Volume 07 Issue 01, JANUARY 2019

A REVIEW ON DIABETES DATASET DETECTION ANALYSIS USING DATA MINING ALGORITHMS

Dipak R Nemade

Page no:01-04

Abstract

Data Mining is used for numerous functions in several applications like industries, medical etc. this can be used for extracting the helpful data from the large quantity of information set. Health observance is additionally used the information mining idea for predict the diagnosing of the diseases. In health observance diabetes is that the common health problem these days, that affects peoples. There are numerous data processing techniques and rule is employed for locating the polygenic disorder. Neural Network, Artificial neural fuzzy interference system, K-Nearest-Neighbor (KNN), Genetic rule, SVM and call Tree etc. These techniques and therefore the algorithms offer the higher result to the individuals and therefore the doctors concerning the diagnosing of the diabetes. From these results the individuals will predict he's affected with the diabetes or non-diabetes additional, performance analysis of various algorithms has been done on this information to diagnose diabetes. The achieved results show the performance of every classification rule.

RFM IS A BETTER METABOLIC PREDICTOR THAN BMI IN A COHORT OF PATIENTS WITH DIABETES MELLITUS

Violeta Hoxha, Gerond Husi, Ruden Cakoni, Eni Celo, Marjeta Kermaj

Page no:05-07

Abstract

As well-known there is a strong connection between obesity and Diabetes Mellitus. Numerous studies are conducted in this field, most of them based on obesity, taking the BMI as a reference value. The aim of our study is to highlight the importance of RFM, especially given that in its formula is incorporated the waist circumference value, as an assessing obesity risk factor for Diabetes Mellitus. We enrolled 137 diabetic patients, hospitalizes in the Endocrine department of “Mother Teresa” Hospital, of Tirana. The variables taken into consideration were age, gender, HbA1c, stature, weight, waist circumference, years with diabetes and from our data we obtained BMI and RFM. We used an IBM SPSS Statistics 23.0 program to analyze the data. Was observed a significative connection between BMI and RFM. It was observed a strong and positive connection between RFM and years with diabetes and HbA1c, that shows that the more the years from diagnosis with diabetes go, the more the chance of obesity increases and the more obese the patient is, the higher get the HbA1c. We do believe that RFM is a better metabolic predictor than BMI in Diabetic patients. Our study was conducted in a small group of patients. We invite researchers to improve the data in this field of study and investigate furtherly.