Table 6 Performance comparisons between Freeze-Missing and SOTA models

From: A three-tier AI solution for equitable glaucoma diagnosis across China’s hierarchical healthcare system

Methods

Accuracy

Sensitivity

Specificity

AUC

MMD

0.7754 ± 0.0200

0.9135 ± 0.0189

0.4956 ± 0.0468

0.8366 ± 0.0178

0.8012

0.9258

0.5487

0.8485

Cheerla et al.

0.7908 ± 0.0134

0.8271 ± 0.0539

0.7172 ± 0.0710

0.8441 ± 0.0132

0.8128

0.9126

0.6106

0.8592

Freeze-Missing

0.7978 ± 0.0046

0.8419 ± 0.0146

0.7085 ± 0.0268

0.8551 ± 0.0110

0.8017

0.8428

0.7193

0.8650

  1. The upright numbers represent the means and standard deviations of the five-fold cross-validation results, with bolded values indicating the best performance. Italicized numbers represent the results of the best model during cross-validation. Values below 0.6 are underlined.
Back to article page