Performance and Accuracy of an AI Enhanced Smart Watch Single Lead ECG (PROCESS)

PeRfOrmance and ACcuracy of an artifiCial Intelligence Enhanced Smart Watch Single Lead ECG (PROCESS)

The purpose of this study is to show the artificial intelligence enhanced single-lead ECG Apple Watch has similar, robust performance comparable to an AI enhanced 12 lead ECG and AI enhanced single lead (LI) of a 12 lead ECG.

Study Overview

• Status: Completed
• Conditions: Healthy

Detailed Description

1. Ambulatory patients undergoing ECG recording in the Mayo Clinic outpatient ECG lab will be asked to consent for this study.
2. Those who consent for the study will be asked to record a ECG using a single-lead watch-based (Apple Watch series 5) recording at a visit for a clinically scheduled 12 lead ECG recording.
3. This watch-based ECG data will be recorded and analyzed in comparison to the near-simultaneously recorded outpatient 12 Lead ECG

• Study Type: Observational
• Enrollment (Actual): 1000

Contacts and Locations

Study Locations

• United States
  • Minnesota
    • Rochester, Minnesota, United States, 55905
      • Mayo Clinic

Participation Criteria

Eligibility Criteria

• Ages Eligible for Study: 18 years to 89 years (Adult, Older Adult)
• Accepts Healthy Volunteers: No
• Genders Eligible for Study: All

• Sampling Method: Non-Probability Sample

Study Population

Ambulatory patients who are undergoing a clinically ordered electrocardiogram in the Mayo Clinic outpatient ECG lab.

Description

Inclusion Criteria:

• Age ≥ 18 years and ≤ 89.
• Able to give verbal consent.
• Able to complete routine clinical 12 lead ECG tracing and single lead Apple Watch ECG tracing.

Exclusion Criteria:

• Individuals < 18 and > 89 years of age.
• Unable to given verbal consent.

Study Plan

How is the study designed?

Design Details

• Observational Models: Cohort
• Time Perspectives: Prospective

Number of groups / cohorts

1

Cohorts and Interventions

Group / Cohort
Ambulatory, outpatient Mayo Clinic ECG lab patients; Patients who are undergoing routine clinical evaluation with a 12 lead ECG recording ordered at the Mayo Clinic ECG lab.

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Comparison of 12 lead ECG features to single-lead watch-based ECG features	The ECG interval differences (in milliseconds) between 12 Lead and collected single-lead watch-based ECG for PR, QRS, QT intervals will be determined and compared for each patient.	12 months
Arrhythmia comparison of 12 lead ECG to single-lead watch-based ECG	A physician interpretation of patients' 12 lead ECG and single-lead watch-based ECG will be performed to determined underlying rhythm (i.e. sinus rhythm, atrial fibrillation etc) from each, and the results from these modalities will be compared.	12 months

Secondary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Arrhythmia classification by physician overread of single-lead watch-based ECG	A physician interpretation of the patient's single-lead watch-based ECG will occur as described in "Outcome 2." The results of this ECG interpretation (i.e. sinus rhythm, atrial fibrillation, or inconclusive) will be compared to the watch/app-based rhythm auto-classification for each recorded single-lead watch-based ECG.	12 months

