Table 12 Comparison of findings based on the CHB-MIT dataset
From: A hybrid CNN-Bi-LSTM model with feature fusion for accurate epilepsy seizure detection
Reference | Year | Strategies | # of patients | Accuracy (%) |
|---|---|---|---|---|
[62] | 2021 | Time and frequency domain feature + fuzzy classifier | 7 | 96.48 |
[63] | 2020 | Channel-embedding spectral-temporal squeeze and excitation network + SVM | 21 | 95.96 |
[64] | 2022 | DWT + compatibility framework + CNN- Bi-LSTM-AM | 23 | 96.87 |
[65] | 2023 | Customized CNN + exhaustive random forest + RNN-Bi-LSTM | 24 | 98.00 |
[66] | 2024 | DWT + time–frequency domain features + LSTM-SNP | 23 | 98.25 |
Proposed Method | 2024 | DWT + (Time domain + Non-linear Features) + SVM-REF + CNN-Bi-LSTM | 23 | 98.43 |