DIFFERENTIAL INCIDENCE OF TYPE II DIABETES MELLITUS BETWEEN MALE AND FEMALE USING BETA-BINOMIAL MODEL (Research Paper Sample)

Anale. Seria Informatică. Vol. XVII fasc. 1 – 2019 Annals. Computer Science Series. 17th Tome 1st Fasc. – 2019 DIFFERENTIAL INCIDENCE OF TYPE II DIABETES MELLITUS BETWEEN MALE AND FEMALE USING BETA BINOMIAL MODEL Matthew Aina 1, Muhammed Besiru Jibrin 2, Usman Abubakar Jauro 3, Olukayode Aiyeniko 4 Department of Statistics, Faculty of Physical Science, Ahmadu Bello University, Zaria, Nigeria 2,3 Department of Computer Science, Federal University of Kashere, Gombe State, Nigeria Department of Computer Science, Faculty of Science, Lagos State University, Lagos State, Nigeria Corresponding Author: Matthew Aina, matthew1969@yahoo.com ABSTRACT: One of the life threaten diseases with a long duration and considerable associated morbidity is type II diabetes mellitus. This disease affects a large populations around the world regardless of economic status and its predominance is rising despite the struggles of clinicians, researchers and public health professionals. This paper intends to analysis gender sensitivity to type II diabetes mellitus. Secondary data used in this study was obtained for a period of 10 years from medical records of General Hospital, Ifaki-Ekiti, Ekiti state, Nigeria. A Beta-Binomial Model was applied by incorporating Bayesian Statistical Technique to determine the hypothesis testing, credible interval and posterior probability distribution of the incidence of type II diabetes mellitus between Male and Female patients. Experimental results revealed that male patients are more sensitive to type II diabetes mellitus, therefore, males are more affected with type II diabetes mellitus than female. KEYWORDS: Beta-Binomial, Bayesian, Diabetes Mellitus, Morbidity, Sensitivity nature. INTRODUCTION Diabetes mellitus of type II can be defined as a long term metabolic disorder that is characterized by hyperglycemia (high blood sugar), disorders of carbohydrates, protein and fat metabolism ([B+13a]). The symptoms usually include increased thirst or hunger, recurrent urination and inexplicable loss of weight, feeling tired and sores that do not heal. The long-term complications from high blood sugar include heart disease, strokes, peripheral neuropathies, blindness, kidney failure, poor blood flow in the limbs which may lead to amputations. and renal disease ([CHF14]), The type-II diabetes mellitus has a very strong genetic component which is more greater than type I ([M+02]). There are no symptoms with is usually diagnosed by tests that indicate glucose intolerance and is treated with changes in diet and an exercise regimen. The cases of diabetes worldwide have increased significantly in the last 3 decades ([B+13a]). It has been identified to be the fifth leading cause of death in the society nowadays ([W+04]). The prediction of studies revealed that in the next twelve years, type II diabetes mellitus will be the seventh leading cause of death in the society ([D+15]). In the developing nations, one in twenty deaths of adults is diabetes mellitus related [D+05]. It was observed that in the world, the region of African experience the incidence rate of diabetes mellitus at 3.8% a total of 13.1 million people [B+13b]. The problem of diabetes is increasing rapidly all over the globe causing an enormous socioeconomic and health challenge, researchers stated that the number of people with diabetes is expected to further increase from that of 415 million in 2015 to 642 million by 2040 [D+18]. There are several risk factors that lead to type II diabetes mellitus, these include unhealthy diet, physical inactivity, increasing age, high blood pressure, impaired glucose tolerance and poor nutrition during pregnancy. Type II diabetes mellitus makes up about 90% of cases of diabetes, with the other 10% due primarily to type I diabetes and gestational diabetes. Globally, there are major intervention sets to prevent incidence of type II diabetes mellitus. The interventions include dietary intervention or physical activity intervention or both. It has been discovered that combining dietary intervention and physical activity intervention is considered to be effective and efficient method for diabetes management. Several studies have mentioned the impact of sexual role in diabetes mellitus in the general public, they have applied different approaches to carry out experimental and statistical analysis to investigate the effect of gender in chronic disease such as type diabetes mellitus. This paper considered majorly the influence of gender on type II diabetes mellitus using beta binomial model to know the gender sensitivity to type 2 diabetes mellitus. 2. RELATED WORK ([D+18]) proposed a framework theoretically demonstrating the link between environment and type II diabetes mellitus. The study briefly 214 Anale. Seria Informatică. Vol. XVII fasc. 1 – 2019 Annals. Computer Science Series. 