A Novel Approach for Cardiac Disease Prediction and Classification Using Intelligent Agents

Intelligent agents are a new paradigm for developing software applications. More than this, agent-based computing has been hailed as "the next significant breakthrough in software development" (Sargent, 1992), and "the new revolution in software" (Ovum, 1994). Currently, agents are the focus of intense interest on the part of many subfields of computer science and artificial intelligence. Agents are being used in an increasingly wide variety of applications, ranging from comparatively small systems such as email filters to large, open, complex, mission critical systems such as air traffic control. At first sight, it may appear that such extremely different types of system can have little in common. And yet this is not the case: in both, the key abstraction used is that of an agent. First, an agent is a computer system situated in some environment, and that is capable of autonomous action in this environment in order to meet its design objectives. Autonomy is a difficult concept to pin down precisely, but we mean it simply in the sense • that the system should be able to act without the direct intervention of humans (or other agents), and should have control over its own actions and internal state. It may be helpful to draw an analogy between the notion • of autonomy with respect to agents and encapsulation • with respect to object oriented systems. An object encapsulates some state, and has some control over this state in that it can only be accessed or modified via the methods that the object provides. Agents encapsulate state in just the same way. However, we also think of agents as encapsulating behavior, in addition to state. An object does not encapsulate behavior: it has no control over the execution of methodsif an object x invokes a method m on an object y, then y has no control over whether m is executed or notit just is. In this sense, object y is not autonomous, as it has no control over its own actions. In contrast, we think of an agent as having exactly this kind of control over what actions it performs. Because of this distinction, we do not think of agents as invoking methods (actions) on agentsrather, we tend to think of them requesting actions to be performed. The decision about whether to act upon the request lies with the recipient. . An intelligent agent is a computer system that is capable of flexible autonomous action in order to meet its design objectives. By flexible, we mean that the system must be: responsive: agents should perceive their environment (which may be the physical world, a user, a collection of agents, the Internet, etc.) and respond in a timely fashion to changes that occur in it. proactive: agents should not simply act in response to their environment, they should be able to exhibit opportunistic, goaldirected behavior and take the initiative where appropriate, and Applications of Intelligent Agents social: agents should be able to interact, when they deem appropriate, with other artificial agents and humans in order to complete their own problem solving and to help others with their activities. Hereafter, when we use the term "agent", it should be understood that we are using it as an abbreviation for "intelligent agent". Other researchers emphasize different aspects of agency (including, for example, mobility or adaptability). Naturally, some agents may have additional characteristics, and for certain types of applications, some attributes will be more important than others. However, we believe that it is the presence of all four attributes in a single software entity that provides the power of the agent paradigm and which distinguishes agent systems from related software paradigmssuch as object-oriented systems, distributed sysems, and expert systems (see Wooldridge (1997) for a more detailed discussion). By an agent-based system, we mean one in which the key abstraction used is that of an agent. In principle, an agent-based system might be conceptualized in terms of agents, but implemented without any software structures corresponding to agents at all. We can again draw a parallel with object-oriented software, where it is entirely possible to design a system in terms of objects, but to implement it without the use of an object-oriented software environment. But this would at best be unusual, and at worst, counterproductive. A similar situation exists with agent technology; we therefore expect an agent-based system to be both designed and implemented in terms of agents. A number of software tools exist that allow a user to implement software systems as agents, and as societies of cooperating agents. Note that an agentbased system may contain any non-zero number of agents. The multi-agent casewhere a system is designed and implemented as several interacting agents, is both more general and significantly more complex than the single-agent case. However, there are a number of situations where the single-agent case is appropriate. Traditional diagnosis in TCM requires long experiences and a high level of skill, and is subjective and deficient in quantitative diagnostic criteria. This seriously affects the reliability and repeatability of diagnosis and limits the popularization of TCM. So the focal problem that needs to be solved urgently is to construct methods or models to quantify the diagnosis in TCM. [1] Recently, a few researchers developed some methods and systems to modernize TCM. But most of them are built incorporating totally or partially rulebased reasoning model, which are lack of the feasibility of implementing all possible inference by chaining rules and limits their practical applications in clinical medicines.An attraction tool for