Tip 1 Diyabette Çok Katmanlı GRU Tabanlı Glikoz Tahmini

Öz

Diyabet teknolojisindeki son gelişmeler, özellikle de sürekli glikoz izleme (CGM) sistemleri, güvenilir glikoz veri kaynakları sağlamaktadır. Bu veriler, yapay zeka ve veri odaklı tekniklerdeki teknolojik ilerlemelerle diyabet hastaları için gelişmiş glikoz tahmin modellerini hızlandırmıştır. Ancak, bu gelişmelere rağmen, modeller karmaşık sıralı verilerdeki bağlamsal örüntüleri öğrenmekte zorlandığından, glikoz seviyelerini doğru bir şekilde tahmin etmek hala bir zorluktur. Bu makalede, kodlayıcı-kod çözücü çerçevesi altında glikoz değerleri dizilerinden özellik çıkarımı için bir konvolüsyonel katman içeren yeni bir çok katmanlı GRU tabanlı model öneriyoruz. Önerilen çok katmanlı GRU tabanlı modeli eğitmek ve test etmek için açık erişimli D1NAMO veri seti kullanılmıştır. Önerilen model, 30 dakikalık glikoz tahmini için 9,88 mg/dL Ortalama Karekök Hatası, 6,46 mg/dL Ortalama Mutlak Hata, 0,92 Belirleme Katsayısı ve %4,83 Ortalama Mutlak Yüzde Hatası elde etmiştir. Ayrıca, tahmin modelinin sağlamlığını değerlendirmek için klinik bir ölçüt olarak Parkes Hata Izgarası kullanılmıştır. Önerilen model, son teknoloji glikoz tahmin modellerine kıyasla üstün performans göstermektedir.

Anahtar Kelimeler

• Yapay Zeka
• Derin Öğrenme
• Kan Şekeri Tahmini
• Kapılı Tekrarlayan Hücre

Stacked GRU-Based Glucose Prediction in Type 1 Diabetes

Abstract

Recent advances in diabetes technology, especially continuous glucose monitoring (CGM) systems, provide reliable sources of glucose data. These data have accelerated advanced glucose prediction models for diabetics with technological advances in artificial intelligence and data-driven techniques. However, despite these advancements, accurately predicting glucose levels still is a challenge as the models struggle to learn contextual patterns in complex sequential data. In this paper, we propose a novel multilayer GRU-based model, including a convolutional layer for feature extraction from sequences of glucose values under the encoder-decoder framework. The open-access D1NAMO dataset was used to train and test the proposed multi-layer GRU-based model. The proposed model achieved a Root Mean Square Error of 9.88 mg/dL, Mean Absolute Error of 6.46 mg/dL, Coefficient of Determination of 0.92, and Mean Absolute Percentage Error of %4.83 for 30-min glucose prediction. Furthermore, the Parkes Error Grid was used as a clinical benchmark to assess the robustness of the prediction model. The proposed model demonstrates superior performance compared to state-of-the-art glucose prediction models.

Keywords

• Artificial Intelligence
• Deep Learning
• Glucose Prediction
• Gated Recurrent Unit

Kaynakça

1. Akosman, Ş. A., Öktem, M., Moral, Ö. T., & Kılıç, V. (2021). Deep Learning-based Semantic Segmentation for Crack Detection on Marbles. 29th Signal Processing and Communications Applications Conference (SIU),
2. Alfian, G., Syafrudin, M., Anshari, M., Benes, F., Atmaji, F. T. D., Fahrurrozi, I., Hidayatullah, A. F., Rhee, J., & Engineering, B. (2020). Blood glucose prediction model for type 1 diabetes based on artificial neural network with time-domain features. Biocybernetics, 40(4), 1586-1599.
3. Aydın, S., Çaylı, Ö., Kılıç, V., & Onan, A. (2022). Sequence-to-sequence video captioning with residual connected gated recurrent units. Avrupa Bilim ve Teknoloji Dergisi(35), 380-386.
4. Chen, G. (2016). A gentle tutorial of recurrent neural network with error backpropagation. arXiv preprint arXiv:.02583.
5. Cho, K., Van Merriënboer, B., Gulcehre, C., Bahdanau, D., Bougares, F., Schwenk, H., & Bengio, Y. (2014). Learning phrase representations using RNN encoder-decoder for statistical machine translation. arXiv preprint arXiv:.
6. Çaylı, Ö., Kılıç, V., Onan, A., & Wang, W. (2022). Auxiliary classifier based residual rnn for image captioning. 30th European Signal Processing Conference (EUSIPCO),
7. Çaylı, Ö., Liu, X., Kılıç, V., & Wang, W. (2023). Knowledge Distillation for Efficient Audio-Visual Video Captioning. arXiv preprint arXiv:.09947.
8. Çaylı, Ö., Makav, B., Kılıç, V., & Onan, A. (2021). Mobile application based automatic caption generation for visually impaired. Intelligent and Fuzzy Techniques: Smart and Innovative Solutions: Proceedings of the INFUS Conference, Istanbul, Turkey,

