Review article

https://doi.org/10.17559/TV-20190421122826

Machine Learning Techniques for Screening and Diagnosis of Diabetes: a Survey

Yun-lei Sun orcid.org/0000-0003-3745-6899 ; College of Computer & Communication Engineering, China University of Petroleum (East China), Qingdao,266580, China
Da-lin Zhang orcid.org/0000-0003-0346-7020 ; National Research Center of Railway Safety Assessment, Beijing Jiaotong University, Beijing, 100044, China

Abstract

Diabetes has become one of the major causes of national disease and death in most countries. By 2015, diabetes had affected more than 415 million people worldwide. According to the International Diabetes Federation report, this figure is expected to rise to more than 642 million in 2040, so early screening and diagnosis of diabetes patients have great significance in detecting and treating diabetes on time. Diabetes is a multifactorial metabolic disease, its diagnostic criteria is difficult to cover all the ethology, damage degree, pathogenesis and other factors, so there is a situation for uncertainty and imprecision under various aspects of medical diagnosis process. With the development of Data mining, researchers find that machine learning is playing an increasingly important role in diabetes research. Machine learning techniques can find the risky factors of diabetes and reasonable threshold of physiological parameters to unearth hidden knowledge from a huge amount of diabetes-related data, which has a very important significance for diagnosis and treatment of diabetes. So this paper provides a survey of machine learning techniques that has been applied to diabetes data screening and diagnosis of the disease. In this paper, conventional machine learning techniques are described in early screening and diagnosis of diabetes, moreover deep learning techniques which have a significance of biomedical effect are also described.

Keywords

Deep Learning; diabetes; feature extraction; Machine Learning

Hrčak ID:

221017

URI

https://hrcak.srce.hr/221017

Publication date:

12.6.2019.

Visits: 4.227 *