Closed WangMengxiao319 closed 4 months ago
Hi @WangMengxiao319 , I'm very sorry for my late reply!
For demonstration, we carried out such a comparable analysis for IVCD in order to understand the comparably weak classification performance on the particular statement compared to other conduction disturbances. Indeed, clustering the model’s output probabilities with k-means clustering revealed two clusters, where one cluster performed much better than the other as can be seen in Fig. 5. Interestingly, it turned out that the two clusters largely align with the presence/absence of NORM as additional ECG statement.
As already stated in the paper, we selected the samples with IVCD
as part of the respective label-set (remember that the labels are multi-label). For this samples, we computed output probabilities (piece-wise sigmoids), which we clustered with K-Means. Each cluster was evaluated separately, revealing clusters with higher and lower errors correlated with the co-occurence of NORM
. It's really as simple as that, no further magic here.
Hi! Thanks for your elaborate work for ECG benchmarkd and dataset.
I wonder that what is the clustering results in Section IV-B (I mean what is high error cluster and low error cluster exactly? And how to get Fig5? Looking forward to your reply!