Diagnosis of diabetes mellitus using extreme learning machine

In 2010, Global Status Report on NCD World Health Organization (WHO) reported that 60 percent of deaths in the world caused by the non-communicable diseases, and one of the non-communicable diseases that consumed a lot of attention was diabetes mellitus. Diabetes is a serious threat to the health development, because diabetes is a disease that caused most other diseases (complications), such as blindness, kidney failure, heart disease, diabetic foot (gangrene) so it had to be amputated, up to the most serious is strokes. Estimated in 2030, Indonesia will have 21.3 million people with diabetes mellitus. An increasing number of diabetes caused by the late diagnosis of this disease. Therefore, we need a new forecast which could be an invaluable tool in determining whether someone has diabetes or not. So many methods used to generate accurate predictions; one of them is artificial neural network. This study will implement a new method of neural network, namely the Extreme Learning Machine (ELM). Extreme learning machine is a feed-forward artificial neural network with one or more hidden layers known as single hidden layer feed-forward neural. Based on the results of experiments conducted, it appears that Extreme Learning Machine is able to provide a good prediction accuracy results with a very good prediction rate.