University of Guadalajara Uses Control Algorithms for Diabetes Care
Adaptive Controllers Simulate Artificial Pancreas Systems
“It is very easy to integrate control and/or identification systems for simulating an artificial pancreas for type 1 diabetes.”
Key Outcomes
- Robust Control for Real-Life Scenarios: The use of H∞ control theory ensures the artificial pancreas system remains effective under physiological uncertainties.
- MATLAB and Simulink Integration: These software tools streamline the design, simulation, and testing of control algorithms, accelerating innovation.
- Global and Regional Relevance: The project addresses a critical health issue in Latin America while contributing to the global advancement of diabetes care.
Continuous glucose monitoring with automated insulin delivery, delivered via an insulin pump, is still an important part of diabetes management.
At the heart of the University of Guadalajara’s Department of Exact Sciences and Engineering, a groundbreaking project is reshaping the future of diabetes management. Led by a seasoned professor-researcher with a Ph.D. in electrical engineering and a deep-rooted passion for public health, Dr. Eduardo Ruiz Velázquez is leading an initiative focused on developing advanced control algorithms for artificial pancreas systems and offering new hope for patients with type 1 diabetes. Alongside Dr. Eduardo Ruiz Velázquez are teammates Dr. Alejandro Ricardo Femat Flores, Dr. Griselda Quiroz Compeán, and Dr. Julio Alberto García Rodríguez.
The team’s primary goal is to mimic the glucose-insulin dynamics of a healthy individual, especially during carbohydrate intake. To achieve this, they rely on MATLAB® and Simulink® to model patient physiology, design robust controllers, and simulate real-life scenarios. Tools like Robust Control Toolbox™ and Symbolic Math Toolbox™ are instrumental in managing physiological variability, ensuring system reliability, calculating equilibrium points, linearizing mathematical models, and more.
The team also integrates the UVA Padova Simulator within Simulink, because according to Dr. Eduardo Ruiz Velázquez, “It is very easy to integrate control and/or identification systems for simulating an artificial pancreas for type 1 diabetes.”
The early results are promising. The control algorithms developed are not only efficient and intuitive to simulate but also hold the potential to be integrated into commercial artificial pancreas systems. These systems, combining continuous glucose monitoring with automated insulin delivery, could reduce the risk of hyperglycemia and hypoglycemia and so transform diabetes care.
Looking ahead, the team is exploring hybrid control algorithms that adapt to various patient conditions, including exercise and circadian rhythms. According to Dr. Eduardo Ruiz Velázquez, “These algorithms, unlike traditional ones, recognize that a single controller is not sufficient for glucose regulation due to the multiple factors involved in glucose dynamics. Instead, they employ a combination of controllers, each suited for a specific scenario in the patient's condition.”
While this research is particularly vital for Latin America, where diabetes affects over 32 million people, it has global relevance. By leveraging the computational power of MATLAB, the team aims to develop low-cost, high-impact solutions for one of the most common health afflictions worldwide. Their work exemplifies how academic innovation can drive global health improvements.
