International Journal on Recent and Innovation Trends in Computing and Communication ISSN: 2321-8169 Volume: 3 Issue: 12 6722 - 6727 ______________________________________________________________________________________ 6722 IJRITCC | December 2015, Available @ http://www.ijritcc.org _______________________________________________________________________________________ Implementing Neural Fuzzy Rough Set and Artificial Neural Network for Predicting PCOS Dr. K. Meena 1 , Dr. M. Manimekalai 2 1 Former Vice-President, Bharathidasan University 2 Director & Head, Department of Computer Application, Shrimati Indira Gandhi College 1,2 Trichy, Tamilnadu, India S. Rethinavalli Assistant Professor, Department of Computer Applications Shrimati Indira Gandhi College Trichy, Tamilnadu, India rethinavalli.17@gmail.com Abstract—Polycystic ovarian syndrome (or polycystic ovary syndrome – PCOS) is a multifarious form in which a woman‟s ovaries are normally larger than standard. The term „Polycystic‟ defines that the ovaries comprise of numerous cysts or follicles to facilitate hardly ever nurture towards ripeness or generate eggs accomplished of being fertilized. One third of women could contain polycystic ovaries observed on an ultrasound, however it does not all have PCOS. PCOS is comparatively universal, especially for sterile women. It concerns about 12 to 18 per cent of women of reproductive age (between late adolescence and menopause). In approximate 70 per cent of this kind of cases remain undiagnosed. In our previous researches, we have proposed a new feature selection technique and hybrid approach and in this present investigation, we implement these proposed algorithms to forecast the PCOS disease among women. In addition to above analysis, we evaluate the effect of the proposed algorithms with other existing methods. Keywords-PCOS, ANN, NFRS, J48, ID3 __________________________________________________*****_________________________________________________ I. INTRODUCTION Polycystic Ovarian Syndrome (PCOS) have an effect on 4 to 12% of women are from reproductive age [1]. In the 1935, Stein and Leventhal primarily depicted the connection of polycystic ovaries, amenorrhea, hirsutism, and obesity. The major characteristics are essential for the identification of PCOS were comprehensive at the conference convened by the National Institute of Health in 1990 moreover, they were menstrual dysfunction and hyperandrogenism, through segregation of other major reasons of hyperandrogenism (congenital adrenal hyperplasia, androgen-secreting tumors, and hyperprolactinemia). The probable criteria included for perimenarchal onset, insulin resistance, elevated leutenizing hormone to follicle-stimulating hormone ratio and polycystic ovaries by ultrasonography (USG) [1]. PCOS was reclassified at a joint consensus meeting of the European Society of Human Reproduction and Embryology (ESHRE) and the American Society of Reproductive Medicine (ASRM), held in Rotterdam in may 2003. And it enclosed the existence of two of the following three major criterias: (a) oligo and/or anovulation, (b) polycystic ovaries on USG and (c) hyperandrogenism (clinical and/or biochemical), with the exclusion of other etiologies. The morphology of the polycystic ovary has been redefined as an ovary with 12 or more follicles measuring 2–9 mm in diameter and/or increased ovarian volume (more than 10 cm 3 ) [2]. Polycystic ovaries are frequently observed in healthy women however it is more common in women with irregular cycles and hyperandrogenism. The polycystic form of the ovary is the distinctive sign of polycystic ovary syndrome (PCOS) however there is a wide range of clinical and biochemical characterisitics like e.g. elevated serum concentrations of androgens, insulin, LH and decreased insulin sensitivity. These conditions are frequently associated with obesity. Since insulin resistance in PCOS patients is predominantly extra-splanchnic [3] (Dunaif et al., 1992), the fasting blood sugar is normal. According to another group [4] (Franks et al., 1997), ovarian morphology is the vital sign of the syndrome and the wide range of associated phenotypes can be elucidated through the interaction of a small quantity of key genes with ecological features. Since these symptoms are found in up to 10% of young women, PCOS is certainly the most frequent endocrine disorder diagnosed in these subjects. Despite the high prevalence of isolated polycystic ovarian morphology (22%), the syndrome may be accompanied by minimal clinical manifestations and, in particular, no uniformly deleterious effect on fertility has been reported [5] (Clayton et al., 1992). A controlled comparative study of patients undergoing an IVF programme found no significant difference in pregnancy and live birth rates between women with and without polycystic ovaries [6] (MacDougall et al., 1993). Nevertheless, in a large group of PCOS patients, a high prevalence of primary (46%) and secondary (26%) infertility was found [7] (Balen et al., 1995), while another group [8] (Regan et al., 1990) found an elevated rate of miscarriages in patients with raised LH concentrations in which PCOS was inferred. II. PCOS PHENOTYPES IN VARIOUS POPULATIONS In view of the high prevalence of affected individuals, a genetic cause of the syndrome was suggested 30 years ago [9]