AI Models for Non-invasive Glycaemic Event Detection Using ECG in Type 1 Diabetics

Development and Validation of Artificial Intelligence Models for Non-invasive Glycaemic Event Detection Using ECG in Type 1 Diabetics

This observational study aims to recruit up to thirty T1DM patients from a diabetic outpatient clinic at the University Hospital Coventry and Warwickshire for a two-phase study. The first phase involves attending an inpatient protocol for up to thirty-six hours in a calorimetry room at the Human Metabolism Research Unit under controlled conditions, followed by a phase of free-living, for up to three days, in which participants will go about their normal daily activities without restriction. Throughout the study, the participants will wear commercially available wearable sensors to measure and record physiological signals (e.g., electrocardiogram and continuous glucose monitor). Data collected will be used to develop and validate an AI model using state-of-the-art deep-learning methods for the purpose of non-invasive glycaemic event detection.

Study Overview

• Status: Not yet recruiting
• Conditions: Metabolic Disease

Detailed Description

The study volunteers will be asked to an attend an 'inpatient' facility for up to 36 hrs dedicated to advanced metabolic measurement (HMRU). They will be asked to consume prepared meals of varying macronutrient content as part of a balanced diet, and performed prescribed physical activity. During this time the volunteers will be measured by instrumentation which will investigate the chemical concentration in respired gases (e.g. whole-body calorimeters, metabolic carts); bloods, saliva and urine samples will be taken. If the participant then wishes, we will ask them to continue to wear the wearable devices in a home setting for a maximum one week.

The data derived from this study will allow new tools and mathematical models to be developed that can be used to analyse and simulate patient metabolic response. It is envisaged this study will give further evidence to support future research into glucose utilisation in diseased metabolic populations.

• Study Type: Observational
• Enrollment (Anticipated): 30

Contacts and Locations

• Study Contact: Name: John G Hattersley, PhD; Phone Number: +44 (0) 24 7696 6068; Email: john.hattersley@uhcw.nhs.uk
• Study Contact Backup: Name: Leandro Pechhia, PhD; Phone Number: +44 (0) 24 7657 3383; Email: L.Pecchia@warwick.ac.uk

Participation Criteria

Eligibility Criteria

• Ages Eligible for Study: 18 years and older (ADULT, OLDER_ADULT)
• Accepts Healthy Volunteers: Yes
• Genders Eligible for Study: All

• Sampling Method: Non-Probability Sample

Study Population

The population will be recruited from the Warwickshire Institute for the study of diabetes, Endocrinology and Metabolism, at the University Hospitals Coventry and Warwickshire. WISDEM is a flagship partnership between the hospitals and the University of Warwick Medical School created to tackle diabetes and related metabolic conditions.

Description

Inclusion Criteria:

The study will be open to all individuals living independently, over 18 years without acute illness or ongoing clinical investigation, or volunteers with a stable medical condition may be included. Volunteers with an ongoing medical condition will only be included after detailed consultation with our clinical and dietetics members of the team; however, it is imperative that volunteers are able to provide written informed consent.

Exclusion Criteria:

Whilst the study employs a deliberately open inclusion criterion, the following exclusion measures will be employed:

• Children (under 18 yrs)
• Any adult who lacks decisional capacity
• Claustrophobia, isolophobia, recent abnormal exercise, radiation exposure within the preceding 24 hours of entering the whole-body calorimeter and feeling unwell in any way.
• Needle phobia
• Any medical/endocrine problem that could affect energy expenditure (e.g. thyroid problems, Cushing's syndrome)
• Chronic inflammatory disorders like rheumatoid arthritis, or long term use of steroids or other immunomodulators like cyclosporine, azathioprine.
• Beta blockers
• Currently actively losing weight
• Depression or any psychiatric illness

Study Plan

How is the study designed?

Number of groups / cohorts

1

Cohorts and Interventions

Group / Cohort
Type1diabetes patients; Males and females diagnosed with T1D, aged over 18 years old who are currently under the care of the Warwickshire Institute for the Study of Diabetes, Endocrinolgy and Metabolism (WISDEM) at the University Hospitals Coventry and Warwickshire.

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Interstitial Glucose	As measured by a continuous glucose monitor [NOTE] Observational study thus a key measurement not a true outcome measure.	For the duration of the study, up to 5 days

Secondary Outcome Measures

Outcome Measure	Measure Description	Time Frame
ECG -Interval across different fiducial points	As measured by an ambulatory ECG device [NOTE] Observational study thus a key measurement not a true outcome measure. The interval across different fiducial points (P.Q.R,S,T) is one of the features that are useful to quantify the difference in ECG signals for different glycaemic events.	For the duration of the study, up to 5 days
ECG - Slope across different fiducial points	As measured by an ambulatory ECG device [NOTE] Observational study thus a key measurement not a true outcome measure. The Slope across different fiducial points (P.Q.R,S,T) is one of the features that are useful to quantify the difference in ECG signals for different glycaemic events.	For the duration of the study, up to 5 days
ECG - Indices of Heart Rate Variability	As measured by an ambulatory ECG device [NOTE] Observational study thus a key measurement not a true outcome measure. Heart rate variability (HRV) is the fluctuation in the time intervals between adjacent heartbeats. There are several indices that are useful to quantify the difference in ECG signals for different glycaemic events such as Ultra Low Frequency (ULF) (≤0.003 Hz), Very Low Frequency (VLF) (0.0033-0.04 Hz), Low Frequency (LF) (0.04-0.15 Hz) and High Frequency (HF) (0.15-0.4 Hz)	For the duration of the study, up to 5 days
Blood Pressure (Systolic and Diastolic)	As measured by an ambulatory blood pressure device [NOTE] Observational study thus a key measurement not a true outcome measure.	For the duration of the study, up to 5 days

Collaborators and Investigators

• Sponsor: University Hospitals Coventry and Warwickshire NHS Trust
• Collaborators: University of Warwick

Publications and helpful links

General Publications

• Porumb M, Stranges S, Pescape A, Pecchia L. Precision Medicine and Artificial Intelligence: A Pilot Study on Deep Learning for Hypoglycemic Events Detection based on ECG. Sci Rep. 2020 Jan 13;10(1):170. doi: 10.1038/s41598-019-56927-5.
• Porumb M, Griffen C, Hattersley J, Pecchia L. Nocturnal low glucose detection in healthy elderly from one-lead ECG using convolutional denoising autoencoders. Biomedical Signal Processing and Control. 2020;62:102054.

Study record dates

Study Major Dates

• Study Start (ANTICIPATED): September 30, 2022
• Primary Completion (ANTICIPATED): May 1, 2026
• Study Completion (ANTICIPATED): May 1, 2027

Study Registration Dates

• First Submitted: June 24, 2022
• First Submitted That Met QC Criteria: July 12, 2022
• First Posted (ACTUAL): July 15, 2022

Study Record Updates

• Last Update Posted (ACTUAL): July 15, 2022
• Last Update Submitted That Met QC Criteria: July 12, 2022
• Last Verified: July 1, 2022

More Information

Terms related to this study

• Keywords: Endochrine
• Additional Relevant MeSH Terms: Metabolic Diseases
• Other Study ID Numbers: JH206817a

Plan for Individual participant data (IPD)

• Plan to Share Individual Participant Data (IPD)?: NO

Drug and device information, study documents

• Studies a U.S. FDA-regulated drug product: No
• Studies a U.S. FDA-regulated device product: No