managing various forms of uncertainty is Bayesian networks (BNs) [2], [3] which is able to represent knowledge with uncertainty and efficiently performing reasoning tasks.Naive Bayesian classifier (NBC) is a simplified form of BNs that assumes independence of the observations. Some research results [4], [5], [6] have demonstrated that the predictive performance of NBC can be competitive with more complicated classifiers.In this study, a novel computerized diagnostic model based on naive Bayesian classifier (NBC) is proposed. Firstly, a Bayesian network structure is learned from a database of cases [7] to find the symptom set that are dependent on the disease directly.Secondly, the symptom set is utilized as attributes of NBC and the mapping relationships between the symptom set and the disease are constructed.To reduce the dimensionality and improve the prediction accuracy of diagnostic model, symptom selection is requisite.Many feature selection methods, such as filters [8] and wrappers [9], have developed. But the dependency relationships among symptoms and the mapping relationships between symptom and syndrome are not considered in these methods, which are important to diagnosis in TCM. To lower the influences of irrelative symptoms, the mutual information between each symptom and disease is computed based on information entropy theory [10], which is utilized to assess the significance of symptoms.The paper [11] presents a multiagent system for supporting physicians in performing clinical studies in real time. The multiagent system is specialized in the controlling of patients with respect to their appointment behavior. Novel types of agents are designed to play a special role as representatives for humans in the environment of clinical studies. OnkoNet mobile agents have been used successfully for patient-centric medical problems solving [12].,emerged from a project covering all relevant issues, from empirical process studies in cancer diagnosis/therapy, down to system implementation and validation. In the paper [13], a medical diagnosis multiagent system that is organized according to the principles of swarm intelligence is proposed. It consists of a large number of agents that interact with each other by simple indirect communication. In the paper [14], a system called Feline composed of five autonomous agents (expert systems with some proprieties of the agents) endowed with medical knowledge is proposed. These agents cooperate to identify the causes of anemia at cats. The paper [39], also presents a development methodology for cooperating expert systems.In the paper [15], a Web-centric extension to a previously developed expert system specialized in the glaucoma diagnosis is proposed. The proposed telehealth solution publishes services of the developed Glaucoma Expert System on the World Wide Web.Each agent member of the CMDS system has problems solving capability and capacity (the notions are defined in [16,17]). The capacity of an agent Agf (Agf U MDUAS)consists in the amount of problems that can be solved by the agent, using the existent problem solving resources.The cooperative medical diagnosis problems solving by the diagnosis system is partially based on the blackboard-based problem solving [18,19]. The problem solving by the BMDS system is similar with the situations, when more physicians with different medical specializations plans a treatment to cure an illness that is in an advanced stage. Treatments known to be effective for the curing of the illness in a less advanced stage cannot be applied. From the entire discussions one can comprehend and classify the medical agent-based IDSS research [20] into two categories, namely Clinical Management and Clinical Research. Clinical Management envelops all clinical systems that are designed to help the doctor with diagnosing and deciding on treatment for medical conditions. Clinical Research on the erstwhile envelopes systems that are used to research facts and connections in attempt to detect new trends and patterns; it covers systems for both diagnosing patients and treating them. Feature selection, as a preprocessing step to machine learning, is effective in reducing dimensionality, removing irrelevant data, increasing learning accuracy, and improving result comprehensibility. In this work, we introduce a novel concept, predominant correlation, and propose a faster method which can identify relevant features as well as redundancy among relevant features without pair wise correlation analysis. The efficiency and effectiveness of our method is demonstrated through extensive comparisons with other methods using real-world data of high dimensionality. In Wrapper based feature selection, the more states that are visited during the search phase of the algorithm the greater the likelihood of finding a feature subset that has a high internal accuracy while generalizing poorly. It removes the irrelevant attributes that are below the threshold value. The classifier agent uses the naïve Bayesian classification algorithm. Bayesian network algorithm is used to classify the collected attributes in to five classes (0-normal,1-starting,2-low,3mild,4serious). The mutual information between each symptom and disease is computed based on information entropy theory. F and C are symptom and disease. f, c are events of F and C. I(F,C)= P(f, c)logP(f,c)/P(f)P(c) Suppose I0 is the prior entropy of C. I0 = (P(c=1)logp(c=1)+p(c=0)logp( c=0)) The significance of each symptom is calculated by S(F,C)=I(F,C)/I0. All the symptoms are evaluated and ranked by significance index S(F,C). Diagnosis results will be accurate in all conditions and then intelligent based decision support system is used to retrieve the patient information from the database.