highly significantp < 0.001
matched samples t-test showed that there was no significant difference between the mean predicted SE by machine learning models and the mean cycloplegic SE ( p = 0.169 for SVR, p = 0.153 for RFR, p = 0.533 for ABR, for DNN p = 0.227), whereas the difference between the mean non-cycloplegic SE and the mean cycloplegic SE was highly significant ( p < 0.001).