Estimation of HbA1c value using artificial neural networks
Main Article Content
Abstract
Diabetes is a life-long disease that occurs because of ineffectiveness or lack of the insulin hormone. Although the blood sugar, fructose and haemoglobin A1c (HbA1c) values are commonly used for diagnosis, the latter give more accurate results. The HbA1c value gives information about the blood sugar levels over the past 2 to 3 months, which is required for treating diabetes. Follow-up data of diabetic patients have been used in this study. In the classification phase, a feed-forward artificial neural network (ANN) was used to estimate the factors affecting HbA1c. The designed ANN has 26 features used as input parameters. The output layer comprises two outputs: normal (HbA1c < 6.5) and high (HbA1c ≥ 6.5). An accuracy rate of 90.33% was obtained with the proposed method. The results, which show accurate estimation of the HbA1c level parameters, will be used in future studies to investigate which parameter affects the HbA1c levels, and in what way.
Keywords: Neural network, haemoglobin A1c, diagnosis of diabetes disease
Downloads
Article Details
- Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (SeeThe Effect of Open Access).