Fig. 8: Estimation of Effect Sizes using Wilcoxon two-sample paired signed-rank test. | npj Digital Medicine

Fig. 8: Estimation of Effect Sizes using Wilcoxon two-sample paired signed-rank test.

From: Comparing decentralized machine learning and AI clinical models to local and centralized alternatives: a systematic review

Fig. 8: Estimation of Effect Sizes using Wilcoxon two-sample paired signed-rank test.The alternative text for this image may have been generated using AI.

The table presents effect sizes calculated using the Wilcoxon two-sample paired signed-rank test across seven performance metrics (AUROC, Accuracy, Precision/Positive Predictive Value, Sensitivity/Recall, F1 score, Specificity, and Dice score). Each metric is compared against both Local Learning and Centralized Learning approaches using the corresponding Decentrralized Learning values. Columns display the estimate of effect size, magnitude classification, and number of comparisons.

Back to article page