Skip to main content
. 2013 Mar 8;8(3):e58695. doi: 10.1371/journal.pone.0058695

Table 3. RMSE for each VRQoL task and overall VRQoL score.

QoVL Task VA (worse eye) VA (better eye) MD MR Random Forest
‘Letters’ and ‘Sentences’ 0.98** 1.03** 0.94** 1.38** 0.84
‘Walking’ 0.38** 0.43** 0.34** 0.49** 0.30
‘Going out’ 0.40** 0.46** 0.42** 0.54** 0.32
‘Dining’ 0.37* 0.40** 0.38* 0.52** 0.33
Overall 2.49** 2.77** 2.42** 3.38** 1.97

Each value is calculated as the absolute difference between predicted VRQoL score and the actual VRQoL score in the testing dataset in the leave-one-out cross validation. RMSE: root mean of the squared prediction error, VRQoL: vision-related quality of life, VA (worse-eye): visual acuity of the eye with worse mean deviation (MD), VA (better-eye): visual acuity of the eye with better MD, MR: Multiple regression with VA (better-eye), VA (worse-eye) and MD of IVF (Integrated Visual Field) (**: p<0.05, *: p<0.01, in comparison with Random Forest, ten fold cross validation).