A physiologically informed mathematical model linking nocturnal glucose levels with resting heart rate dynamics provides a non-invasive, interpretable tool for early T2DM risk stratification based on nocturnal physiological patterns in cancer survivors.
Key Findings
Methods
A coupled system of ordinary differential equations was developed to model the relationship between nocturnal glucose levels and resting heart rate dynamics in cancer survivors.
The model is described as 'physiologically informed' and links nocturnal glucose levels with resting heart rate dynamics.
The framework was designed to be non-invasive and interpretable.
The model targets early detection of T2DM risk, motivated by sleep-related alterations in glucose metabolism and autonomic regulation.
The model was fitted to wearable sensor data.
Methods
Particle swarm optimization was used to fit the mathematical model to wearable sensor data.
Particle swarm optimization was the parameter estimation method selected for model fitting.
The data used for fitting came from wearable sensors.
This approach enabled personalized parameter estimation for individual patients.
Results
Sensitivity analysis identified key parameters driving metabolic dysregulation, enabling formulation of a personalized risk score.
Sensitivity analysis was carried out to determine which model parameters most strongly influence metabolic dysregulation.
The results of sensitivity analysis directly informed the construction of a personalized risk score.
The risk score is intended for T2DM risk stratification in cancer survivors.
Results
The model was successfully validated using both synthetic patient data and real cancer survivor patient data.
Validation was performed on two data types: synthetic patient data and cancer survivor patient data.
The paper describes the validation as 'successful.'
Validation supported the framework's utility for risk stratification based on nocturnal physiological patterns.
Background
Sleep-related alterations in glucose metabolism and autonomic regulation play a critical role in the early development of type 2 diabetes mellitus, especially among cancer survivors.
The paper identifies this interplay as the primary motivation for developing the model.
Cancer survivors are highlighted as a specific at-risk population for T2DM development.
Nocturnal physiological patterns—specifically resting heart rate and glucose levels—are the focus of monitoring.
What This Means
This research suggests that mathematical modeling of nighttime body signals—specifically blood sugar levels and resting heart rate measured by wearable devices—can help predict which cancer survivors are at risk of developing type 2 diabetes. The researchers built a set of equations that captures how these two signals interact during sleep, then used a computer algorithm called particle swarm optimization to tune the model to individual patients' data. By analyzing which parts of the model had the biggest influence on unhealthy metabolic patterns, they created a personalized risk score for each patient.
The model was tested against both simulated patient data and real data from cancer survivors, and the results were described as successful. This matters because cancer survivors face elevated risks of metabolic problems, partly due to the effects of cancer treatment on the body's regulatory systems, and catching these problems early is difficult with traditional clinical tests.
This research suggests that continuous, wearable sensor data collected during sleep—a relatively non-invasive approach—could be combined with mathematical modeling to flag individuals who may be developing diabetes before obvious symptoms appear. If further validated in larger clinical studies, this kind of tool could allow doctors to intervene earlier and more precisely in cancer survivor populations.
Liao Y, Spadola C, Roy S. (2026). Mathematical modeling of glucose regulation and sleep resting heart rate to predict risk of metabolic dysregulation in cancer survivors.. PloS one. https://doi.org/10.1371/journal.pone.0339461