Additional Signals for Exercise, Stress and Sleep and Prediction of Glucose Levels for AP Systems

Type1 Diabetes Mellitus Clinical Trial

Official title:

Additional Signals for Detection of Exercise, Stress and Sleep Effects and Prediction of Glucose Levels for Next Generation AP Systems

Clinical Trial Details — Status: Completed

Administrative data

• NCT number: NCT04725799
• Other study ID #: 2017-1016
• Status: Completed
• Start date: May 24, 2018
• Est. completion date: July 13, 2021

Study information

• Verified date: March 2022
• Source: University of Illinois at Chicago
• Contact: n/a
• Is FDA regulated: No
• Study type: Observational

Clinical Trial Summary

The objective of this research is to determine the most informative variables for detecting exercise, acute stress and sleep, identify select sensors that report these variables, and develop the algorithms to detect the occurrence of exercise, stress and sleep, to discriminate them and to determine their characteristics. Research is needed to identify which wearable devices report the most informative and predictive variables of exercise, acute stress and sleep with desired precision and accuracy, determine the best location to wear them for collecting reliable and informative data, and to distill accurate knowledge from data reported by wearable sensors. Data and their interpretation should be informative for various types of physical activities, stages of sleep, and types and intensities of acute stress, and concurrent occurrence of these factors. The investigators will use several devices (chest band, wristband and skin patches) to collect data and evaluate their information content and contribution to improvement of glucose concentration prediction, best locations for collecting accurate and reliable information by conducting clinical and free-living experiments at-home to assess the contributions of the wearable device in improving the accuracy of glucose concentration prediction and the performance of the multivariable artificial pancreas.

Description:

The focus of the proposed work is to determine the most informative variables for meals, exercise, acute stress and sleep (MESS), identify select sensors that report these variables, and develop the algorithms to detect the occurrence of MESS, to discriminate them and to determine their characteristics. Research is needed to identify which wearable devices report the most informative and predictive variables of MESS with desired precision and accuracy, determine the best location to wear them for collecting reliable and informative data, and to distill accurate knowledge from data reported by wearable sensors. Data and their interpretation should be informative for various types of physical activities, stages of sleep, and types and intensities of acute stress, and concurrent occurrence of MESS factors. In the first year of the research, the investigators will use several devices to collect data and evaluate their information content and contribution to improvement of serum glucose prediction, best locations for collecting accurate and reliable information, and their acceptance by users. In the second year of the project, we will select a single wearable device, conduct additional clinical and free-living experiments at-home to assess the contributions of the information from the wearable device selected in improving the accuracy of glucose prediction and the performance of the multivariable AP. The objectives of the proposed research are to determine the most informative variables and sensor locations to capture reliable and accurate information that complement glucose concentration measurements (CGM) and to develop algorithms to determine the presence of exercise, stress or sleep at any given time for estimating glucose concentrations and making control decisions by a multivariable artificial pancreas system. The proposed research will be conducted by achieving the following specific aims: To systematically perturb the MESS factors within standardized experiments in the laboratory and at home for analyzing their effects and interactions on glucose concentrations; to identify the most informative and reliable measurement approaches (i.e. types of wearable sensors and locations on body) for discriminating MESS factors; to develop quantitative relations that identify the type and features of specific MESS activities, to develop algorithms to detect the presence of specific MESS activities, identify them, and predict glucose concentrations with recursive models in real time (for future integration with adaptive AP control systems); and to evaluate the performance of the multivariable models with data from wearable devices and CGM in closed-loop control with AP during exploratory clinical experiments. The study will take place over a period of 3-4 weeks for a total of 6 study visits. The subjects will be continuously monitored using CGM (DEXCOM [US G5 PLATINUM]); 3 activity of the following activity monitor(s); (Empatica E4, BiostampRC™, Equivital™ Life Monitor) and Actigraph throughout the night. In addition, the subjects will use a sleep monitor (Zmachine® Insight & Insight+ Model DT-200 and Zephyr Bioharness) throughout the 3-week period. Subjects will come in for a study visit 3 times during the first week during which the investigators will measure the subjects' oxygen capacity and muscular strength during a peak exercise stress test (peak V02) and a test of maximal muscle strength (1-RM), respectively. The results of this testing will establish a baseline from which we can calculate the intensity of submaximal aerobic and resistance exercise bouts. The next 3 study visits will occur over the following 3-4 weeks. During this time the subjects will perform activities that produce physiological and psychological stress alone, in combination or proximal to each other. Women will have a urine pregnancy test before each day of data collection. In addition, the subjects will perform a number of physical activities at home that will be scheduled with the research assistant. The research assistant will telephone the subjects periodically and have the subjects perform stressful activities, such as mental challenges.

