Automatic Sleep Scoring on SleepEDF
83.9AccuracyIITNet
Evaluation results
| Method | Links | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| IITNetArchitecture=CNN + RNN, Channel=Fpz-Cz, Input=Raw Signal, Approach=Many-to-One, Sequence Length=10 (9 past), Subjects=202019.02 | 83.9 | 77.6 | 0.78 | 87.7 | 43.4 | 87.7 | 86.7 | 82.5 | |
| IITNetArchitecture=CNN + RNN, Channel=Fpz-Cz, Input=Raw Signal, Approach=Many-to-One, Sequence Length=4 (3 past), Subjects=202019.02 | 83.6 | 76.5 | 0.77 | 87.1 | 39.2 | 87.7 | 87.7 | 80.9 | |
| SupratakArchitecture=CNN + RNN, Channel=Fpz-Cz, Input=Raw Signal, Approach=Many-to-One, Sequence Length=Whole previous night epochs, Subjects=202019.02 | 82 | 76.9 | 0.76 | 84.7 | 46.6 | 89.8 | 84.8 | 82.4 | |
| PhanArchitecture=Multitask 1-max CNN, Channel=Fpz-Cz, Input=Spectrogram, Approach=One-to-Many, Sequence Length=1, Subjects=202019.02 | 81.9 | 73.8 | 0.74 | — | — | — | — | — | |
| VilamalaArchitecture=CNN, Channel=Fpz-Cz, Input=Spectrogram, Approach=Many-to-One, Sequence Length=5 (2 past, future), Subjects=202019.02 | 81.3 | 76.5 | 0.74 | 80.9 | 47.4 | 86.2 | 86.2 | 81.9 | |
| IITNetArchitecture=CNN + RNN, Channel=Fpz-Cz, Input=Raw Signal, Approach=One-to-One, Sequence Length=1, Subjects=202019.02 | 80.6 | 72.1 | 0.73 | 84.7 | 29.8 | 86.3 | 87.1 | 72.8 | |
| TsinalisArchitecture=CNN, Channel=Fpz-Cz, Input=Raw Signal, Approach=Many-to-One, Sequence Length=5 (2 past, future), Subjects=202019.02 | 74.8 | 69.8 | 0.65 | 43.7 | 80.6 | 84.9 | 74.5 | 65.4 |