Key result
An IoMT-enabled wearable device using an attention-based evidential recurrent neural network achieved superior RMSE and hypoglycemia detection accuracy compared to baseline machine learning methods.
Why the study?
Deep learning models for blood glucose prediction require high computational resources making clinical implementation challenging, while smartphone-based implementations face short battery life and portability constraints.
Does an IoMT-enabled deep learning model improve real-time blood glucose prediction and hypoglycemia detection in type 1 diabetes compared to baseline machine learning methods?
Does an IoMT-enabled deep learning model improve real-time blood glucose prediction and hypoglycemia detection in type 1 diabetes compared to baseline machine learning methods?
An IoMT-enabled wearable device using an embedded deep learning model provides superior real-time blood glucose prediction and hypoglycemia detection compared to baseline machine learning methods.
Answers are grounded in this paper and cite their sources.
May support IoMT glucose prediction in T1D but should not yet change practice; leaves open prospective validation in clinical trials.
Blood glucose (BG) prediction is essential to the success of glycemic control in type 1 diabetes (T1D) management. Empowered by the recent development of the Internet of Medical Things (IoMT), continuous glucose monitoring (CGM) and deep learning technologies have been demonstrated to achieve the state of the art in BG prediction. However, it is challenging to implement such algorithms in actual clinical settings to provide persistent decision support due to the high demand for computational resources, while smartphone-based implementations are limited by short battery life and require users to carry the device. In this work, we propose a new deep learning model using an attention-based evidential recurrent neural network and design an IoMT-enabled wearable device to implement the embedded model, which comprises a low-cost and low-power system on a chip to perform Bluetooth connectivity and edge computing for real-time BG prediction and predictive hypoglycemia detection. In addition, we developed a smartphone app to visualize BG trajectories and predictions, and desktop and cloud platforms to backup data and fine-tune models. The embedded model was evaluated on three clinical data sets including 47 T1D subjects. The proposed model achieved superior performance of root mean square error (RMSE), mean absolute error, and glucose-specific RMSE, and obtained the best accuracy for hypoglycemia detection when compared with a group of machine learning baseline methods. Moreover, we performed hardware-in-the-loop in silico trials with ten virtual T1D adults to test the whole IoMT system with predictive low-glucose management, which significantly reduced hypoglycemia and improved BG control.
No takes yet. Share an insight, caveat, or question.
Zhu et al. (2022) studied Type 1 diabetes (n=47). IoMT-enabled wearable device with attention-based evidential recurrent neural network vs. Machine learning baseline methods was evaluated on Root mean square error (RMSE), mean absolute error, glucose-specific RMSE, and hypoglycemia detection accuracy. An IoMT-enabled wearable device using an attention-based evidential recurrent neural network achieved superior RMSE and hypoglycemia detection accuracy compared to baseline machine learning methods.
Synapse has enriched 10 related papers. Consider them for comparative context: