highly significantP < 0.0001
Our quantitative Z-scores showed a strong and highly significant correlation with the expert's clinical ratings (Spearman's ρ = 0.866, P < 0.0001).
Our quantitative Z-scores showed a strong and highly significant correlation with the expert's clinical ratings (Spearman's ρ = 0.866, P < 0.0001).
The 62-feature model showed a visible separation trend between ‘Good’ and ‘Poor’ recovery groups but failed to reach statistical significance ( P = 0.072), likely due to noise from non-informative features ( Figure 4B ).