Bayesian Artificial Neural Network (ANN) Model Approach to AIDS Associated Illness

AIDS associated illness (Stroke) is a major health problem and its impact is likely to increase in the future due to ongoing demographic transitional changes including ageing of population and health index observed worldwide (Baker RD et al: IMA J Manag Sci 12:1–17, 2001). Data from the global burden of disease (GBD) 2015 study showed that although age standardized rates of stroke mortality have decreased globally over the past two decades, the absolute number of people affected by stroke worldwide increased significantly between 1990 and 2015 (Craig PS, Goldstein M, Rougier JC, Seheul AH :J Am Stat Assoc 96:717–729, 2001; Chang LC, Lander HS, Lue MT: IEEE Trans Reliab 43(Suppl 3):457–469, 1994). Virtually no country in the world has seen a reduction in stroke burden in terms of absolute number of incidents and fatal strokes. The literature showed that various confounders viz age, gender, ethnicity and heredity have been identified as markers of risk of stroke (Baker RD et al: IMA J Manag Sci 12:1–17, 2001; Chang LC, Lander HS, Lue MT: IEEE Trans Reliab 43(Suppl 3):457–469, 1994; Chaurasia S, Chakrabarti P, Chourasia N: Int J Comput Appl 59(Suppl 3):6–10, 2012; Chevret S, Roguin H, Ganne P, Lefrere JJ: AIDS 6:1349–1352, 1992; Chu H, Gange SJ, Yamashita TE, Hoover DR, Chmiel JS, Margolick JB, Jacobson LP: Am J Epidemiol 162:787, 2005; Craig PS, Goldstein M, Rougier JC, Seheul AH: J Am Stat Assoc 96:717–729, 2001; Currin C, Mitchell TJ, Morris M, Ylvisaker D: J Am Stat Assoc 86:953–963, 1991; Diebolt J, Robert C: Bayesian estimation of finite mixture distributions, part I: theoretical aspects. Technical Report 110, LSTA, University Paris VI, Paris, 1990; Diebolt J, Robert C: J Royal Stati Soc Ser B 56:363–375, 1994; Doguc O, Emmanuel Ramirez-Marquez J: J Integr Des Process Sci 13(Suppl 1):33–48, 2009; Dragic T, Litwin V, Allaway GP, Martin SR, Huang Y, Nagashima KA, Cayanan C, Maddon PJ, Koup R, Moor JP, Paxton WA: Nature 381:667–673, 1996). Although, these factors cannot be modified, their presence helps to identify those at greatest risk for enabling vigorous treatment of those risk factors that can be modified. However, age is the single most important risk factors for stroke. For each successive 10 years after age 55 the stroke rate more than doubles in both men and women worldwide (Embretson J, Zupancic M, Ribas J, Burke A, Raca P, Tenner-Racz K, Haase AT: Nature 362:359–362, 1993; Fauci AS: Science 239:617–622, 1988; Gelfand AE, Hills SE, Racine-Poon A, Smith AFM: J Am Stat Assoc 85:972–985, 1990; Gelman A, Rubin DB: Stat Sci 7:457–511, 1992; Gray RH, Wawer MJ, Brookmeyer R, Sewankambo NK, Serwadda D, Wabwire-Mangen F, Lutalo T, Li X, van Cott T, Quinn TC, akai Project Team: Lancet 357:1149–1153, 2001; Grewal A, Stephan DA: J Personalized Med 10(Suppl 8):835–848, 2013). On the contrary, stroke treatment and prevention strategies have been very well documented, large volumes of stroke follow up data sets have been generated from the patient’s side. However, in the light of up gradation of comprehensive treatment care and prevention health care services have started digitization of stroke data, resulting from increasingly widespread adoption of electronic health records and electroencephalogram EEG. Electronic activity of brain has greatly facilitated its analysis by varied type of estimation computational methods and thereby enabled large-scale secondary use thereof. This can be exploited to support public health activities; programme implementation such as pharmacovigilance and estimation of DALY (disability adjusted life year) etc. The safety of administered drugs is monitored to inform regularity well versed guidelines and decisions about sustained use. To that larger extent, electronic health records have emerged as potentially valuable data source, providing access to prospective, longitudinal observations of treatment response and drug use. In the context of policy intervention, mathematical and statistical modelling has tremendous potential in improving the quality and efficiency of health care delivery, tool for decision making by health care professionals. New innovative model based discoveries could be necessary to prevent and implementation of new health policy guidelines consideration with risk assessment of genetic interaction, behavioral health related quality life domain (BHQOL) characteristics and associated clinical determinants of stroke (Dempster AP, Laird N, Rubin DB: J R Stat Soc Ser B 39:1–38, 1977; Guo X, Carlin BP: Am Stat 58:16–24, 2004; Hagan A et al: J Stat Plann Infer 91:245–260, 1991; Gelfand AE, Hills SE, Racine-Poon A, Smith AFM: J Am Stat Assoc 85:972–985, 1990).