Sample size important for performance of GA prediction algorithm

NEW ORLEANS — Training sample size is an important factor in the performance of geographic atrophy prediction algorithms, according to research presented at the Association for Research in Vision and Ophthalmology meeting.
Robert Slater, PhD, and colleagues wrote that the availability of fundus autofluorescence limits the number of examples that can be used for artificial intelligence models in geographic atrophy (GA). In their study, they sought to determine the effect of sample size on the performance of AI models.
They developed an algorithm based on a training set of 1,515 images and

NEW ORLEANS — Training sample size is an important factor in the performance of geographic atrophy prediction algorithms, according to research presented at the Association for Research in Vision and Ophthalmology meeting.
Robert Slater, PhD, and colleagues wrote that the availability of fundus autofluorescence limits the number of examples that can be used for artificial intelligence models in geographic atrophy (GA). In their study, they sought to determine the effect of sample size on the performance of AI models.
They developed an algorithm based on a training set of 1,515 images and