computer model comprised of the experience and knowledge of specialists. The development of the model was based on knowledge from the literature and then it was successfully tested on four clinical cases (Georgopoulos et al. 2003). In the medical application area, FCMs have been used to model and analyze temporal medical data. In this proposed framework ,the division and merging of concepts depending on the time scale used for learning and learning the FCM with the use of a virtual time scale with logical time slots of variable real-time length ,the two new propositions to the adaptive algorithm used for learning FCMs are introduced. The experimental results presented by Froelich & Wakulicz-Deja (2009) show the impact of the introduced enhancements on the process of learning of FCM .A new framework for the construction of augmented Fuzzy Cognitive Maps based on Fuzzy Rule-Extraction methods for decisions in medical informatics is investigated. Fuzzy cognitive maps are knowledge-based techniques which combine elements of fuzzy logic and neural networks and work as artificial cognitive networks. The knowledge extraction methods used in this study extract the available knowledge from data in the form of fuzzy rules and insert them into the FCM, contributing to the development of a dynamic decision support system (Papageorgiou 2011). The idea of combining decision tress with FCMs was explored in order to maintain the potential advantages of both techniques. The new integrated system has been introduced for medical decision making process. This work proposes a new framework of Fuzzy Cognitive Map utilizing Decision Trees that updates the traditional Fuzzy Cognitive Map and has better characteristics. The inclusion of decision tree generators in the structure of the FCM, and the new DTFCM system gives better results. The performance of the new methodology can deal with different kind of input data eliminating numerical errors (Papageorgiou et al. 2006). The fuzzy cognitive map (FCM) is an efficient technique for characterization of Brain tumor. Brain tumors are considered as one of the most lethal and difficult to identify and be treated forms of cancer (DeAngelis 2001). Pathologists evaluate the aggressiveness of brain tumors by visually examining tissue section (biopsies) based on guidelines determined by the World Health Organization (WHO) (Kleihues 1993). A new method for characterizing brain tumors is presented (Papageorgiou et al. 2008a), which models the human thinking approach and the classification results