17th Tome 1st Fasc. – 2019 mentioned methodological challenges, potential solutions and future research chances. Walkability, air pollution, food, physical activity environment and roadways proximity were the most common environmental characteristics studied. Of the more than 200 reported and extracted relationships assessed in 60 studies, 82 showed significant association in the expected direction. In general, higher levels of walkability and green space were associated with lower type II diabetes mellitus, while increased levels of noise and air pollution were associated with greater risk. Current evidence is limited in terms of volume and study quality prohibiting causal inferences. However, the proof put forward that environmental characteristics may influence type II diabetes mellitus occurrence and also give a tangible basis for further investigation with better quality data and longitudinal studies with policy-relevant environmental measures. [Nij17] applied negative binomial distribution to define the null hypothesis, to show the fraction of successes in each performed experiment and also the number of failure. The experiment showed that a good approximation for the negative binomial distribution is Poisson distribution, and it also established that the negative Binomial could be used to calculate the confidence interval percentage in based on the generalized hypergeometric function. The study recommended the areas of application of negative binomial such as health care, quality control, science and social science. It was further stressed that Negative Binomial may be applied in Aging research in clinical epidemiology, thunderstorm probability of occurrence in meteorology, stock control problems and accident statistics are good examples among other applications of negative Binomial distribution. [ANH16] applied Bayesian Mixture Model Averaging (BMMA), which is a Bayesian solution to form the best single model from all the possible mixture models. BMMA. The modeling of BMM was used for Diabetes Mellitus data, while Bayesian Model Averaging was used for microarray data. The BMMA was applied to model blood sugar levels of Diabetes Mellitus patient through simulation studies. The experimental results of this study have succeeded in developing the BMMA normal model with two components mixture that could accommodate the real condition of the diabetes mellitus data with driven data concept. The model is capable of accommodating the real condition of Diabetes mellitus driven data concept in the future. [B+15] employed Bayesian Logistic Regression Model (BLM) to model the risk factors associated with Diabetes Mellitus. The study compared the Bayesian logistic regression model with those from the Frequentist Logistic Regression Model (FLR) based on the significant factors. The results showed that BLM with non – informative flat prior distribution and Frequentist logistic regression model yielded similar results while non-informative flat model shared a different result compared to the model. The study suggested that the use of non-informative did not perfectly give improved model than the maximum likelihood estimate (MLE). [JPJ14] applied negative regression to examine data on intravenous catheter (iv) insertion on sample of children. The study determined the fitness and applied negative binomial regression to examine data in southern stern U.S hospital. The assumptions governing negative binomial model was examined, that data may violate the assumption independence. The result showed that assumption of negative binominal is a good fit for modeling the (iv) insertion process data, with the AIC of the negative binomial regression model indicating an improved fit than the OLS regression model. The result revealed that negative binomial regression model appeared to give more precise effects. [B+13a] presented a gender differential in the incidence of Diabetes mellitus among the patients at Udi village in Enugu State, Nigeria. The study formulated research questions and hypothesis, the method of descriptive survey design was applied. A sample of 343 diabetes mellitus patients were randomly selected and used for the study. Data documentary profoma was employed to collect the data from selected health facilities in respect to the patients’ gender and their fasting sugar level as contained in their folder or hospital record/register. Frequency and percentages was used to answer research question while the null hypothesis was tested using Chi- square at 0.05 level of significance. The findings of the result showed a high incidence of diabetes mellitus in 2012, females recorded a high incidence of diabetes mellitus from 2008-2012 and high percentage of fasting blood sugar level than male. The findings also revealed no difference in the fasting blood sugar based on gender. 3. METHODOLOGY In this proposed study, Beta Binomial model was used to carry out the analysis. This method shows how to update knowledge by applying Bayesian inference techniques. Beta binomial is a joint distribution which comprises beta as con...