Analysis Diabetes Mellitus on Complications with Data Mining

Analysis Diabetes Mellitus on Complications with Data Mining M. Mayilvaganan T.Sivaranjani Abstract: Diabetes mellitus is incredible growing and seems to be emerging as a main public health problem for our country.The prevalence of diabetes is rapidly increasing all over the world. Data mining provides more no of tools and techniques that can be applied to this processed data to discover hidden patterns. It is used to provide healthcare professionals an additional source of knowledge for making decisions. This research is analysis about diabetes prevalence, complications, and preventing from complications. Keywords— diabetes mellitus, data analysis, data mining, diabetes prevalence, complications INTRODUCTION: Diabetes is a group of metabolic diseasescaused by the lack of insulin in the body or inability to produce as normal. In contemporary world most of folk are distressed by diabetes, which affects a large population across the world. The prevalence of diabetes for all age-groups worldwide was estimated to be 2.8% in 2000 and 4.4% in 2030. The total number of people with diabetes is projected to rise from 171 million in 2000 to 366 million in 2030. The prevalence of diabetes is higher in men than women, but there are more women with diabetes than men. The urban population in developing countries is projected to double between 2000 and 2030[9]. TYPES OF DIABETES Type 1 diabetes This type of diabetes usually develops during childhood or teens and is characterized by a severe deficiency of insulin secretion resulting from atrophy of the islets of Langerhans and causing hyperglycemia and a marked tendency toward ketoacidosis—called alsoinsulin-dependent diabetes, insulin-dependent diabetes mellitus, juvenile diabetes, juvenile-onset diabetes, type 1 diabetes mellitus [6]. Type 2 diabetes It’s mostly distressed in adulthood and is exacerbated by obesity and an inactive lifestyle. This disease often has no symptoms, is usually diagnosed by tests that indicate glucose intolerance, and is treated with changes in diet and an exercise regularly [7]. Gestational diabetes Gestational diabetes mellitus (GDM) is defined as any degree of glucose intolerance with onset or first recognition during pregnancy [8]. The definition applies whether insulin or only diet modification is used for treatment and whether or not the condition persists after pregnancy. It does not exclude the possibility that unrecognized glucose intolerance may have antedated or begun concomitantly with the pregnancy. Fig 1: Prevalence estimates of diabetes mellitus (DM), 2010 – South-East Asian Region To estimating the prevalence of diabetes for the years 2000, 2010 and 2030, data on case numbers and national prevalence of impaired glucose tolerance are presented in chart [10].The total populations of the regions and the populations aged from 20-79 years are shown in Figure 2. From the figure we clearly known Western Pacific Region, which includes China, and the South-East Asian Region, which has India as a member, have the greatest numbers of people [10]. Fig2: Top 5- Number of people with diabetes (20-79 age group), 2000, 2010 and 2030 Fig 3 Top 5-Prevalence of impaired glucose tolerance (20-79 age group), 2010 and 2030 COMPLICATIONS OF DIABETES Skin Complications To be more consciousness for symptoms of skin infections and other skin disorders common in people with diabetes. Eye Complications Yearly or six months once keep regular check up; avoid risk of glaucoma, cataracts and other eye problems. Due to nation survey in India eye complication was rare. Neuropathy Nerve damage from diabetes is called diabetic neuropathy .The majority of people with diabetes have any one of type of nerve damage. Foot Complications The largest parts of diabetes patients have foot damages. Take care of our foot as much as like face. Before bed we have to clean and dry our foot. Through the regular excise and walking we can avoid this complication. Kidney Disease (Nephropathy) High BP and glucose is major cause this. Keep your diabetes and blood pressure under control to lesser the chance of getting kidney disease. High Blood Pressure High blood pressure is also called hypertension. It raises more complications like heart attack, stroke, eye problems, and kidney disease. Stroke Keep up blood glucose, blood pressure, and cholesterol in good level. It to be reduces your risk of stroke. Most of the patients affected stroke by hypertension. DATA MINING TECHNIQUES In healthcare industry nowadays generates huge amounts of complex data about patients, hospitals resources, disease diagnosis, electronic patient records, and medical devices etc.These data are a key resource to be stored, processed and analyzed for knowledge extraction that enables to support for cost-savings and decision making. Data mining is the process of exploration and analysis, by automatic or semiautomatic means, of large quantities of data in order to discover meaningful patterns and rules [11]. Data mining could be on the whole useful in medicine when there is no dispositive evidence favoring an exacting treatment option. Based on patients’ profile, history, physical inspection, diagnosis and utilizing previous treatment patterns, new treatment policy can be successfully recommended. Data mining is finding interesting structure (patterns, statistical models, relationships) in databases. [12]. Logistic regression models are used to compare hospital profiles and based on that risk-factors are analyses in data mining. Artificial neural networks are used in medical diagnosis. It produces a clinically relevant output based on sample database, and constructs the probability of a certain pathology or classification of biomedical objects. Due to the generous plasticity of input data, ANNs have verified useful in the analysis of blood and urine samples of