چکیده:
Introduction: The differential diagnosis of Erythemato-Squamous disease is a very important issue in dermatology. These diseases include dermatitis seborrhea, lichen planus, Pityriasis rosea, Chronic
Dermatiticand Pityriasis rubra pilaris. The differential diagnosis of psoriasis need to spend a lot of time and costs and incorrect or delayed diagnosis will lead to risky consequences for the patient. In this regard, a fuzzy expert system was created to diagnose psoriasis at any time and in any place.
Methods: In this study, an expert system is created to diagnose psoriasis based on fuzzy logic and rulebased database using MATLAB software. Also to test system the data from 190 patients with psoriasis was used among which 126 people with psoriasis were selected.
Results: The accuracy, sensitivity and accuracy of the system, was obtained 96%, 97% and 95%,
respectively. The area under the ROC curve was obtained 0.97 and that of Kappa test was obtained 0.92 (P < 0.001).
Conclusion: Given the importance of early diagnosis of psoriasis, the use of expert systems can play an important role in preventing disease progression, reducing severe pain and costs.
خلاصه ماشینی:
"In addition, clinical symptoms with scales of very low, low, medium, high and very high were described Data collection method: In order to obtain medical knowledge and identifying the important factors affecting the diagnosis of psoriasis in the step of design and development of system, four dermatologists were interviewed, while the check list was prepared for evaluating fuzzy expert system in psoriasis diagnosis.
Table 1 - Value of experts in diagnostic parameters diagnosis parameter The average value (0 to 4) Redness 4 scaling 4 Itching 2 Lesion with distinct borders 3 Worsening the injury by 4 hitting Third step: multifaceted lesion 0 oral involvement 0 knee and elbow involvement 3 scalp involvement 3 lesions around hair follicles 1 Data analysis: To assess the extent of compliance of the recognition systemThe purpose of this step was to ensure the completeness and accuracy of the system.
Table 2 – Table of created Expert System Decision-making redness scaling lesion with distinct borders itching worsening of injury by being hit multiple lesions lesions around hair follicles oral mucosal involvement involvement of knees and elbows involvement of scalp Psoriasis 4 4 3 2 4 0 1 0 3 3 Seborrhea dermatitis 2 2 1 3 0 0 1 0 0 3 lichen planus 1 0 3 3 3 3 2 4 2 2 Pityriasis Rosea 3 3 2 1 0 0 0 0 2 1 Chronic Dermatitic 3 1 2 3 0 0 0 0 2 2 PRP 3 2 2 2 0 0 2 0 2 0 Semantic system Network of Psoriasis Recognition Semantic networks are a graphical notation to represent knowledge that are a set of Node and bows."