Deep Multi-Output Forecasting: Learning to Accurately Predict Blood Glucose Trajectories

@article{Fox2018DeepMF,
  title={Deep Multi-Output Forecasting: Learning to Accurately Predict Blood Glucose Trajectories},
  author={Ian Fox and Lynn Ang and Mamta Jaiswal and Rodica Pop-Busui and Jenna Wiens},
  journal={Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery \& Data Mining},
  year={2018}
}
  • Ian Fox, Lynn Ang, +2 authors J. Wiens
  • Published 2018
  • Computer Science, Mathematics
  • Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining
In many forecasting applications, it is valuable to predict not only the value of a signal at a certain time point in the future, but also the values leading up to that point. This is especially true in clinical applications, where the future state of the patient can be less important than the patient's overall trajectory. This requires multi-step forecasting, a forecasting variant where one aims to predict multiple values in the future simultaneously. Standard methods to accomplish this can… Expand
Deep Personalized Glucose Level Forecasting Using Attention-based Recurrent Neural Networks
TLDR
This work studies the problem of blood glucose forecasting and proposes a deep personalized solution that learns a personalized model for each patient as well as a global model and introduces a new, robust training procedure for time series data. Expand
Benchmarking Machine Learning Algorithms on Blood Glucose Prediction for Type I Diabetes in Comparison With Classical Time-Series Models
TLDR
There was no significant advantage observed from the ML models compared to the classic ARX model in predicting BG levels for T1D, except that TCN's performance was more robust with respect to BG trajectories with spurious oscillations. Expand
Deep learning-based data imputation on time-variant data using recurrent neural network
TLDR
A novel approach for predicting the blood glucose levels with the aid of Maclaurin series-based expectation maximization, estimation of correlation relationship and dissimilarities, kernel-based Hilbert–Schmidt optimization, optimized features, and classification using the deep learning methodology of RNN—recurrent neural network is proposed. Expand
Deep learning and regression approaches to forecasting blood glucose levels for type 1 diabetes
  • Meng Zhang, K. Flores, Hien T. Tran
  • Computer Science
  • Biomed. Signal Process. Control.
  • 2021
TLDR
A neural network model using an encoder-decoder architecture has the most stable performance and is able to recover missing dynamics in short time intervals and may enable a feasible solution with low computational cost for the time-dependent adjustment of artificial pancreas for diabetes patients. Expand
Deep Physiological Model for Blood Glucose Prediction in T1DM Patients
TLDR
A new hybrid model is proposed that decomposes a deep machine learning model in order to mimic the metabolic behavior of physiological blood glucose methods and the results show Root Mean Square Error (RMSE) values under 5 mg/dL for simulated patients and under 10 mg/DL for real patients. Expand
Multi-step Deep Autoregressive Forecasting with Latent States
In many forecasting applications, it is important to predict an overall trajectory within a period of time, i.e. a forecasting horizon. Various deep neural network models have been introduced toExpand
Benchmark Machine Learning Approaches with Classical Time Series Approaches on the Blood Glucose Level Prediction Challenge
TLDR
In this paper, experiments are conducted to compare the performances of multiple popular machine learning algorithms including two major types of deep learning approaches, with the classic autoregression with exogenous inputs (ARX) model on this particular Blood Glucose Level Prediction (BGLP) Challenge. Expand
Comparative Analysis of Deep Learning Models for Multi-Step Prediction of Financial Time Series
TLDR
Experimental results show that the latest deep learning models such as N-BEATS, ES-LSTM and TCN produced better results for all stock market related datasets by obtaining around 50% less Root Mean Squared Error (RMSE) and Mean Absolute Error (MAE) scores for each prediction horizon as compared to other models. Expand
Deep Learning for Blood Glucose Prediction: CNN vs LSTM
TLDR
The results show that the proposed CNN outperformed significantly the LSTM model for both one-step and multi-steps prediction and no MSF strategy outperforms the others for CNN. Expand
Comparing Machine Learning Techniques for Blood Glucose Forecasting Using Free-living and Patient Generated Data
TLDR
A rigorous comparison of BG forecasting using both a small controlled research dataset and large heterogeneous PGHD is introduced, providing insight into how methods may fare when moving beyond small controlled studies to real-world use. Expand
...
1
2
3
4
5
...

References

SHOWING 1-10 OF 32 REFERENCES
Jump neural network for real-time prediction of glucose concentration.
TLDR
Results with a prediction horizon of 30 min show that prediction of glucose concentration in T1D via NN is feasible and sufficiently accurate and the average time anticipation obtained is compatible with the generation of preventive hypoglycemic and hyperglycemic alerts and the improvement of artificial pancreas performance. Expand
A Machine Learning Approach to Predicting Blood Glucose Levels for Diabetes Management
TLDR
A generic physiological model of blood glucose dynamics is used to generate informative features for a Support Vector Regression model that is trained on patient specific data and could be used to anticipate almost a quarter of hypoglycemic events 30 minutes in advance. Expand
Using LSTMs to learn physiological models of blood glucose behavior
TLDR
A recursive neural network (RNN) approach that uses long short-term memory (LSTM) units to learn a physiological model of blood glucose that is competitive with previous state-of-the-art model based on manually engineered physiological equations is described. Expand
Hypoglycemia Prediction with Subject-Specific Recursive Time-Series Models
TLDR
Compared to the absolute value method, both CUSUM and EWMA methods behaved more conservatively before raising an alarm (reduced time to detection), which significantly decreased the false alarm rate and increased the specificity. Expand
Hypoglycemia Prediction Using Machine Learning Models for Patients With Type 2 Diabetes
TLDR
This work trained a probabilistic model using machine learning algorithms and SMBG values from real patients to predict hypoglycemia events with a high degree of sensitivity and specificity and validated this model retrospectively and if implemented in real time. Expand
A Multivariate Timeseries Modeling Approach to Severity of Illness Assessment and Forecasting in ICU with Sparse, Heterogeneous Clinical Data
TLDR
This work evaluates the use of multivariate timeseries modeling with the multi-task Gaussian process (GP) models using noisy, incomplete, sparse, heterogeneous and unevenly-sampled clinical data, including both physiological signals and clinical notes to assess and forecast patient acuity. Expand
Probabilistic short-term wind power forecasting based on deep neural networks
High-precision wind power forecasting is an essential operation issue of power systems integrated with large numbers of wind farms. In addition to traditional forecasting methods, probabilisticExpand
Deep Learning for solar power forecasting — An approach using AutoEncoder and LSTM Neural Networks
TLDR
This work uses different Deep Learning and Artificial Neural Network algorithms, such as Deep Belief Networks, AutoEncoder, and LSTM, to show their forecast strength compared to a standard MLP and a physical forecasting model in the forecasting the energy output of 21 solar power plants. Expand
Nonlinear model predictive control of glucose concentration in subjects with type 1 diabetes.
TLDR
Adapt nonlinear model predictive control is promising for the control of glucose concentration during fasting conditions in subjects with type 1 diabetes. Expand
A review and comparison of strategies for multi-step ahead time series forecasting based on the NN5 forecasting competition
TLDR
Three findings appear to be consistently supported by the experimental results: Multiple-Output strategies are the best performing approaches, deseasonalization leads to uniformly improved forecast accuracy, and input selection is more effective when performed in conjunction with dese Masonalization. Expand
...
1
2
3
4
...