Literature DB >> 33726723

Stacked LSTM based deep recurrent neural network with kalman smoothing for blood glucose prediction.

Md Fazle Rabby1, Yazhou Tu2, Md Imran Hossen2, Insup Lee3, Anthony S Maida2, Xiali Hei2.   

Abstract

BACKGROUND: Blood glucose (BG) management is crucial for type-1 diabetes patients resulting in the necessity of reliable artificial pancreas or insulin infusion systems. In recent years, deep learning techniques have been utilized for a more accurate BG level prediction system. However, continuous glucose monitoring (CGM) readings are susceptible to sensor errors. As a result, inaccurate CGM readings would affect BG prediction and make it unreliable, even if the most optimal machine learning model is used.
METHODS: In this work, we propose a novel approach to predicting blood glucose level with a stacked Long short-term memory (LSTM) based deep recurrent neural network (RNN) model considering sensor fault. We use the Kalman smoothing technique for the correction of the inaccurate CGM readings due to sensor error.
RESULTS: For the OhioT1DM (2018) dataset, containing eight weeks' data from six different patients, we achieve an average RMSE of 6.45 and 17.24 mg/dl for 30 min and 60 min of prediction horizon (PH), respectively.
CONCLUSIONS: To the best of our knowledge, this is the leading average prediction accuracy for the ohioT1DM dataset. Different physiological information, e.g., Kalman smoothed CGM data, carbohydrates from the meal, bolus insulin, and cumulative step counts in a fixed time interval, are crafted to represent meaningful features used as input to the model. The goal of our approach is to lower the difference between the predicted CGM values and the fingerstick blood glucose readings-the ground truth. Our results indicate that the proposed approach is feasible for more reliable BG forecasting that might improve the performance of the artificial pancreas and insulin infusion system for T1D diabetes management.

Entities:  

Keywords:  Blood glucose level prediction; Kalman smoothing; Recurrent neural network; Sensor fault correction; Stacked long short-term memory

Mesh:

Substances:

Year:  2021        PMID: 33726723      PMCID: PMC7968367          DOI: 10.1186/s12911-021-01462-5

Source DB:  PubMed          Journal:  BMC Med Inform Decis Mak        ISSN: 1472-6947            Impact factor:   2.796


  33 in total

1.  Neural network incorporating meal information improves accuracy of short-time prediction of glucose concentration.

Authors:  Chiara Zecchin; Andrea Facchinetti; Giovanni Sparacino; Giuseppe De Nicolao; Claudio Cobelli
Journal:  IEEE Trans Biomed Eng       Date:  2012-02-24       Impact factor: 4.538

2.  Modeling Transient Disconnections and Compression Artifacts of Continuous Glucose Sensors.

Authors:  Andrea Facchinetti; Simone Del Favero; Giovanni Sparacino; Claudio Cobelli
Journal:  Diabetes Technol Ther       Date:  2016-02-16       Impact factor: 6.118

3.  Learning long-term dependencies with gradient descent is difficult.

Authors:  Y Bengio; P Simard; P Frasconi
Journal:  IEEE Trans Neural Netw       Date:  1994

4.  Neural network approach for non-invasive detection of hyperglycemia using electrocardiographic signals.

Authors:  Linh Lan Nguyen; Steven Su; Hung T Nguyen
Journal:  Conf Proc IEEE Eng Med Biol Soc       Date:  2014

5.  An early warning system for hypoglycemic/hyperglycemic events based on fusion of adaptive prediction models.

Authors:  Elena Daskalaki; Kirsten Nørgaard; Thomas Züger; Aikaterini Prountzou; Peter Diem; Stavroula Mougiakakou
Journal:  J Diabetes Sci Technol       Date:  2013-05-01

6.  Timing of insulin delivery with meals.

Authors:  E W Kraegen; D J Chisholm; M E McNamara
Journal:  Horm Metab Res       Date:  1981-07       Impact factor: 2.936

7.  Kalman Smoothing for Objective and Automatic Preprocessing of Glucose Data.

Authors:  Odd Martin Staal; Steinar Salid; Anders Fougner; Oyvind Stavdahl
Journal:  IEEE J Biomed Health Inform       Date:  2018-03-01       Impact factor: 5.772

8.  An ARIMA Model With Adaptive Orders for Predicting Blood Glucose Concentrations and Hypoglycemia.

Authors:  Jun Yang; Lei Li; Yimeng Shi; Xiaolei Xie
Journal:  IEEE J Biomed Health Inform       Date:  2018-05-25       Impact factor: 5.772

9.  Hypoglycemia Early Alarm Systems Based On Multivariable Models.

Authors:  Kamuran Turksoy; Elif S Bayrak; Lauretta Quinn; Elizabeth Littlejohn; Derrick Rollins; Ali Cinar
Journal:  Ind Eng Chem Res       Date:  2013-09-04       Impact factor: 3.720

10.  An online failure detection method of the glucose sensor-insulin pump system: improved overnight safety of type-1 diabetic subjects.

Authors:  Andrea Facchinetti; Simone Del Favero; Giovanni Sparacino; Claudio Cobelli
Journal:  IEEE Trans Biomed Eng       Date:  2012-11-15       Impact factor: 4.538

View more
  1 in total

1.  Reservoir Production Prediction Model Based on a Stacked LSTM Network and Transfer Learning.

Authors:  Yukun Dong; Yu Zhang; Fubin Liu; Xiaotong Cheng
Journal:  ACS Omega       Date:  2021-12-07
  1 in total

北京卡尤迪生物科技股份有限公司 © 2022-2023.