Skip to main content

Table 3 Machine learning algorithm compared with other published methods

From: Discrimination of indoor versus outdoor environmental state with machine learning algorithms in myopia observational studies

Author

Method

Cut point

AUC

Sensitivity

Specificity

Youden Index

Tandon [20]

Luminance segmentation method

110 lx

0.82

74%

86%

0.60

Tandon [20]

GPS segmentation method

250 SNRa

0.89

82%

88%

0.70

Flynn [21]

Luminance segmentation method

240 lx

0.96

92%

90%

0.82

Dharani [19]

Luminance segmentation method

1000 lx

–

–

–

–

In our study

SVM Machine learning algorithm

–

0.99

99%

99%

0.99

  1. aSignal-to-noise ratio (SNR); Area under the ROC curve (AUC)