Table 3 Prediction results of the three different walking speed models of the XGBoost in the risk of falls.

From: XGBoost based machine learning approach to predict the risk of fall in older adults using gait outcomes

Models

AUC (95% CI)

Sensitivity (95% CI)

Specificity (95% CI)

PPV (95% CI)

NPV (95% CI)

PLR (95% CI)

NLR (95% CI)

Slower speed

0.71 (0.64–0.78)

0.43 (0.33–0.54)

0.84 (0.76–0.89)

0.63 (0.50–0.75)

0.69 (0.62–0.76)

2.65 (1.69–4.16)

0.68 (0.56–0.83)

Preferred speed

0.71 (0.64–0.78)

0.53 (0.42–0.64)

0.81 (0.73–0.87)

0.64 (0.52–0.75)

0.73 (0.65–0.80)

2.77 (1.87–4.12)

0.58 (0.45–0.73)

Faster speed

0.72 (0.66–0.79)

0.51 (0.40–0.62)

0.77 (0.69–0.84)

0.59 (0.47–0.70)

0.71 (0.63–0.78)

2.23 (1.54–3.23)

0.63 (0.53–0.83)

  1. AUCarea under the curve, NLR negative likelihood ratio, NPV negative predictive value, PLR positive likelihood ratio, PPV positive predictive value, CI confidence interval.
Back to article page