R-DECO: an open-source Matlab based graphical user interface for the detection and correction of R-peaks

Figure 1. A normal heartbeat as recorded by an ECG. The QRS-complex can be observed in the center. The detection of this complex is crucial for almost all ECG analysis algorithms.

Figure 2. Enveloping procedure. The flattened ECG (F) is constructed by subtracting the lower envelope (L) from the upper envelope (U). This enhances the QRS-complex and flattens the rest of the ECG signal.

Figure 3. Procedure to select R-peaks. The resulting flat ECG denoted F is indicated by the black line. The samples with an amplitude lower than the sample 80 ms further are indicated by the magenta circles, with the last sample indicated by the green circle. The search window is indicated by the green line. The selected R-peaks are indicated by the black circles. A.u. stands for arbitrary units.

Figure 4. Performance of the algorithm, compared to the choice of window width. (A) The positive predictive value and (B) the sensitivity of the algorithm. A window width between 250 and 350 ms results in the best performance.

Figure 5. The graphical user interface of R-DECO. The user interface can be divided in five segments: (A) Data, (B) Filter, (C) Analysis Period, (D) R-peak Detection and (E) R-peak Correction. The ECG signal and the resulting tachogram are shown in respectively (F) and (G). The detected R-peaks are depicted as small blue circles.

Figure 6. Example of a high pass filter. By clicking the “Show Original” checkbox, the original signal (light blue) is overlaid on the filtered signal (blue).

Figure 7. The R-peak parameter selection window. The user can adjust the envelope size, the average heart rate and indicate if RR post-processing is necessary. Pressing the Default button restores the default values.

Figure 8. The template selection window. The user can select either the positive or “negative” R-peak template and can shift the location of the R-peak if necessary.

Figure 9. The data browsing options of R-DECO.

Figure 10. Result of the high pass filtering in the power spectrum.

Figure 11. Example of an nsVT segment with correction (A); the resulting tachogram is shown in (B).

Figure 12. Example of an nsVT segment with correction (A); the resulting tachogram is shown in (B).