Ali Cinar, United States of America
Illinois Institute of Technology Chemical and Biological EngineeringPresenter of 2 Presentations
INCORPORATING PHYSICAL ACTIVITY AND STRESS ESTIMATES TO IMPROVE GLUCOSE PREDICTIONS FOR MULTIVARIABLE ARTIFICIAL PANCREAS SYSTEMS
- Ali Cinar, United States of America
- Mert Sevil, United States of America
- Iman Hajizadeh, United States of America
- Mudassir Rashid, United States of America
- Zacharie Maloney, United States of America
- Sediqeh Samadi, United States of America
- Rachel Brandt, United States of America
- Mohammad R. Askari, United States of America
- Nicole Hobbs, United States of America
- Minsun Park, United States of America
- Lauretta Quinn, United States of America
Abstract
Background and Aims
Maintaining glucose concentration (GC) in the target range in spite of physical activities and events causing acute psychological stress (PS) is challenging for artificial pancreas (AP) systems. PS and physical activities affect GC in different ways; aerobic exercise decreases GC while PS can increase it. Many stressful events and physical activities cannot be manually entered to the AP in a timely manner. Hence, they are unknown disturbances for AP systems and reduce glycemic control. In this work, wristband biosignals are utilized in a novel algorithm to estimate the psychological and physiological state of a subject, and improve the GC prediction for use in AP systems.
Methods
Biosignals from Empatica E4 wristband are collected in real-time and machine learning algorithms are utilized to determine the physical state of a subject, obtain energy expenditure estimates, and predict her/his PS levels. These estimates are incorporated in a GC prediction model along with CGM readings and insulin infusion data from pump. These BGC estimates are compared to estimates from a model that uses only CGM readings and insulin infusion data.
Results
Data from 50 experiments with thirteen different subjects with T1D who performed physical activities and various PS causing events were used and proposed method improved to mean absolute percentage error of GC prediction by 6.5%.
Conclusions
Wristband biosignals used to determine the psychological and physiological state of people with T1D provide valuable information to improve GC estimates and the performance of an AP system in response to unannounced physical activities and stressful events.
ADAPTIVE PERSONALIZED MULTIVARIABLE AP WITH PLASMA INSULIN CONCENTRATION ESTIMATION AND LEARNING OF UNANNOUNCED PHYSICAL ACTIVITIES AND MEALS
Abstract
Background and Aims
Most physical activities increase the risk of hypoglycemia for people with T1D. This study assesses the performance of our adaptive personalized multivariable artificial pancreas (AP-mAP) system without any exercises and meal unannouncements.
Methods
AP-mAP utilizes model predictive control, CGM readings, plasma insulin concentration (PIC) estimates and biosignals from a wristband. A patient’s historical data is leveraged to identify daily life activities using machine learning techniques. Incorporating online learning of probable times of significant glycemic disturbances from historical data improves AP performance by proactively mitigating the effects of impending disturbances. AP-mAP parameters such as controller set-point and PIC safety constraints are proactively modified for anticipated types and periods of disturbances.
Results
Simulations with the multivariable simulator mGIPsim illustrate the performance of AP-mAP. Twenty virtual subjects were simulated for 30 days with varying times and quantities of meals and different types, intensities, and durations of physical activities. AP-mAP reduced the 30-day average of total number of rescue carbohydrates from 20 to 6 without any hypoglycemia and improved glycemic control from 68.3% to 82.4% (in 70-180mg/dL range). One subject was simulated for 120 days with randomly varying times and quantities of meals and different types, intensities, and durations of physical activities (two exercise bouts per day). AP-mAP reduced the total number of rescue carbohydrates intakes for 120 days from 494 to 191 and kept good glycemic control (~83% in 70-180mg/dL range).
Conclusions
Integrating fully automated AP with machine leaning to adapt the AP and PIC during physical activities and meals improve diabetes management.