Table 5 Comparison of our approach to the state of the art methods from other studies.
From: Novel AI driven approach to classify infant motor functions
Study | Classification | Acc. (%) | Sens. (%) | Spec. (%) |
|---|---|---|---|---|
Current study | FM+ vs. FM− | 88 | 88 | 88 |
Adde et al.54 | FM+ vs. FM− | 90 | 80 | |
Machireddy et al.55 | FM+ vs. FM− | 70 | ||
Tsuji et al.39 | Normal (WMs, FMs) vs. Abnormal (CS, PR) | 84.52 | ||
Adde et al.40 | CP vs. no-CP | 85 | 88 | |
Karch et al.33 | CP vs. no-CP | 90 | 96 | |
Philippi et al.56 | CP vs. no-CP | 90 | 95 | |
Orlandi et al.29 | CP vs. no-CP | 92.13 | ||
Ihlen et al.57 | CP vs. no-CP | 87 | 92.7 | 81.6 |
Meinecke et al.58 | Healthy vs. at-risk | 73 | 100 | 70 |
Heinze et al.59 | Healthy vs. pathologic | 89.66 | ||
Rahmati et al.60 | Healthy vs. affected | 87 | ||
Rahmati et al.61 | Healthy vs. affected | 91 | ||
Stahl et al.62 | Impaired vs. unimpaired | 93.7 | 85.3 | 95.5 |
Dai et al.63 | Normal vs. abnormal | 93.3 | 95 | 91.7 |
McCay et al.64 | Normal vs. abnormal (synthetic data) | 87.05 | ||
Raghuram et al.65 | Motor-impairment vs. no-motor-impairment | 66 | 95 | 95 |
Gao et al.66 | Typical development vs. abnormal movements | 79 | ||
Doroniewicz et al.67 | Normal WM vs. PR movements | 80.93 |
