Table 4 Comparative results of different methods for comorbidity prediction
From: Representation learning for clinical time series prediction tasks in electronic health records
RNN-DAE | RNN-AE | SDAs | PCA | k-means | GMM | Hand | |
|---|---|---|---|---|---|---|---|
HD | 0.745(2) | 0.771(1) | 0.736(3) | 0.545(6) | 0.537(7) | 0.652(5) | 0.654(4) |
DM | 0.671(1) | 0.660(2) | 0.422(7) | 0.619(6) | 0.631(3) | 0.627(4.5) | 0.627(4.5) |
CAD | 0.744(2) | 0.746(1) | 0.609(6) | 0.601(7) | 0.617(5) | 0.741(3) | 0.740(4) |
AF | 0.743(1) | 0.522(6) | 0.537(4) | 0.535(5) | 0.404(7) | 0.645(2) | 0.644(3) |
CRD | 0.700(2) | 0.727(1) | 0.554(6) | 0.373(7) | 0.566(5) | 0.699(3) | 0.698(4) |
VHD | 0.586(5) | 0.842(3) | 0.601(4) | 0.258(7) | 0.500(6) | 0.882(2) | 0.902(1) |
DCM | 0.785(1) | 0.777(2) | 0.406(7) | 0.416(6) | 0.440(5) | 0.675(3) | 0.674(4) |
HCM | 0.718(2) | 0.814(1) | 0.201(7) | 0.222(6) | 0.396(5) | 0.438(3) | 0.437(4) |
COPD | 0.747(1) | 0.547(3) | 0.522(5) | 0.577(2) | 0.457(7) | 0.522(5) | 0.522(5) |
CID | 0.790(3) | 0.739(5) | 0.474(7) | 0.697(6) | 0.762(4) | 0.872(2) | 0.873(1) |
Avg.rank | 2.000 | 2.500 | 5.600 | 5.800 | 5.400 | 3.250 | 3.450 |