چكيده به لاتين
In this thesis, we provide a new method based on a combination of high-order fuzzy cognitive map (HFCM) and support vector machine to classify celiac disease (CD). CD is a complex disorder whose development is affected by genetics (HLA alleles) and gluten ingestion. The celiac patients who are not treated are at a high risk of cancer, malignant lymphoma, and small-bowel neoplasia. Therefore, CD diagnosis and grading is of paramount importance. To improve the efficiency and increase the capability of HFCM classification, we use particle swarm optimization (PSO) algorithm. This approach, which models the human thinking process, uses the most recent method of CD grading, including the A, B1 and B2 grades. In order to evaluate the performance, this method is applied to 89 patients. The simulation results show the superiority of the proposed method compared with rule-based Bayesian networks. Given that extended models of PSO enhance its efficiency in terms of rate of convergence, global optimality, the solution accuracy and reliability of the algorithm, in this paper, we use algorithms such as adaptive PSO (APSO) and chaotic PSO (CPSO). APSO shows better search efficiency than classical PSO and can do a global search on all search space with more rate of convergence. Also CPSO methods maintain particles' population diversity in classical PSO to prevent premature convergence. The results of applying different CPSOs and APSO are compared with classical PSO. The best results in this case, which are achieved by applying the CPSO, are 91.67%, 90.91% and 93.75% for the A, B1 and B2 grades, respectively. Therefore, the highest grading precisions are achieved by using the combination of fourth order learned HFCM by CPSO and least squares SVM.
Keywords: Celiac disease, fuzzy cognitive maps, bayesian networks, grading
