A hybrid Transformer-LSTM model apply to glucose prediction
The global prevalence of diabetes is escalating, with estimates indicating that over 536.6 million individuals were afflicted by 2021, accounting for approximately 10.5% of the world’s population. Effective management of diabetes, particularly monitoring and prediction of blood glucose levels, remai...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Public Library of Science
2024
|
| Online Access: | http://psasir.upm.edu.my/id/eprint/114513/ http://psasir.upm.edu.my/id/eprint/114513/1/114513.pdf |
| _version_ | 1848866516021280768 |
|---|---|
| author | Bian, QingXiang As’arry, Azizan Cong, XiangGuo Md Rezali, Khairil Anas |
| author_facet | Bian, QingXiang As’arry, Azizan Cong, XiangGuo Md Rezali, Khairil Anas |
| author_sort | Bian, QingXiang |
| building | UPM Institutional Repository |
| collection | Online Access |
| description | The global prevalence of diabetes is escalating, with estimates indicating that over 536.6 million individuals were afflicted by 2021, accounting for approximately 10.5% of the world’s population. Effective management of diabetes, particularly monitoring and prediction of blood glucose levels, remains a significant challenge due to the severe health risks associated with inaccuracies, such as hypoglycemia and hyperglycemia. This study addresses this critical issue by employing a hybrid Transformer-LSTM (Long Short-Term Memory) model designed to enhance the accuracy of future glucose level predictions based on data from Continuous Glucose Monitoring (CGM) systems. This innovative approach aims to reduce the risk of diabetic complications and improve patient outcomes. We utilized a dataset which contain more than 32000 data points comprising CGM data from eight patients collected by Suzhou Municipal Hospital in Jiangsu Province, China. This dataset includes historical glucose readings and equipment calibration values, making it highly suitable for developing predictive models due to its richness and real-time applicability. Our findings demonstrate that the hybrid Transformer-LSTM model significantly outperforms the standard LSTM model, achieving Mean Square Error (MSE) values of 1.18, 1.70, and 2.00 at forecasting intervals of 15, 30, and 45 minutes, respectively. This research underscores the potential of advanced machine learning techniques in the proactive management of diabetes, a critical step toward mitigating its impact. |
| first_indexed | 2025-11-15T14:21:50Z |
| format | Article |
| id | upm-114513 |
| institution | Universiti Putra Malaysia |
| institution_category | Local University |
| language | English |
| last_indexed | 2025-11-15T14:21:50Z |
| publishDate | 2024 |
| publisher | Public Library of Science |
| recordtype | eprints |
| repository_type | Digital Repository |
| spelling | upm-1145132025-01-16T11:59:31Z http://psasir.upm.edu.my/id/eprint/114513/ A hybrid Transformer-LSTM model apply to glucose prediction Bian, QingXiang As’arry, Azizan Cong, XiangGuo Md Rezali, Khairil Anas The global prevalence of diabetes is escalating, with estimates indicating that over 536.6 million individuals were afflicted by 2021, accounting for approximately 10.5% of the world’s population. Effective management of diabetes, particularly monitoring and prediction of blood glucose levels, remains a significant challenge due to the severe health risks associated with inaccuracies, such as hypoglycemia and hyperglycemia. This study addresses this critical issue by employing a hybrid Transformer-LSTM (Long Short-Term Memory) model designed to enhance the accuracy of future glucose level predictions based on data from Continuous Glucose Monitoring (CGM) systems. This innovative approach aims to reduce the risk of diabetic complications and improve patient outcomes. We utilized a dataset which contain more than 32000 data points comprising CGM data from eight patients collected by Suzhou Municipal Hospital in Jiangsu Province, China. This dataset includes historical glucose readings and equipment calibration values, making it highly suitable for developing predictive models due to its richness and real-time applicability. Our findings demonstrate that the hybrid Transformer-LSTM model significantly outperforms the standard LSTM model, achieving Mean Square Error (MSE) values of 1.18, 1.70, and 2.00 at forecasting intervals of 15, 30, and 45 minutes, respectively. This research underscores the potential of advanced machine learning techniques in the proactive management of diabetes, a critical step toward mitigating its impact. Public Library of Science 2024-09-11 Article PeerReviewed text en cc_by_4 http://psasir.upm.edu.my/id/eprint/114513/1/114513.pdf Bian, QingXiang and As’arry, Azizan and Cong, XiangGuo and Md Rezali, Khairil Anas (2024) A hybrid Transformer-LSTM model apply to glucose prediction. PLoS ONE, 19 (9). art. no. e0310084. pp. 1-13. ISSN 1932-6203; eISSN: 1932-6203 https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0310084 10.1371/journal.pone.0310084 |
| spellingShingle | Bian, QingXiang As’arry, Azizan Cong, XiangGuo Md Rezali, Khairil Anas A hybrid Transformer-LSTM model apply to glucose prediction |
| title | A hybrid Transformer-LSTM model apply to glucose prediction |
| title_full | A hybrid Transformer-LSTM model apply to glucose prediction |
| title_fullStr | A hybrid Transformer-LSTM model apply to glucose prediction |
| title_full_unstemmed | A hybrid Transformer-LSTM model apply to glucose prediction |
| title_short | A hybrid Transformer-LSTM model apply to glucose prediction |
| title_sort | hybrid transformer-lstm model apply to glucose prediction |
| url | http://psasir.upm.edu.my/id/eprint/114513/ http://psasir.upm.edu.my/id/eprint/114513/ http://psasir.upm.edu.my/id/eprint/114513/ http://psasir.upm.edu.my/id/eprint/114513/1/114513.pdf |