Blood glucose prediction with deep neural networks using weighted decision level fusion

• Autorzy: Dudukcu Hatice Vildan , Taskiran Murat , Yildirim Tulay

Identyfikatory

DOI

10.1016/j.bbe.2021.08.007

• Języki publikacji: EN

Abstrakty

EN

Background and Objective: Diabetes mellitus is a chronic disease that requires regular monitoring of blood glucose in the circulatory system. If the amount of glucose in the blood is not regulated constantly, this may have vital consequences for the individual. For this reason, there are many studies in the literature that perform blood glucose (BG) prediction. Methods: Blood glucose prediction is generally performed by using many parameters. In this paper, it was attempted to predict the future blood glucose values of the patient by using only the blood glucose values of diabetes patients’ history. For this purpose, Long short term memory (LSTM), WaveNet and Gated Recurrent Units (GRU) and decision-level combinations of these architectures were used to predict blood glucose. First of all, hyper-parameters were selected for the most efficient operation of these network architectures and experimental studies were conducted using the extended OhioT1DM data set which has blood glucose history of 12 diabetes patients. Results: Experimental studies using 30, 45 and 60 min prediction horizon (PH), the average lowest RMSE value were obtained by the fusion of three networks as 21.90 mg/dl, 29.12 mg/dl, 35.10 mg/dl respectively. Conclusions: When the obtained RMSE value compared to state-of-art studies in the literature, the results show that the proposed method is quite successful for short-term blood glucose prediction. In addition, the proposed fusion method gives a new perspective for future studies in the literature for BG prediction.

Słowa kluczowe

EN

blood glucose prediction; long short term memory; LSTM; WaveNet; gated recurrent units; GRU; decision level fusion

PL

stężenie glukozy; LSTM; WaveNet; GRU

• Wydawca: Nałęcz Institute of Biocybernetics and Biomedical Engineering of the Polish Academy of Sciences ; Elsevier
• Czasopismo: Biocybernetics and Biomedical Engineering
• Rocznik: 2021
• Tom: Vol. 41, no. 3
• Strony: 1208--1223
• Opis fizyczny: Bibliogr. 37 poz., rys., tab., wykr.

Twórcy

autor

Dudukcu Hatice Vildan

• Yildiz Technical University Dept. of Electronics and Communication Engineering, 34220 Istanbul, Turkey

autor

Taskiran Murat

• Yildiz Technical University Dept. of Electronics and Communication Engineering, Istanbul, Turkey

autor

Yildirim Tulay

• Yildiz Technical University Dept. of Electronics and Communication Engineering, Istanbul, Turkey