diabetic patients [13]. Unsupervised learning engrosses identifying clusters and associations. Clusters are faction the analogous subtypes and make group. Using regression analysis, associate the following attributes as age, family history, increasing socio-economic status and decreasing physical activity and find high frequency of cause which type of diabetes distressed. No one can tell literally, which algorithm is best for any problem, because data sets from various data sources. To applying algorithm in training set and came to the solution, which is suite .data set be consists of missing values, noise, and outliers. Cleaning data from noise and outliers and handling missing values, and then finding the exact subset of data and prepares them for successful data mining. Missing values are filled up with the most familiar value and combinations of particular attribute-value pairs are significant within a dataset. DATA SET REPRESENTATION Collecting patient’s medical details based on that calculated BMI, body type, required calories, actual calories, complications, risk factors. The table 1 specified for risk analysis and table 2 for diagnosed complications. Some of the attributes of datasets are BMI, require weight, BMI index, working industry, eating habit, blood group, life style, and require calorie based on sex, family history,PCOS,HBA1c,Smoker, drinker, type of DM,dignosed age, symptoms, no of years, Gestational diabetes history, baby weight, order of baby, control DM,Fast food,BP,food intervals, intake limits. Table 1. Characteristics of risk analysis data set 2 3 4 2 2 7 2 1 22 1 4 3 1 3 1 2 0 Monitor the following attributes as high HBA1c, stationary, job, BP, Life style, disease caused after diabetes diagnosed, undiet, smoking, drinking habits regularly can avoid more complications. Table 2. Characteristics of complications data set 1 3 6 1 1 1 2 1 1 1 67 2 4 4 2 0 2 2 1 0 5 Conclusion India is top most country in prevalence of diabetes. Number of people with diabetes in our country in 2010 50.8 million and will be estimated 87.0 in 2030[10]. Diabetes complication fatality rates also raised and prevent these government or social organizations, health care’s must provide education or training focuses on self-care behaviors, such as healthy eating, being active, and monitoring blood sugar. Many of the steps necessitate to take to avert one of those complications may really help to prevent them all. This kind of education or training is a mutual process in which diabetes educators help people with or at risk for diabetes gain the knowledge. Data mining bring a set of tools, techniques and method that can be functional to this processed data to determine hidden patterns. Data mining algorithms are used to extract informative patterns from raw data. Physicians can identify effective treatments and best observation, and also patients receive improved and more affordable healthcare services. It is help to manage and monitor patients can have important utility in diabetes mellitus and analysis complicates. In the future, we plan to demonstrate the usefulness of this kind of study by measuring the extent to which data mining approaches empower clinical research and practice. References: [1]. Dandona, Lalit, et al. Population based assessment of diabetic retinopathy in an urban population in southern India.British journal of ophthalmology83.8 (1999): 937-940. [2]. Sanders, Reginald J., and M. Roy Wilson. Diabetes-related eye disorders.Journal of the National Medical Association85.2 (1993): 104. [3]. Gà ¤ckler, D., et al. [Diabetes and kidneys].Deutsche medizinische Wochenschrift (1946)138.18 (2013): 949-955. [4]. Berger, A. and Berger, C.R. â€Å"Data mining as a tool for research and knowledge development in nursing.†CINMay/June 2004. [5]. Stephens, S. and Tamayo, P. â€Å"Supervised and unsupervised data mining techniques for life sciences.†Curr Drug DiscJune 2003. [6]. Ewing, D. J., I. W. Campbell, and B. F. Clarke. The natural history of diabetic autonomic neuropathy.QJM49.1 (1980): 95-108. [7].  http://www.merriam-webster.com/dictionary/type%201%20diabetes [8]. Metzger BE, Coustan DR (Eds.): Proceedings of the Fourth International Work-shop-Conference on Gestational Diabetes Mellitus.Diabetes Care21 (Suppl. 2):B1–B167,1998 [9]. Wild, Sarah, et al. Global prevalence of diabetes estimates for the year 2000 and projections for 2030.Diabetes care27.5 (2004): 1047-1053. [10]. Sicree, Richard, et al. The global burden.Diabetes and impaired glucose tolerance. Baker IDI Heart and Diabetes Institute(2010). [11]. Berry, Michael JA, and Gordon Linoff. Data Mining Techniques . J. (2004). [12]. Bradley, Paul S., Usama M. Fayyad, and Olvi L. Mangasarian. Mathematical programming for data mining: formulations and challenges.INFORMS Journal on Computing11.3 (1999): 217-238. [13]. Amato, Filippo, et al. Artificial neural networks in medical diagnosis.Journal of Applied Biomedicine11.2 (2013): 47-58. [13]. Data Mining Technologies for Blood Glucose and Diabetes Management 603 , Riccardo Bellazzi, Ph.D.,and Ameen Abu-Hanna, Ph.D. [14].  http://health.india.com/diseases-conditions/sweet-nothings-discard-myths-to-successfully-manage-diabetes/ [15]. Application of data mining: Diabetes health care in young and old patients Abdullah A. Aljumah, Mohammed Gulam Ahamad, Mohammad Khubeb Siddiqui [16]. An Analysis of Diabetes Risk Factors Using Data Mining Approach Akkarapol Sangasoongsong and Jongsawas Chongwatpol Oklahoma State University, Stillwater, OK 74078, USA [17]. The need for obtaining accurate nationwide estimates of diabetes prevalence in India Rationale for a national study on diabetes R.M. Anjana, M.K. Ali*, R. Pradeepa, M. Deepa, M. Datta, R. Unnikrishnan, M. Rema V. Mohan [18]. Am I at risk for type 2 diabetes? Taking Steps to Lower Your Risk of Getting Diabetes [19]. http://www.diabetes.org/living-with-diabetes/complications/