Recruitment information / eligibility

• Status: Completed
• Enrollment: 23
• Est. completion date: July 13, 2021
• Est. primary completion date: July 13, 2021
• Accepts healthy volunteers: No
• Gender: All
• Age group: 18 Years to 60 Years

Eligibility

Inclusion Criteria:

• Men and Women with T1DM
• 18-60 years of age
• Insulin pump user

Exclusion Criteria:

• Metabolic instability as evidenced by hospitalizations for diabetes or other diabetes-related complications (e.g., diabetic ketoacidosis and hypoglycemic seizures) within the preceding three months;
• Severe macrovascular disease, as evidenced by severe peripheral artery disease; history of myocardial infarction, heart failure, thromboembolic disease, or unstable angina; uncontrolled hypertension; abnormal resting EKG;
• Maximal exercise stress test with significant brady/tachy arrhythmia, ectopic beats, bundle branch block, or signs of acute ischemia;
• Severe microvascular disease as evidenced by history of vision-threatening proliferative or non-proliferative retinal disease; kidney disease;
• Any uncontrolled non-musculoskeletal condition that would limit the subject's ability to participate in the exercise program (e.g., chronic obstructive airways disease);
• Musculoskeletal conditions such as neurological or orthopedic conditions affecting lower limb strength and mobility (e.g., stroke; insensitive foot);
• Pregnancy;
• Documented medical condition or physical impairment that is judged by the health care practitioner to contraindicate exercise.

Related Conditions & MeSH terms

• Diabetes Mellitus, Type 1
• Type1 Diabetes Mellitus

Locations

Country	Name	City	State
United States	University of Illinois at Chicago	Chicago	Illinois

Sponsors (3)

Lead Sponsor	Collaborator
University of Illinois at Chicago	Illinois Institute of Technology, University of Chicago

Country where clinical trial is conducted

United States, 

References & Publications (5)

Bayrak ES, Turksoy K, Cinar A, Quinn L, Littlejohn E, Rollins D. Hypoglycemia early alarm systems based on recursive autoregressive partial least squares models. J Diabetes Sci Technol. 2013 Jan 1;7(1):206-14. — View Citation

Eren-Oruklu M, Cinar A, Rollins DK, Quinn L. Adaptive System Identification for Estimating Future Glucose Concentrations and Hypoglycemia Alarms. Automatica (Oxf). 2012 Aug;48(8):1892-1897. Epub 2012 Jun 22. — View Citation

Turksoy K, Bayrak ES, Quinn L, Littlejohn E, Rollins D, Cinar A. Hypoglycemia Early Alarm Systems Based On Multivariable Models. Ind Eng Chem Res. 2013 Sep 4;52(35). doi: 10.1021/ie3034015. — View Citation

Turksoy K, Roy A, Cinar A. Real-Time Model-Based Fault Detection of Continuous Glucose Sensor Measurements. IEEE Trans Biomed Eng. 2017 Jul;64(7):1437-1445. doi: 10.1109/TBME.2016.2535412. Epub 2016 Feb 25. — View Citation

Turksoy K, Samadi S, Feng J, Littlejohn E, Quinn L, Cinar A. Meal Detection in Patients With Type 1 Diabetes: A New Module for the Multivariable Adaptive Artificial Pancreas Control System. IEEE J Biomed Health Inform. 2016 Jan;20(1):47-54. doi: 10.1109/JBHI.2015.2446413. Epub 2015 Jun 16. — View Citation

Outcome

Type	Measure	Description	Time frame	Safety issue
Primary	Measurement of Glucose Concentration	• Glucose concentration (mg/dl) will be measured by: Continuous Glucose Monitor (DEXCOM [US G5 PLATINUM/G6]	3.5 years