LSTM and Reinforcement Learning Based Peak Analyzer for Electrocardiogram Characteristic Waves Delineation (performance results + annotated ECG record datasets)

Supplementary data containing three databases storing ECG records annotated by three different annotators in addition to the performance results of the proposed method.
The data comprise the following 6 files:
1. PDF file describing the content of the supplementary data, presenting a guide for reading the annotated ECG records, and discussing additional performance comparison of the proposed method with the related works.
2. Excel file presenting the performance of the reinforcement learning model in manipulating the peak analyzer using different episode-ending parameters.
3. Second Excel file describing in detail the selected training and validating datasets from the provided three ECG record databases aligned with the performance results of the proposed method.
4. Three SQLite database files consisting of:
4.a. 73 annotated ECG records selected from the ecgpuwave annotated records of the QTDB.
4.b. 71 annotated ECG records selected from the expert annotated records of the QTDB.
4.c. 160 of our own manually annotated records.