9. Doǧan, V., Isık, T., Kılıç, V., & Horzum, N. (2022). A field-deployable water quality monitoring with machine learning-based smartphone colorimetry. Analytical Methods, 14(35), 3458-3466.
10. Dubosson, F., Ranvier, J.-E., Bromuri, S., Calbimonte, J.-P., Ruiz, J., & Schumacher, M. (2018). The open D1NAMO dataset: A multi-modal dataset for research on non-invasive type 1 diabetes management. Informatics in Medicine Unlocked 13, 92-100.
11. Fetiler, B., Çaylı, Ö., Moral, Ö. T., Kılıç, V., & Onan, A. (2021). Video captioning based on multi-layer gated recurrent unit for smartphones. Avrupa Bilim ve Teknoloji Dergisi(32), 221-226.
12. Hossain, M. Z., Sohel, F., Shiratuddin, M. F., & Laga, H. (2019). A comprehensive survey of deep learning for image captioning. ACM Computing Surveys, 51(6), 1-36.
13. Keskin, R., Çaylı, Ö., Moral, Ö. T., Kılıç, V., & Onan, A. (2021). A benchmark for feature-injection architectures in image captioning. Avrupa Bilim ve Teknoloji Dergisi(31), 461-468.
14. Kılıç, V. (2021). Yapay Zeka Tabanlı Akıllı Telefon Uygulaması ile Kan Şekeri Tahmini. Avrupa Bilim ve Teknoloji Dergisi(26), 289-294.
15. Kılıç, V., Barnard, M., Wang, W., & Kittler, J. (2014). Audio assisted robust visual tracking with adaptive particle filtering. Transactions on Multimedia, 17(2), 186-200.
16. Kılıç, V., Mercan, Ö. B., Tetik, M., Kap, Ö., & Horzum, N. (2022). Non-enzymatic colorimetric glucose detection based on Au/Ag nanoparticles using smartphone and machine learning. Analytical Sciences, 38(2), 347-358.
17. Li, W., Wu, H., Zhu, N., Jiang, Y., Tan, J., & Guo, Y. (2021). Prediction of dissolved oxygen in a fishery pond based on gated recurrent unit (GRU). Information Processing in Agriculture, 8(1), 185-193.
18. McShinsky, R., & Marshall, B. (2020). Comparison of Forecasting Algorithms for Type 1 Diabetic Glucose Prediction on 30 and 60-Minute Prediction Horizons. KDH@ ECAI,
19. Mercan, Ö. B., Doğan, V., & Kılıç, V. (2020). Time Series Analysis based Machine Learning Classification for Blood Sugar Levels. Medical Technologies Congress (TIPTEKNO),
20. Mercan, Ö. B., & Kılıç, V. (2020). Deep learning based colorimetric classification of glucose with au-ag nanoparticles using smartphone. Medical Technologies Congress (TIPTEKNO),
21. Niu, Z., Zhong, G., & Yu, H. (2021). A review on the attention mechanism of deep learning. Neurocomputing, 452, 48-62.
22. Palaz, Z., Doğan, V., & Kılıç, V. (2021). Smartphone-based Multi-parametric Glucose Prediction using Recurrent Neural Networks. Avrupa Bilim ve Teknoloji Dergisi(32), 1168-1174.
23. Saeed, A., Li, C., Gan, Z., Xie, Y., & Liu, F. (2022). A simple approach for short-term wind speed interval prediction based on independently recurrent neural networks and error probability distribution. Energy, 238, 122012.
24. Saiti, K., Macaš, M., Lhotská, L., Štechová, K., Pithová, P., & Biomedicine, P. i. (2020). Ensemble methods in combination with compartment models for blood glucose level prediction in type 1 diabetes mellitus. Computer Methods Programs in Biomedicine, 196, 105628.
25. Shi, X., Chen, Z., Wang, H., Yeung, D.-Y., Wong, W.-K., & Woo, W.-c. (2015). Convolutional LSTM network: A machine learning approach for precipitation nowcasting. Advances in neural information processing systems 28.
26. Song, J., Wang, Y., Li, F., Akutsu, T., Rawlings, N. D., Webb, G. I., & Chou, K.-C. (2019). iProt-Sub: a comprehensive package for accurately mapping and predicting protease-specific substrates and cleavage sites. Briefings in bioinformatics, 20(2), 638-658.
27. Strollo, F., Furia, A., Verde, P., Bellia, A., Grussu, M., Mambro, A., Petrelli, M., Gentile, S., & practice, c. (2021). Technological innovation of Continuous Glucose Monitoring (CGM) as a tool for commercial aviation pilots with insulin-treated diabetes and stakeholders/regulators: A new chance to improve the directives? diabetes research, 172, 108638.
28. Şen, M., Yüzer, E., Doğan, V., Avcı, İ., Ensarioğlu, K., Aykaç, A., Kaya, N., Can, M., & Kılıç, V. (2022). Colorimetric detection of H2O2 with Fe3O4@ Chi nanozyme modified µPADs using artificial intelligence. Microchimica Acta, 189(10), 373.
29. Uslu, B., Çaylı, Ö., Kılıç, V., & Onan, A. (2022). Resnet based deep gated recurrent unit for image captioning on smartphone. Avrupa Bilim ve Teknoloji Dergisi(35), 610-615.
30. Van Geffen, L., van Herpen, E., & van Trijp, H. (2020). Household Food waste—How to avoid it? An integrative review. Food waste management: Solving the wicked problem 27-55.
31. Wang, Y., & Wang, T. (2020). Application of improved LightGBM model in blood glucose prediction. Applied Sciences, 10(9), 3227.
32. Zhu, T., Li, K., Chen, J., Herrero, P., & Georgiou, P. (2020). Dilated recurrent neural networks for glucose forecasting in type 1 diabetes. Journal of Healthcare Informatics Research, 4, 308-324.