Research Article

A Cascaded Convolutional Neural Network for Assessing Signal Quality of Dynamic ECG

Table 6

Results per record on public dataset using cascaded CNN model.

RecordDominant rhythmAc (%)RecordDominant rhythmAc (%)RecordDominant rhythmAc (%)

100N94.7117N97.6212N95.3
101N96.4118N96.0213N, B96.7
102P93.6119N, B, T93.1214N, T91.8
103N96.2121N96.4215N92.4
104P95.3122N97.1217P, AF82.7
105N95.8123N96.4219N, AF79.3
106N, B93.6124N95.8220N96.7
107P94.2200N, B95.3221AF, T78.9
108N96.7201N, AF, T80.2222N, AFL, NOD, AB77.3
109N94.7202N, AF79.6223N, B, VT, T94.0
111N93.6203AFL, AF75.1228N, B93.6
112N93.1205N94.9230N, PREX94.4
113N92.9207N, VFL, SVTA, B, IVR77.6231N, BII95.3
114N94.9208N, T96.4232SBR94.0
115N94.2209N, SVTA90.4233N, B93.1
116N96.2210AF80.9234N, SVTA92.0

N, normal sinus rhythm; AB, atrial bigeminy; AFL, atrial flutter; AF, atrial fibrillation; B, bigeminy; BII, 2nd-degree heart block; IVR, idioventricular rhythm; NOD, nodal rhythm; P, paced rhythm; PREX, preexcitation (WPW); SBR, sinus bradycardia; SVTA, supraventricular tachyarrhythmia; T, ventricular trigeminy; VFL, ventricular flutter; VT, ventricular tachycardia.