Compelling new electrocardiographic markers for automatic diagnosis

A unique cardiac electrocardiographic 3D model. Toward interpretable AI diagnosis

“…On the other hand, Table 1 also shows that and are directly associated with CLBBB and CRBBB, respectively, whereas does with HYP, as long as is within the percentile range of NORM. The interest of the FMM parameters in the diagnosis had already been evidenced in Rueda et al., 42 where simple diagnostic rules for CLBBB and CBBB (Complete Bundle Branch Block) have been defined, and can be used even when only one of the lead signals is available. To better illustrate the diagnostic ability of binary classifier systems based on these markers, ROC curves have been plotted in Figure 3 .…”

“… QRS Revision. The median length of the fragment is used as benchmark to remove too short/long detections 40 , 42 , 46 preventing false positive detections. is the time elapsed between two consecutive detections.…”

“…As it was done in Rueda et al., 42 the heartbeat segmentation across leads is given by . ECG Segmentation Revision.…”