Trial Review

The safety and scientific validity of this study is the responsibility of the study sponsor and investigators. Listing a study does not mean it has been endorsed by the ANZCTR. Before participating in a study, talk to your health care provider and refer to this information for consumers
Trial registered on ANZCTR


Registration number
ACTRN12617001374358
Ethics application status
Approved
Date submitted
31/08/2017
Date registered
27/09/2017
Date last updated
29/10/2018
Date data sharing statement initially provided
29/10/2018
Type of registration
Prospectively registered

Titles & IDs
Public title
Continuous monitoring of heart rhythm with a wearable device for detection of atrial fibrillation.
Scientific title
Development and validation of a novel, wearable continuous ECG monitor for automated detection of atrial fibrillation in patients attending a hypertension outpatient clinic.
Secondary ID [1] 291987 0
DP03246
Universal Trial Number (UTN)
Trial acronym
Linked study record

Health condition
Health condition(s) or problem(s) studied:
Atrial fibrillation 303351 0
Cardiac arrhythmia 303352 0
Hypertension 303353 0
Stroke 303354 0
Condition category
Condition code
Cardiovascular 302779 302779 0 0
Hypertension
Stroke 302780 302780 0 0
Ischaemic
Cardiovascular 302781 302781 0 0
Other cardiovascular diseases

Intervention/exposure
Study type
Observational
Patient registry
False
Target follow-up duration
Target follow-up type
Description of intervention(s) / exposure
Participants will be exposed to 24 hours of monitoring with a wearable ECG device.

The Alerte Digital Health ECG device is a small wrist watched sized device with two electrodes attached to their body. The device comprises of circuitry allowing live streaming to a smartphone via Bluetooth, and a rechargeable battery with approximate life of 24 hours. It has been developed to provide automated continuous, real time ECG interpretation with artificial intelligence and it has been configured for this study to focus on automated detection of atrial fibrillation.

The device has previously been validated for safety and accuracy in a laboratory environment, and subsequently in a clinical environment compared against a standard ward based hospital monitoring system.
Intervention code [1] 298117 0
Early Detection / Screening
Comparator / control treatment
No control group
Control group
Uncontrolled

Outcomes
Primary outcome [1] 302169 0
Our goal is to produce automated continuous, real time ECG interpretation with artificial intelligence (AI), for detection of atrial fibrillation, utilising our wristwatch-sized ECG machine.
Timepoint [1] 302169 0
Patients will be observed with 24 hours of continuous monitoring.
Secondary outcome [1] 335049 0
To document the frequency and duration of atrial fibrillation, identified by the artificial intelligence, by applying the artificial intelligence to the ECG data acquired by the wristwatch sized ECG acquisition device. The ECG strips showing both sinus rhythm and atrial fibrillation will be independently adjudicated by a cardiologist blinded to the diagnosis made by the artificial intelligence. Accuracy of the AI system will be determined by comparing the AF and sinus rhythm diagnosis from the AI system against the cardiologist gold standard.
Timepoint [1] 335049 0
Following 24 hours of monitoring.
Secondary outcome [2] 335050 0
To document the frequency and duration of other arrhythmias, specifically supraventricular arrhythmias but not atrial fibrillation (including atrial flutter), ventricular arrhythmias (including ventricular ectopics), and artifact identified by the artificial intelligence. This will be measured by applying the artificial intelligence to the ECG data acquired by the wristwatch sized ECG acquisition device. The ECG strips showing supraventricular arrhythmias, ventricular arrhythmias and artifact will be independently adjudicated by a cardiologist blinded to the diagnosis made by the artificial intelligence. Accuracy of the AI system will be determined by comparing the atrial and ventricular arrhythmias and artifact diagnosis from the AI system against the cardiologist gold standard.
Timepoint [2] 335050 0
Following 24 hours of monitoring.
Secondary outcome [3] 335051 0
To document any discrepancy between the cardiologist scored ECG and the AI analysis. This data will be used to improve the training of the AI system.
Timepoint [3] 335051 0
Following 24 hours of monitoring.
Secondary outcome [4] 338567 0
The exact AF burden (paroxysmal and persistent) in patients wearing the device, helping identify those in need of further investigation and/or anticoagulant therapy
Timepoint [4] 338567 0
After 24 hours of monitoring
Secondary outcome [5] 338568 0
The heart rate in those with known persistent AF, to examine methods of using AI to devise a better surveillance strategy to detect (or prevent) early heart failure
Timepoint [5] 338568 0
After 24 hours of monitoring
Secondary outcome [6] 338569 0
Levels of anxiety / stress reported by patients through a questionnaire that will be used to determine the heart rate variability in response to stress.
Timepoint [6] 338569 0
Before the 24 hours of monitoring, participants will be asked to complete a standardised medical questionnaire (DASS-21) to determine levels of stress or anxiety in their life. They will also be asked to complete a diary to record sleep quality and activities whilst wearing the device.