Bibliografia

• [1] Pouya Saeedi, Paraskevi Salpea, Suvi Karuranga, Inga Petersohn, Belma Malanda, Edward W Gregg, Nigel Unwin, Sarah H Wild, Rhys Williams. Mortality attributable to diabetes in 20–79 years old adults, 2019 estimates: Results from the international diabetes federation diabetes atlas. Diabetes research and clinical practice; 2020. p. 108086.
• [2] Ashenafi Zebene Woldaregay, Eirik Ðrsand, Ståle Walderhaug, David Albers, Lena Mamykina, Taxiarchis Botsis, Gunnar Hartvigsen. Data-driven modeling and prediction of blood glucose dynamics: Machine learning applications in type 1 diabetes. Artif Intell Med 2019;98:109–134.
• [3] Oviedo Silvia, Vehí Josep, Calm Remei, Armengol Joaquim. A review of personalized blood glucose prediction strategies for t1dm patients. Int J Numer Methods Biomed Eng 2017;33(6)e2833.
• [4] J Ignacio Hidalgo, J Manuel Colmenar, Gabriel Kronberger, Stephan M Winkler, Oscar Garnica, Juan Lanchares. Data based prediction of blood glucose concentrations using evolutionary methods. J Med Syst 2017;41(9):142.
• [5] Doike Takuyoshi, Hayashi Kenya, Arata Shigeki, Mohammad Karim Nissar, Kobayashi Atsuki, Niitsu Kiichi. A blood glucose level prediction system using machine learning based on recurrent neural network for hypoglycemia prevention. In: 2018 16th IEEE International New Circuits and Systems Conference (NEWCAS), IEEE. p. 291–5.
• [6] John Martinsson, Alexander Schliep, Bjorn Eliasson, Christian Meijner, Simon Persson, Olof Mogren. Automatic blood glucose prediction with confidence using recurrent neural networks. In: KHD@ IJCAI, 2018.
• [7] Taiyu Zhu, Kezhi Li, Pau Herrero, Jianwei Chen, and Pantelis Georgiou. A deep learning algorithm for personalized blood glucose prediction. In: KHD@ IJCAI, 2018, p. 64–78.
• [8] Takoua Hamdi, Jaouher Ben Ali, Véronique Di Costanzo, Farhat Fnaiech, Eric Moreau, and Jean-Marc Ginoux. Accurate prediction of continuous blood glucose based on support vector regression and differential evolution algorithm. Biocybern Biomed Eng 2018;38(2):362–372.
• [9] Jaouher Ben Ali, Takoua Hamdi, Nader Fnaiech, Véronique Di Costanzo, Farhat Fnaiech, Jean-Marc Ginoux. Continuous blood glucose level prediction of type 1 diabetes based on artificial neural network. Biocybern Biomed Eng 2018;38 (4):828–840.
• [10] Aliberti Alessandro, Pupillo Irene, Terna Stefano, Macii Enrico, Di Cataldo Santa, Patti Edoardo, Acquaviva Andrea. A multi-patient data-driven approach to blood glucose prediction. IEEE Access 2019;7:69311–25.
• [11] Yang Jun, Li Lei, Shi Yimeng, Xie Xiaolei. An arima model with adaptive orders for predicting blood glucose concentrations and hypoglycemia. IEEE J Biomed Health Inf 2018;23(3):1251–60.
• [12] Wei Song, Wanyuan Cai, Jing Li, Fusong Jiang, Shengqi He. Predicting blood glucose levels with emd and lstm based cgm data. In: 2019 6th International Conference on Systems and Informatics (ICSAI), IEEE; 2019, p. 1443–1448.
• [13] Martinsson John, Schliep Alexander, Eliasson Björn, Mogren Olof. Blood glucose prediction with variance estimation using recurrent neural networks. J Healthcare Inf Res 2020;4(1):1–18.
• [14] Xia Yu, Rashid Mudassir, Feng Jianyuan, Hobbs Nicole, Hajizadeh Iman, Samadi Sediqeh, et al. Online glucose prediction using computationally efficient sparse kernel filtering algorithms in type-1 diabetes. IEEE Trans Control Syst Technol 2018.
• [15] Li Kezhi, Daniels John, Liu Chengyuan, Herrero Pau, Georgiou Pantelis. Convolutional recurrent neural networks for glucose prediction. IEEE J Biomed Health Inf 2019;24(2):603–13.
• [16] Li Kezhi, Liu Chengyuan, Zhu Taiyu, Herrero Pau, Georgiou Pantelis. Glunet: a deep learning framework for accurate glucose forecasting. IEEE J Biomed Health Inf 2019;24(2):414–23.
• [17] Xie Jinyu, Wang Qian. Benchmarking machine learning algorithms on blood glucose prediction for type 1 diabetes in comparison with classical time-series models. IEEE Trans Biomed Eng 2020.
