Table 4 Comparison of classification performance between previous SOTA methods and ours MESEL on the PTB-XL dataset in the sub-diagnosis class-level

From: Multi-expert ensemble ECG diagnostic algorithm using mutually exclusive–symbiotic correlation between 254 hierarchical multiple labels

Methods

Year

Sub classes (23)

AUROC

AUPRC

ATICNN47

2020

88.2

-

DMSFNet48

2020

89.7

-

XResNet 1D 10149

2020

92.9

-

ASTLNet50

2023

93.2

-

MSDNN20

2023

93.3

52.8

MESEL

ours

93.6

56.3

  1. The bold line represents the classification performance of the proposed MESEL, outperforming these methods in both AUROC and AUPRC metrics.
Back to article page