Table 2 Out-of-fold model performance in the primary and secondary datasets.

From: A machine learning algorithm to differentiate bipolar disorder from major depressive disorder using an online mental health questionnaire and blood biomarker data

 

Baseline MDD

Baseline low mood

Baseline BD

 

126 BD vs. 187 MDD

98 BD vs. 112 MDD

98 BD vs. 120 low mood

45 BD

AUROC

0.92 (0.86–0.97)

0.89 (0.86–0.91)

0.90 (0.87–0.91)

NA

AUPRC

0.85 (0.73–0.95)

0.87 (0.81–0.90)

0.84 (0.80–0.88)

NA

Accuracy

0.83 (0.76–0.91)

0.80 (0.75–0.84)

0.82 (0.77–0.86)

NA

Sensitivity

0.84 (0.66–1.00)

0.77 (0.61–0.90)

0.77 (0.61–0.90)

0.86 (0.57–0.96)

Specificity

0.83 (0.65–0.95)

0.83 (0.76–0.92)

0.86 (0.77–0.94)

NA

PPV

0.77 (0.64–0.91)

0.80 (0.76–0.88)

0.83 (0.76–0.89)

NA

NPV

0.89 (0.79–1.00)

0.81 (0.72–0.90)

0.83 (0.74–0.91)

NA

  1. Values are shown as mean (95% confidence intervals).
  2. AUPRC area under the precision-recall curve, AUROC area under the receiver operating characteristic curve, BD bipolar disorder, MDD major depressive disorder, NA not applicable, NPV negative predictive value, PPV positive predictive value.
Back to article page