The Early Detection of Diabetes Mellitus (DM) Using Fuzzy Hierarchical Model

In 2013 the number of patients with Diabetes Mellitus (DM) in the world has reached 382 million. It is estimated that the preva- lence will increase 55% in 20351. As a form of our efforts to contribute to the prevention of this phenomenon we propose an application of computational intelligence by using fuzzy hierarchical model that has the ability to perform early detection against DM. In order to achieve the success of our propose method, we cooperate with one of the eastern Jakarta hospital laboratory in Indonesia as a facilitator of the data that we need during research and conducting interviews with two medical doctors at the same hospital as a form of knowledge acquisition process. The architecture of our proposed method is designed based on how a medical doctor concluded related to indication that someone has the potential against DM, which is the model has been adjusted with the data we have obtained from the authorities in the laboratory. As a form of evaluation that we do, we did a comparison of the data we have obtained from our method with the medical doctor decision which is equipped with the data from laboratory, and the result is 87.46% of the 311 relevant data is equal to medical doctor's statement. In interpreting the conclusions that we get with the hospitals that cooperate with us, the results showed that the method we propose has meet the needs of the effectiveness and efficiency in performing early detection against DM and can help people in knowing the potential of DM since early.