Table 4 Experimental results for related deep learning methods for cancer classification with omics data.
Authors | Models | Pan-Cancer | Multi-Omics Data type | Accuracy | |||
|---|---|---|---|---|---|---|---|
mRNA | miRNA | DNA methylation | |||||
Mostavi et al., 202043 | 1D-CNN | 34 Classes | \(\:\surd\:\) | - | - | 95.50 ± 0.1 | |
2D-Vanilla-CNN | 94.87 ± 0.4 | ||||||
2D-Hybrid-CNN | 95.70 ± 1.0 | ||||||
Ramirez et al., 202044 | GCNN-PPI graph | 34 Classes | \(\:\surd\:\) | - | - | 88.98 ± 0.9 | |
GCNN-PPI + singleton graph | 94.61 ± 1.0 | ||||||
Kaczmarek et al., 202245 | GTN | 12 Classes | \(\:\surd\:\) | \(\:\surd\:\) | - | 93.56 ± 0.9 | |
Proposed MOGKAN-PPI graph | 32 Classes | \(\:\surd\:\) | \(\:\surd\:\) | \(\:\surd\:\) | 96.28 ± 0.0035 | ||
