A Study to Evaluate Accuracy and Validity of the Chang Gung ECG Abnormality Detection Software

"Chang Gung ECG Abnormality Detection Software" is a is an artificial intelligence medical signal analysis software that detect whether patients have abnormal ECG signals of 14 diseases by static 12-lead ECG. The 14 diseases were

• Long QT syndrome
• Sinus bradycardia
• Sinus Tachycardia
• Premature atrial complexes
• Premature ventricular complexes
• Atrial Flutter, Right bundle branch block
• Left bundle branch block
• Left Ventricular hypertrophy
• Anterior wall Myocardial Infarction
• Septal wall Myocardial Infarction
• Lateral wall Myocardial Infarction
• Inferior wall Myocardial Infarction
• Posterior wall Myocardial Infarction

The main purpose of this study is to verify whether "Chang Gung ECG Abnormality Detection Software" can correctly identify abnormal ECG signals among patients of 14 diseases. The interpretation standard is the consensus of 3 cardiologists. The results of the software analysis will be used to evaluate the performance of the primary and secondary evaluation indicators.

Study Overview

• Status: Enrolling by invitation
• Conditions: Myocardial Infarction; Left Bundle-Branch Block; Atrial Flutter; Left Ventricular Hypertrophy; Long QT Syndrome; Sinus Bradycardia; Right Bundle-Branch Block; Sinus Tachycardia; Premature Atrial Complexes; Premature Ventricular Complexes
• Intervention / Treatment: Drug: Chang Gung ECG Abnormality Detection Software

Detailed Description

Detailed procedure:

1. Sample source:

   This is a retrospective study, and the data comes from the Chang Gung Medical Research Database(CGRD) which was an database form 6 hospitals of Chang Gung Memorial hospital. We collected de-identified static 12-lead ECG data from the database during 2006.01.01~2019.12.31, and the length of the ECG was 10 seconds.

2. Sampling:

   In this experiment, the training dataset and the test dataset ECG were separated. Afterwards, the ECG signals are stratified according to the distribution as the test sample, and all abnormal ECG signals of 14 diseases will be independently sampled from the ECG database of the test set.

3. Confirmation criteria:

   The ECG data will be preliminarily screened and selected by the inclusion and exclusion criteria and compiled serial numbers. Then, a cardiologist confirms that the sampling results of the ECG data do not include the exclusion criteria again.

4. Physician interpretation:

   The ECG data will be converted into graphic files and submitted to 3 cardiologists for interpretation abnormal ECG signals of 14 related diseases. The results will be used as the standard of this study (Reference).

5. Software interpretation:

   After confirming the test standard, input the ECG signal into Chang Gung ECG Abnormality Detection Software to analyze abnormal ECG signals of 14 diseases and interpret each ECG data.

6. Statistical analysis:

After the software interpretation is completed, it will be compared with the results of the physician's interpretation and analyze the primary and secondary evaluation indicators.

• Study Type: Observational
• Enrollment (Estimated): 4306
• Phase: Not Applicable

Contacts and Locations

Study Locations

• Taiwan
  • Taoyuan city, Taiwan, 333
    • Chang Gung Memorial Hospital

Participation Criteria

Eligibility Criteria

• Ages Eligible for Study: Adult; Older Adult
• Accepts Healthy Volunteers: No

• Sampling Method: Probability Sample

Study Population

This is a retrospective study, and the data comes from the Chang Gung Medical Research Database(CGRD) which was an database form 6 hospitals of Chang Gung Memorial hospital. According to inclusion and exclusion criteria, de-identified static 12-lead ECG data from the database during 2006.01.01~2019.12.31 was collected, and the length of the ECG was 10 seconds.

Description

Inclusion Criteria:

• Equal or greater than twenty years old.
• Static 12-lead electrocardiogram of General Electric MUSE XML format file.
• The data comes from the static 12-lead electrocardiogram device of General Electric (model MAC5500).
• The electrocardiogram signal is 500 Hz.
• The Alternating current (AC) filter of the electrocardiogram signal is 60 Hz.
• The resource of original diagnosis was a cardiologist.

Exclusion Criteria:

• Cases used in the model development process.
• Lacks any electrode.
• Contain any electrode lacks a segment.

Study Plan

How is the study designed?

