Skip to main content
. Author manuscript; available in PMC: 2023 May 1.
Published in final edited form as: Ophthalmology. 2022 Jan 3;129(5):571–584. doi: 10.1016/j.ophtha.2021.12.017

Table 3.

Performance of the Deep Learning Models on the Full Test Set

NS CLO PSC CLO-Center PSC-Center
MSE: mean (SD) 0.23 (0.01) 13.1 (1.6) 16.6 (2.4) 53.7 (4.9) 51.9 (6.5)

For each variable, 10 deep learning models were trained and tested separately.

CLO = cortical lens opacity; MSE = mean squared error; NS = nuclear sclerotic cataract; PSC = posterior subcapsular cataract; SD = standard deviation.