3.1 PROBLEM STATEMENT :-
Real time problem faced in medical domain are to be analyzed and diagnose properly, so treatment should be started in timely manner. Risk due to disease or any disorder to be diagnosed to avoid the severe stage/condition.
This thesis report defines the AI and AI Techniques used for medical disease diagnosis. This report involves how AI helps in medical diagnosis .Different computer based expert system, difficulties in medical diagnosis. Challenges to implement AI methods, Risk in medical disease covered by AI technique explained. The Different AI techniques /methods that are Neural network/Genetic algorithm/Fuzzy Logic architecture are described .Also the combination of all three …show more content…
The AI based Medical expert system, not meant to replace the doctor/specialist. The system assist specialist and general practitioner’s. in diagnosing and predicting the patient’s condition either normal or severe from experience or certain set of rules. whereas the patient with high risk symptoms/ factors or predicted to be highly effected with certain illness , are shortlisted to see the expert /specialist doctor for further treatment AI techniques in medical and biomedical reduce cost and time of patient needed for …show more content…
The Artificial neural network solves the problem/disease which are too complex for conventional technology. NN gives powerful tool to help the doctor to analyze ,model and make idea of complex clinical (patient) data across broad range of medical applications.
The main task is to categorized the patient problem on the basis of measured features to assign the patient to one of small set of classes. The Feed forward back propagation network used as Classifier to differentiate between infected or non infected person. The Back propagation ,RBF and LVQ algorithms used for NN. The hybrid structure ,like combination of GA-NN and FL-NN implemented .which makes the medical diagnosis expert system.
2.Methodology of Disease diagnosis using genetic algorithm.
The person health affected by Genetic features and environmental factors. Algorithm isolate the most relevant group of feature and then to class individual that have the considered disease according to these association. Operator to allow GA to explore the search space. GA plays crucial role in disease diagnosis .The dimension of data reduced using GA algorithm techniques. The supervised feature selection of disease used in GA (GA-BN Algorithm combined with SVM (Support vector machine)).The hybrid structure of GA-NN implemented for better performance