• [18] Wang Wenbo, Tong Meng, Min Yu. Blood glucose prediction with vmd and lstm optimized by improved particle swarm optimization. IEEE Access 2020.
• [19] Zhu Taiyu, Yao Xi, Li Kezhi, Herrero Pau, Georgiou Pantelis. Blood glucose prediction for type 1 diabetes using generative adversarial networks. CEUR Workshop Proc 2020;2675:90–4.
• [20] Ganjar Alfian, Muhammad Syafrudin, Jongtae Rhee, Muhammad Anshari, M Mustakim, Imam Fahrurrozi. Blood glucose prediction model for type 1 diabetes based on extreme gradient boosting. In IOP Conference Series: Materials Science and Engineering, vol. 803, IOP Publishing, 2020. p. 012012.
• [21] Cindy Marling and Razvan C Bunescu. The ohiot1dm dataset for blood glucose level prediction. In: KHD@ IJCAI, 2018. p. 60–63.
• [22] Maxime De Bois, Mehdi Ammi, Mounı_m A El Yacoubi. Glyfe: Review and benchmark of personalized glucose predictive models in type-1 diabetes. arXiv preprint arXiv:2006.15946, 2020.
• [23] Md Fazle Rabby, Yazhou Tu, Md Imran Hossen, Insup Lee, Anthony S Maida, Xiali Hei. Stacked lstm based deep recurrent neural network with kalman smoothing for blood glucose prediction. BMC Med Inf Decis Mak 2021;21 (1):1–15.
• [24] Asiye Şahin, Ahmet Aydın. Personalized advanced time blood glucose level prediction. Arab J Sci Eng 2021 p. 1–12.
• [25] Cindy Marling, Razvan Bunescu. The ohiot1dm dataset for blood glucose level prediction: Update 2020. KHD@ IJCAI, 2020.
• [26] Gábor Petneházi. Recurrent neural networks for time series forecasting. arXiv preprint arXiv:1901.00069, 2019.
• [27] Hochreiter Sepp, Bengio Yoshua, Frasconi Paolo, Schmidhuber Jürgen, et al., editors. Gradient flow in recurrent nets: the difficulty of learning long-term dependencies.
• [28] Hochreiter Sepp, Schmidhuber Jürgen. Long short-term memory. Neural Comput 1997;9(8):1735–80.
• [29] Kyunghyun Cho, Bart Van Merriënboer, Caglar Gulcehre, Dzmitry Bahdanau, Fethi Bougares, Holger Schwenk, Yoshua Bengio. Learning phrase representations using rnn encoderdecoder for statistical machine translation. arXiv preprint arXiv:1406.1078, 2014.
• [30] Aaron van den Oord, Sander Dieleman, Heiga Zen, Karen Simonyan, Oriol Vinyals, Alex Graves, Nal Kalchbrenner, Andrew Senior, Koray Kavukcuoglu. Wavenet: A generative model for raw audio. arXiv preprint arXiv:1609.03499, 2016.
• [31] LaValle Steven M, Branicky Michael S, Lindemann Stephen R. On the relationship between classical grid search and probabilistic roadmaps. Int J Robot Res 2004;23(7–8):673–92.
• [32] Chai Tianfeng, Draxler Roland R. Root mean square error (rmse) or mean absolute error (mae)?–arguments against avoiding rmse in the literature. Geosci Model Develop 2014;7(3):1247–50.
• [33] Ummul Khair, Hasanul Fahmi, Sarudin Al Hakim, Robbi Rahim. Forecasting error calculation with mean absolute deviation and mean absolute percentage error. In: Journal of Physics: Conference Series, vol. 930, IOP Publishing, 2017, p. 012002.
• [34] David C Klonoff, Courtney Lias, Robert Vigersky, William Clarke, Joan Lee Parkes, David B Sacks, M Sue Kirkman, Boris Kovatchev, Error Grid Panel. The surveillance error grid. J Diabet Sci Technol 2014;8(4):658–672.
• [35] Kovatchev Boris P, Wakeman Christian A, Breton Marc D, Kost Gerald J, Louie Richard F, Tran Nam K, Klonoff David C. Computing the surveillance error grid analysis: procedure and examples. J Diabet Sci Technol 2014;8(4):673–84.
• [36] Sampath Sivananthan, Valeriya Naumova, Chiara Dalla Man, Andrea Facchinetti, Eric Renard, Claudio Cobelli, Sergei V Pereverzyev. Assessment of blood glucose predictors: the prediction-error grid analysis. Diabet Technol Therap 2011;13(8):787–796.
• [37] Naumova Valeriya, Nita Lucian, Poulsen Jens Ulrik, Pereverzyev Sergei V. Meta-learning based blood glucose predictor for diabetic smartphone app. In: Prediction methods for blood glucose concentration. Springer; 2016. p. 93–105.

Uwagi

Opracowanie rekordu ze środków MNiSW, umowa Nr 461252 w ramach programu "Społeczna odpowiedzialność nauki" - moduł: Popularyzacja nauki i promocja sportu (2021).