抽象的な

Computational Tools for Human Papillomavirus (HPV) Risk Prediction by Fuzzy Logic

Ilka Kassandra Pereira Belfort*, Sally Cristina Moutinho Monteiro, Allan Kardec Duailibe Barros

Introduction: Human Papillomavirus (HPV) is one of the most common sexually transmitted infections (STIs) and responsible for approximately 99% of cervical cancers in the world. Thus, the objective of this work was to develop a computational tool for HPV risk prediction by fuzzy logic.

Material and Methods: This involves the development of a computational model using fuzzy logic tools to predict women with a greater predisposition to exposure and infection by HPV. The health, lifestyle, and sexual data of adult women were collected using a semistructured questionnaire, oncotic cytology occurred from the analysis of the vaginal smear, and DNA-HPV research was carried out through the chain reaction of the Nested polymerase (PGMY09/11 (first-round PCR) and GP + 5/GP + 6 (second round PCR) using the Platinum ™ Taq DNA Polymerase system (Invitrogen ™, NY, USA). in the literature and later these were added to the results obtained from the analysis of the participants' data to construct the calculation with the determination of the risk.

Results: After the statistical analysis, the RISK set was concatenated into 3 sets (green, yellow, and red), the fuzzified data obtained as variables for availability in the risk calculator the following items: Green = [0-30%], low risk; Yellow = [31-50%], medium risk; RED = above 50%, high risk of HPV infection. os: 400 results of the epidemiological and cervical findings of the women participating in the research were used for training the software and 162 for system validation. After evidenced statistical data, the insertion of the results in the database started.

Conclusion: With the results obtained Fuzzy inference system can be as well adopted for the screening for HPV as this will in turn helps to reduce the mortality rate in cases with cancer. This expert system is user-friendly and carries out screening based on patients 'complain (clinical and laboratory data) to a medical expert.