Table 2. Other feature extraction algorithms in spatial domain.
| A | M | P | D | S | DE | E |
|---|---|---|---|---|---|---|
| Wang et al. (2021) | CN, CSP BQPSO ELM |
Chinese characters “壹(one)” | Private data | 10 (7 m and 3 f/age from 22–28) | SynAmps 2 | Average: 85.4 |
| Huang et al. (2022) | STWFBCSP | Chinese characters “右(right)” | Private data | 12 (8 m and 4 f/age from 22–26) | N/A | Average: 84.87 |
| Alizadeh & Omranpour (2023) | EM-CSP | Words, English phonemes | 1. Kara One (Zhao & Rudzicz, 2015) | 12 (8 m and 4 f/age mean 27.4) | SynAmps RT | Highest: 98.47 |
| 2. Private data | ||||||
| Panachakel & Ramakrishnan (2022) | CSP | “in, cooperate” | ASU (Nguyen, Karavas & Artemiadis, 2017) | 15 (11 m and 4 f/age from 22–32) | BrainProducts | Highest: 85.2 |
| LSTM | ActiCHamp amplifier |
Note:
A, authors; M, methods; P, pronunciation materials; D, datasets; S, subjects (number); DE, device; E, evaluation indicators (accuracy: %); m, males; f, females; CN, causal network; CSP, common spatial pattern; BQPSO, binary quantum particle swarm optimization; ELM, Extreme Learning Machine; STWFBCSP, sub-time window filter bank common spatial pattern; EM-CSP, efficient-multiclass CSP; LSTM, long short-term memory; SI, speech imagery; MI, motor imagery.