Development of an Expert System as a Diagnostic Support of Cervical Cancer in Atypical Glandular Cells, Based on Fuzzy Logics and Image Interpretation

• Published in: Computational and mathematical methods in medicine Vol. 2013; no. 2013; pp. 1 - 17
• Main Authors: Domínguez Hernández, Karem R., Aguilar Lasserre, Alberto Alfonso, Posada Gómez, Rubén, Palet Guzmán, José A., González Sánchez, Blanca E.
• Format: Journal Article
• Language: English
• Published: Cairo, Egypt Hindawi Puplishing Corporation 01.01.2013; Hindawi Publishing Corporation
• Subjects: Cervical Intraepithelial Neoplasia - diagnosis; Computational Biology; Diagnosis, Computer-Assisted - methods; Diagnosis, Computer-Assisted - statistics & numerical data; Expert Systems; Female; Fuzzy Logic; Humans; Image Interpretation, Computer-Assisted - methods; Risk Factors; Uterine Cervical Dysplasia - diagnosis; Uterine Cervical Neoplasms - diagnosis; Vaginal Smears
• ISSN: 1748-670X; 1748-6718
• DOI: 10.1155/2013/796387

Cervical cancer is the second largest cause of death among women worldwide. Nowadays, this disease is preventable and curable at low cost and low risk when an accurate diagnosis is done in due time, since it is the neoplasm with the highest prevention potential. This work describes the development of an expert system able to provide a diagnosis to cervical neoplasia (CN) precursor injuries through the integration of fuzzy logics and image interpretation techniques. The key contribution of this research focuses on atypical cases, specifically on atypical glandular cells (AGC). The expert system consists of 3 phases: (1) risk diagnosis which consists of the interpretation of a patient’s clinical background and the risks for contracting CN according to specialists; (2) cytology images detection which consists of image interpretation (IM) and the Bethesda system for cytology interpretation, and (3) determination of cancer precursor injuries which consists of in retrieving the information from the prior phases and integrating the expert system by means of a fuzzy logics (FL) model. During the validation stage of the system, 21 already diagnosed cases were tested with a positive correlation in which 100% effectiveness was obtained. The main contribution of this work relies on the reduction of false positives and false negatives by providing a more accurate diagnosis for CN.