Applying an Artificial Intelligence-Enabled Electrocardiographic System for Reducing Mortality

NCT05118035 | Completed

• 1 other identifier: NDMC2021005
• Study Type: interventional
• Target: 15,965
• Locations: 1 country
• Sites: 1

Brief Summary

This is a randomized controlled trial (RCT) to test a novel artificial intelligence (AI)-enabled electrocardiogram (ECG)-based screening tool for early detection of clinical deterioration for reducing mortality.

Conditions

Cardiovascular Diseases; Intensity Care; Morality

Outcome Measures

Primary Outcomes (1)

• All cause mortality (death)

  After performing an electrocardiogram, the patient's survival is tracked.

  Within 90 days

Secondary Outcomes (4)

• Cardiovascular cause mortality (death)

  Within 90 days

• Arrhythmia medication

  Within 12 hours

• Electrolyte examination

  Within 3 days

• Cadiac examination

  Within 3-7 days

Study Arms (2)

Intervention

EXPERIMENTAL

Patients randomized to intervention will have access to the screening tool. Once the AI-ECG indicates high risk of mortality, a warning message would be immediately triggered and sent to the corresponding attending physicians. Notifications appear in the recipient's smartphone message system for the prompt attention. The message notified the physician that, "An ECG was received for patient X. An ECG indicates high risk of mortality. Please intensively attend to patient's conditions. If the physicians need to further identify the ECG, click on the following link to connect the ECG and the result of AI-ECG prediction." Of note, although we will actively send a warning message for high risk cases, the AI-ECG report for low risk cases still presented the degree of risk. Physicians can check the relative severity by access EHR for patients in the intervention group.

Other: AI-enabled ECG-based Screening Tool

Control

NO INTERVENTION

Patients will continue routine practice.

Interventions

AI-enabled ECG-based Screening Tool OTHER

Primary care clinicians in the intervention group had access to the report, which shows the risk prediction results for each patients. Moreover, the clinicians will recieve a short message when patients with a high risk ECG identified by AI.

Intervention

Eligibility Criteria

• Age: 18 Years+
• Sex: all
• Healthy Volunteers: No
• Age Groups: Adult (18-64), Older Adult (65+)

You may qualify if:

• Patients in emergency department or inpatient department.
• Patients recieved at least 1 ECG examination.

You may not qualify if:

• The patients recieved ECG at the period of inactive AI-ECG system.

Sponsors & Collaborators

• National Defense Medical Center, Taiwan: lead

Study Sites (1)

National Defense Medical Center

Taipei, 114, Taiwan

Location

Related Publications (2)

• Hsieh PH, Lin C, Lin CS, Liu WT, Lin TK, Tsai DJ, Hung YJ, Chen YH, Lin CY, Lin SH, Tsai CS. Economic analysis of an AI-enabled ECG alert system: impact on mortality outcomes from a pragmatic randomized trial. NPJ Digit Med. 2025 Jun 11;8(1):348. doi: 10.1038/s41746-025-01735-7.

  PMID: 40494963 DERIVED

• Lin CS, Liu WT, Tsai DJ, Lou YS, Chang CH, Lee CC, Fang WH, Wang CC, Chen YY, Lin WS, Cheng CC, Lee CC, Wang CH, Tsai CS, Lin SH, Lin C. AI-enabled electrocardiography alert intervention and all-cause mortality: a pragmatic randomized clinical trial. Nat Med. 2024 May;30(5):1461-1470. doi: 10.1038/s41591-024-02961-4. Epub 2024 Apr 29.

  PMID: 38684860 DERIVED

MeSH Terms

Conditions

Cardiovascular Diseases

Study Design

• Study Type: interventional
• Phase: not applicable
• Allocation: RANDOMIZED
• Masking: SINGLE
• Who Masked: PARTICIPANT
• Purpose: SCREENING
• Intervention Model: PARALLEL
• Sponsor Type: OTHER
• Responsible Party: PRINCIPAL INVESTIGATOR
• PI Title: Associate Professor

Model Details: intervention group:8001 control group:7964

Study Record Dates

• First Submitted: November 1, 2021
• First Posted: November 11, 2021
• Study Start: December 15, 2021
• Primary Completion: April 30, 2022
• Study Completion: December 31, 2022
• Last Updated: February 8, 2023

Record last verified: 2023-02

Data Sharing

• IPD Sharing: Will not share

Locations

TW (1)