Action fuzzy rule based classifier for analysis of dermatology databases. (1st January 2014)

• Record Type: Journal Article
• Authors: Deepa, T.; Sathiyabhama, B.; Akilandeswari, J.; Gopalan, N.P.
• Is Part Of: International journal of biomedical engineering and technology. Volume 15:Number 4(2014)
• Page Start: 360
• Page End: 379
• Publication Date: 2014-01-01
• Subjects: action rule -- fuzzy rule -- clinical decision support system -- medical informatics -- knowledge management; Biomedical engineering -- Periodicals; 610.28
• Languages: English
• ISSNs: 1752-6418

Abstract

In healthcare industries, the classification systems are applied to localise primary tumours, prognosis and recurrence of breast cancer, diagnosis of thyroid diseases, and rheumatology. The major requirements for such a system are prediction accuracy and comprehensibility of the learned knowledge by human experts. Success of a classification learning algorithm, in terms of these criteria, is directly related to the scheme used for representing the learned classification knowledge. An Adaptive Neuro Fuzzy Inference System (ANFIS) mechanism is proposed for erythemato-squamous disease analysis and to manage the knowledge acquired and found from the system. In addition, this system utilises Action Rule based Diagnostic System (ARDS) action and fuzzy rule based classifiers in accordance with the severity levels of the diseases. Empirical results endorse that this system is an authentic and reliable tool to reduce the human errors and improve the quality of medical care provided to the public.