Gluformer: Transformer-Based Personalized Glucose Forecasting with Uncertainty Quantification
This work addresses the need for reliable uncertainty estimation in personalized glucose forecasting to enhance clinical decision-making, representing an incremental improvement over existing methods.
The paper tackled the problem of slow clinical adoption of deep learning models for blood glucose forecasting due to lack of uncertainty quantification, by proposing a Transformer-based method that models glucose trajectories as an infinite mixture of distributions to learn uncertainty and improve accuracy, achieving superior results on synthetic and benchmark datasets.
Deep learning models achieve state-of-the art results in predicting blood glucose trajectories, with a wide range of architectures being proposed. However, the adaptation of such models in clinical practice is slow, largely due to the lack of uncertainty quantification of provided predictions. In this work, we propose to model the future glucose trajectory conditioned on the past as an infinite mixture of basis distributions (i.e., Gaussian, Laplace, etc.). This change allows us to learn the uncertainty and predict more accurately in the cases when the trajectory has a heterogeneous or multi-modal distribution. To estimate the parameters of the predictive distribution, we utilize the Transformer architecture. We empirically demonstrate the superiority of our method over existing state-of-the-art techniques both in terms of accuracy and uncertainty on the synthetic and benchmark glucose data sets.