Skip to content

Research on quality control of ECG based on multi-scale cross-domain transfer learning

Research on quality control of ECG based on multi-scale cross-domain transfer learning

Status
Active, not recruiting
Phases
Unknown
Study type
Observational
Source
ChiCTR
Registry ID
ChiCTR2400088271
Enrollment
Unknown
Registered
2024-08-14
Start date
2024-09-01
Completion date
Unknown
Last updated
2024-08-19

For informational purposes only — not medical advice. Sourced from public registries and may not reflect the latest updates. Terms

Conditions

None

Interventions

ECG Diagnostic Department ECG Data:None

Sponsors

The Second Affiliated Hospital of Shanxi Medical University
Lead Sponsor

Eligibility

Sex/Gender
All

Inclusion criteria

Inclusion criteria: 1. The electrocardiogram data included in this study were selected from the past 5 years from March 2019 to present. Data from patients of all age groups can be included in this study. 2. All patients who undergo routine 12 lead electrocardiogram examination should have complete recorded electrocardiogram data (each lead recording time should not be less than 10 seconds), and electrocardiogram data that has not been severely disturbed can be included in this study.

Exclusion criteria

Exclusion criteria: 1) Signals with incomplete data recording or severe interference; 2) ECG data that cannot be accurately labeled manually.

Design outcomes

Primary

MeasureTime frame
Average recognition accuracy of ECG abnormality;The detection rate of error diagnosis samples in ECG quality control;Detection rate of abnormal electrocardiogram;

Secondary

MeasureTime frame
The inference time of a single ECG sample ;

Countries

China

Contacts

Public ContactWang Hongyu

The Second Affiliated Hospital of Shanxi Medical University

wanghongyu@sxmu.edu.cn+86 139 0341 0058

Outcome results

None listed

Source: ChiCTR (via WHO ICTRP) · Data processed: Feb 4, 2026