Eligibility
Key inclusion criteria
Age over 18 and less than 80 years.
Ability to give informed consent in English without the need for a translator.
Attendance at the Royal Perth Hospital hypertension clinic.
Presence of sinus rhythm or atrial fibrillation.
Willingness to return devices to Royal Perth Hospital hypertension clinic.
Minimum age
18 Years
Maximum age
79 Years
Sex
Both males and females
Can healthy volunteers participate?
No
Key exclusion criteria
Permanent pacemaker or defibrillator.
Presence of arrhythmias other than atrial fibrillation on baseline ECG. (Presence of ventricular ectopy is not an exclusion criteria).
Dementia or severe physical dependence.
Inability to attend on two consecutive days to allow for return of the monitoring system, or other social circumstances in which an ECG system and smart phone may not be practical.

Study design
Purpose
Screening
Duration
Cross-sectional
Selection
Convenience sample
Timing
Prospective
Statistical methods / analysis

Recruitment
Recruitment status
Recruiting
Date of first participant enrolment
Anticipated
1/10/2017
Actual
10/04/2018
Date of last participant enrolment
Anticipated
1/12/2018
Actual
Date of last data collection
Anticipated
14/12/2018
Actual
Sample size
Target
50
Accrual to date
40
Final
Recruitment in Australia
Recruitment state(s)
WA
Recruitment hospital [1] 8043 0
Royal Perth Hospital - Perth
Recruitment postcode(s) [1] 16085 0
6000 - Perth

Funding & Sponsors
Funding source category [1] 296499 0
Commercial sector/Industry
Name [1] 296499 0
Alerte Digital Health
Country [1] 296499 0
Australia
Funding source category [2] 296510 0
Charities/Societies/Foundations
Name [2] 296510 0
Royal Perth Hospital Medical Research Foundation
Country [2] 296510 0
Australia
Primary sponsor type
Commercial sector/Industry
Name
Alerte Digital Health
Address
174 Reynolds Road
Mt Pleasant, Western Australia 6153
Country
Australia
Secondary sponsor category [1] 295458 0
None
Name [1] 295458 0
Address [1] 295458 0
Country [1] 295458 0

Ethics approval
Ethics application status
Approved
Ethics committee name [1] 297724 0
Royal Perth Hospital Human Research Ethics Committee
Ethics committee address [1] 297724 0
Ethics committee country [1] 297724 0
Australia
Date submitted for ethics approval [1] 297724 0
Approval date [1] 297724 0
26/05/2017
Ethics approval number [1] 297724 0

Summary
Brief summary
Our goal is to produce automated continuous, real time ECG interpretation with artificial intelligence (AI), utilising our wristwatch-sized ECG machine. Our initial AI system will be focused on automated detection of atrial fibrillation. We have already developed a prototype ECG detection device and trained our initial AI system, which has undergone validation testing in a clinical environment. This study is intended to further develop our ECG system to be suitable for more widespread use.
Trial website
Trial related presentations / publications
Public notes

Contacts
Principal investigator
Name 74938 0
Prof David Playford
Address 74938 0
University of Notre Dame, Fremantle
Professor of Cardiology, Mount Hospital
Suite 41
Mount Medical Centre
146 Mounts Bay Road
PERTH WA 6000
Country 74938 0
Australia
Phone 74938 0
+61 8 9485 0945
Fax 74938 0
Email 74938 0
[email protected]
Contact person for public queries
Name 74939 0
Prof David Playford
Address 74939 0
University of Notre Dame, Fremantle
Professor of Cardiology, Mount Hospital
Suite 41
Mount Medical Centre
146 Mounts Bay Road
PERTH WA 6000
Country 74939 0
Australia
Phone 74939 0
+61 8 9485 0945
Fax 74939 0
Email 74939 0
[email protected]
Contact person for scientific queries
Name 74940 0
Prof David Playford
Address 74940 0
University of Notre Dame, Fremantle
Professor of Cardiology, Mount Hospital
Suite 41
Mount Medical Centre
146 Mounts Bay Road
PERTH WA 6000
Country 74940 0
Australia
Phone 74940 0
+61 8 9485 0945
Fax 74940 0
Email 74940 0
[email protected]

Data sharing statement
Will individual participant data (IPD) for this trial be available (including data dictionaries)?
Yes
What data in particular will be shared?
The deidentified ECG data from each patient will be available on request.
When will data be available (start and end dates)?
January 2019
Available to whom?
Researchers requesting source data used to train the Alerte artificial intelligence system
Available for what types of analyses?
Verification of type and nature of arrhythmia present on ECG traces
How or where can data be obtained?
De-identified patient ECG file (single file, most likely in .edf format).


What supporting documents are/will be available?

No Supporting Document Provided



Results publications and other study-related documents

Documents added manually
No documents have been uploaded by study researchers.

Documents added automatically
No additional documents have been identified.