Atrial Fibrillation Detecting Software Gung Atrial Fibrillation Detecting Software

This study is a retrospective study, and the data is from the six hospitals of Chang Gung Medical Research Database (CGRD). We collected de-identified static 12-lead electrocardiogram (ECG) data from the database during the period of January 1, 2006, to December 31, 2019.

We created a training set and a testing set of ECG data from the CGRD. Then, we stratified and sampled ECG signals from the testing set according to the actual proportion to obtain the experimental sample.

The computer first preliminarily screened and selected ECG data that met the inclusion and exclusion criteria, and then numbered them sequentially. A cardiologist confirmed that the sampled ECG data did not include exclusion criteria.

The ECG data were converted into images and interpreted for the presence or absence of atrial fibrillation by three cardiologists. Their results were used as the gold standard (reference) for this study.

After determining the experimental standards, the ECG signals were inputted into the Chang Gung Atrial Fibrillation Detection software for analysis and interpretation of each ECG data.

After the software interpretation was completed, the results were compared with the interpretations of the physicians, and the primary and secondary evaluation indicators were analyzed accordingly.

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.

Exclusion criteria

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