Other Outcome Measures

Outcome Measure	Measure Description	Time Frame
Artificial intelligence detection of heart failure by single-lead watch-based ECG	A previously developed AI algorithm to predict potential underlying cardiac pathology assess from 12 lead ECG via convolutional neural network will be adapted applied to the ECGs for patients who undergo single-lead watch-based ECG recording. This neural network uses PQRST complexes to yield a probability of heart failure which may not be readily apparent via manual review. Each recorded single-lead watch-based ECG will undergo evaluation by this neural network and will produce a probability of heart failure (0-100%) for each individual patient. This probability will be compared to the AI ECG result (probability 0-100%) from the patient's recently recorded 12 lead ECG which is routinely available for all patients with a recorded 12 lead ECG at our medical system.	12 months
Artificial intelligence detection of silent/paroxysmal atrial fibrillation by single-lead watch-based ECG	A previously developed AI algorithm to predict potential underlying cardiac pathology assess from 12 lead ECG via convolutional neural network will be adapted applied to the ECGs for patients who undergo single-lead watch-based ECG recording. This neural network uses PQRST complexes to yield a probability of silent/paroxysmal atrial fibrillation which may not be readily apparent via manual review. Each recorded single-lead watch-based ECG will undergo evaluation by this neural network and will produce a probability of silent/paroxysmal atrial fibrillation (0-100%) for each individual patient. This probability will be compared to the AI ECG result (probability 0-100%) from the patient's recently recorded 12 lead ECG which is routinely available for all patients with a recorded 12 lead ECG at our medical system.	12 months
Artificial intelligence detection of aortic stenosis by single-lead watch-based ECG	A previously developed AI algorithm to predict potential underlying cardiac pathology assess from 12 lead ECG via convolutional neural network will be adapted applied to the ECGs for patients who undergo single-lead watch-based ECG recording. This neural network uses PQRST complexes to yield a probability of aortic stenosis which may not be readily apparent via manual review. Each recorded single-lead watch-based ECG will undergo evaluation by this neural network and will produce a probability of aortic stenosis (0-100%) for each individual patient. This probability will be compared to the AI ECG result (probability 0-100%) from the patient's recently recorded 12 lead ECG which is routinely available for all patients with a recorded 12 lead ECG at our medical system.	12 months
Artificial intelligence determination of patient age by single-lead watch-based ECG	A previously developed AI algorithm to predict patient age from 12 lead ECG via convolutional neural network will be adapted applied to the ECGs for patients who undergo single-lead watch-based ECG recording. This neural network uses PQRST complexes to yield an ECG-predicted age. Each recorded single-lead watch-based ECG will undergo evaluation by this neural network and determine "ECG age" for each individual patient. This single-lead "ECG age" will be compared to the AI ECG "age" result determined from the patient's recently recorded 12 lead ECG which is routinely available for all patients with a recorded 12 lead ECG at our medical system.	12 months
Artificial intelligence detection of amyloidosis by single-lead watch-based ECG	A previously developed AI algorithm to predict potential underlying cardiac pathology assess from 12 lead ECG via convolutional neural network will be adapted applied to the ECGs for patients who undergo single-lead watch-based ECG recording. This neural network uses PQRST complexes to yield a probability of amyloidosis which may not be readily apparent via manual review. Each recorded single-lead watch-based ECG will undergo evaluation by this neural network and will produce a probability of amyloidosis (0-100%) for each individual patient. This probability will be compared to the AI ECG result (probability 0-100%) from the patient's recently recorded 12 lead ECG which is routinely available for all patients with a recorded 12 lead ECG at our medical system.	12 months

Collaborators and Investigators

• Sponsor: Mayo Clinic
• Investigators: Principal Investigator: Itzhak Zachi Attia, PhD, Mayo Clinic

Publications and helpful links

Helpful Links

• Mayo Clinic Clinical Trials

Study record dates

Study Major Dates

• Study Start (Actual): November 5, 2021
• Primary Completion (Actual): November 2, 2022
• Study Completion (Actual): November 2, 2022

Study Registration Dates

• First Submitted: July 5, 2022
• First Submitted That Met QC Criteria: July 5, 2022
• First Posted (Actual): July 11, 2022

Study Record Updates

• Last Update Posted (Estimate): December 28, 2022
• Last Update Submitted That Met QC Criteria: December 27, 2022
• Last Verified: December 1, 2022

More Information

Terms related to this study

• Keywords: wearable; electrocardiogram; artificial intellegence; Apple Watch
• Other Study ID Numbers: 21-006770

Plan for Individual participant data (IPD)

• Plan to Share Individual Participant Data (IPD)?: No
• IPD Plan Description: Due to patient confidentiality and IRB rules, we will not make individual patient data available

Drug and device information, study documents

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