PMB-NN: Physiological-Model-Based Neural Network for Glucose Forecasting in Type 1 Diabetes
Abstract
Accurate, physiologically consistent glucose forecasts could help people with type 1 diabetes respond to impending excursions. We applied our physiological-model-based neural network (PMB--NN) framework to glucose forecasting. A temporal convolutional network (TCN) used 180 min of history, constrained during training by a personalized four-parameter physiological model. The main PMB--NN combined forecast and neural--physiological agreement losses; a three-term variant additionally fitted physiological predictions to measured glucose. Data comprised 6--8 weeks of continuous glucose monitoring, pump, and meal records from 25 AZT1D participants. Person-specific models used chronological 70/10/20 train/validation/test splits and 30-, 60-, and 120-min horizons. Median mean absolute errors were 16.24, 25.00, and 30.20 mg/dL with basal insulin records and 15.38, 24.21, and 31.79 mg/dL before these records became available. In the latter setting, PMB--NN had the lowest median error at 30 and 60 min; it outperformed the physiological model alone in all six settings. In input-increase tests, glucose predictions changed in the expected direction for 85\% of correction boluses and 100\% of carbohydrate rescues, versus 57\% for active insulin and 56\% for active carbohydrates. Most standalone physiological fits showed locally distinguishable parameter effects, with glucose-related coefficients more stable under resampling than insulin- or carbohydrate-related coefficients. The three-term variant improved hyperglycaemia accuracy with basal records, but did not significantly change overall MAE and reduced directional plausibility in three tests. PMB--NN combines personalized forecasting with an explicit physiological model, allowing assessment of predictive accuracy, physiological response behavior, and participant-specific coefficients.
Related
- Large-Scale Benchmarking of Quantum Neural Network Configurations for Financial Time Series Forecasting
- Scaling-law-informed neural point processes for earthquake sequence forecasting
- Reducing boundary conditions for the Thomas–Fermi equation using a hybrid q-deformation and neural network method
- PINNIES: An efficient physics-informed neural network framework for integral operator problems
- Probe-aware self-supervised neural network for X-ray holographic reconstruction
BibTeX
@misc{zhang2026pmbnn,
title = {{PMB-NN: Physiological-Model-Based Neural Network for Glucose Forecasting in Type 1 Diabetes}},
author = {Yaowen Zhang and Shuhao Que and Ying Wang},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.15456},
eprint = {2610.15456},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.15456},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - PMB-NN: Physiological-Model-Based Neural Network for Glucose Forecasting in Type 1 Diabetes
AU - Yaowen Zhang
AU - Shuhao Que
AU - Ying Wang
PY - 2026
DA - 2026/10/15
PB - arXiv
DO - 10.48550/arxiv.2610.15456
UR - https://arxiv.org/abs/2610.15456
AB - Accurate, physiologically consistent glucose forecasts could help people with type 1 diabetes respond to impending excursions. We applied our physiological-model-based neural network (PMB--NN) framework to glucose forecasting. A temporal convolutional network (TCN) used 180 min of history, constrained during training by a personalized four-parameter physiological model. The main PMB--NN combined forecast and neural--physiological agreement losses; a three-term variant additionally fitted physiological predictions to measured glucose. Data comprised 6--8 weeks of continuous glucose monitoring, pump, and meal records from 25 AZT1D participants. Person-specific models used chronological 70/10/20 train/validation/test splits and 30-, 60-, and 120-min horizons. Median mean absolute errors were 16.24, 25.00, and 30.20 mg/dL with basal insulin records and 15.38, 24.21, and 31.79 mg/dL before these records became available. In the latter setting, PMB--NN had the lowest median error at 30 and 60 min; it outperformed the physiological model alone in all six settings. In input-increase tests, glucose predictions changed in the expected direction for 85\% of correction boluses and 100\% of carbohydrate rescues, versus 57\% for active insulin and 56\% for active carbohydrates. Most standalone physiological fits showed locally distinguishable parameter effects, with glucose-related coefficients more stable under resampling than insulin- or carbohydrate-related coefficients. The three-term variant improved hyperglycaemia accuracy with basal records, but did not significantly change overall MAE and reduced directional plausibility in three tests. PMB--NN combines personalized forecasting with an explicit physiological model, allowing assessment of predictive accuracy, physiological response behavior, and participant-specific coefficients.
ER -