Number of groups / cohorts

1

Cohorts and Interventions

Group / Cohort	Intervention / Treatment
Software diagnosis; Software diagnosis with gold standard of 3 cardiologists' interpretation.	Drug: Chang Gung ECG Abnormality Detection Software; This device is expected to be used for the static 12-lead ECG to detect whether there are abnormal ECG signals related to diseases and outputs the results.; Other Names: CGMH-EAD-001

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Sensitivity and Specificity	The rate of test results that correctly indicate the presence and absence.	baseline

Secondary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Area Under the receiver operating characteristic Curve	A graphical plot that illustrates the diagnostic ability of a binary classifier system as its discrimination threshold is varied.	Baseline

Collaborators and Investigators

• Sponsor: Chang Gung Memorial Hospital
• Investigators: Study Chair: Chang-Fu Kuo, MD/Ph.D, Associate Professor and Director Division of Rheumatology

Publications and helpful links

General Publications

• Ribeiro AH, Ribeiro MH, Paixao GMM, Oliveira DM, Gomes PR, Canazart JA, Ferreira MPS, Andersson CR, Macfarlane PW, Meira W Jr, Schon TB, Ribeiro ALP. Automatic diagnosis of the 12-lead ECG using a deep neural network. Nat Commun. 2020 Apr 9;11(1):1760. doi: 10.1038/s41467-020-15432-4. Erratum In: Nat Commun. 2020 May 1;11(1):2227.
• Malik J, Devecioglu OC, Kiranyaz S, Ince T, Gabbouj M. Real-Time Patient-Specific ECG Classification by 1D Self-Operational Neural Networks. IEEE Trans Biomed Eng. 2022 May;69(5):1788-1801. doi: 10.1109/TBME.2021.3135622. Epub 2022 Apr 21.
• Acharya U.R., Fujita H., Lih O.S., Adam M., Tan J.H., Chua C.K. Automated detection of coronary artery disease using different durations of ECG segments with convolutional neu-ral network Knowl.-Based Syst., 132 (sep.15) (2017), pp. 62-71
• Bos JM, Attia ZI, Albert DE, Noseworthy PA, Friedman PA, Ackerman MJ. Use of Artificial Intelligence and Deep Neural Networks in Evaluation of Patients With Electrocardiographically Concealed Long QT Syndrome From the Surface 12-Lead Electrocardiogram. JAMA Cardiol. 2021 May 1;6(5):532-538. doi: 10.1001/jamacardio.2020.7422.
• U. Rajendra Acharya, Hamido Fujita, Oh Shu Lih, Yuki Hagiwara, Jen Hong Tan, Muhammad Adam, Automated detection of arrhythmias using different intervals of tachycardia ECG seg-ments with convolutional neural network, Information Sciences, Volume 405, 2017, Pages 81-90, ISSN 0020-0255
• Jeong DU, Lim KM. Convolutional neural network for classification of eight types of arrhythmia using 2D time-frequency feature map from standard 12-lead electrocardiogram. Sci Rep. 2021 Oct 14;11(1):20396. doi: 10.1038/s41598-021-99975-6.
• Kwon JM, Jeon KH, Kim HM, Kim MJ, Lim SM, Kim KH, Song PS, Park J, Choi RK, Oh BH. Comparing the performance of artificial intelligence and conventional diagnosis criteria for detecting left ventricular hypertrophy using electrocardiography. Europace. 2020 Mar 1;22(3):412-419. doi: 10.1093/europace/euz324.
• Makimoto H, Hockmann M, Lin T, Glockner D, Gerguri S, Clasen L, Schmidt J, Assadi-Schmidt A, Bejinariu A, Muller P, Angendohr S, Babady M, Brinkmeyer C, Makimoto A, Kelm M. Performance of a convolutional neural network derived from an ECG database in recognizing myocardial infarction. Sci Rep. 2020 May 21;10(1):8445. doi: 10.1038/s41598-020-65105-x.

Study record dates

Study Major Dates

• Study Start (Actual): October 6, 2023
• Primary Completion (Estimated): January 31, 2024
• Study Completion (Estimated): February 28, 2024

Study Registration Dates

• First Submitted: June 6, 2023
• First Submitted That Met QC Criteria: June 6, 2023
• First Posted (Actual): June 15, 2023

Study Record Updates

• Last Update Posted (Actual): January 11, 2024
• Last Update Submitted That Met QC Criteria: January 9, 2024
• Last Verified: May 1, 2023

More Information

Terms related to this study

• Additional Relevant MeSH Terms: Ischemia; Pathologic Processes; Necrosis; Myocardial Ischemia; Heart Diseases; Cardiovascular Diseases; Vascular Diseases; Congenital Abnormalities; Pathological Conditions, Anatomical; Arrhythmias, Cardiac; Cardiac Conduction System Disease; Pregnancy Complications; Obstetric Labor Complications; Obstetric Labor, Premature; Infarction; Heart Defects, Congenital; Cardiovascular Abnormalities; Cardiomegaly; Cardiac Complexes, Premature; Tachycardia, Supraventricular; Female Urogenital Diseases and Pregnancy Complications; Urogenital Diseases; Myocardial Infarction; Bundle-Branch Block; Heart Block; Premature Birth; Bradycardia; Tachycardia; Hypertrophy; Atrial Flutter; Hypertrophy, Left Ventricular; Ventricular Premature Complexes; Long QT Syndrome; Atrial Premature Complexes; Tachycardia, Sinus
• Other Study ID Numbers: 202300710A5

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