TY - JOUR
AU - Bressler, Neil M.
AB - 
Key PointsQuestionWhat is the association of having a very low number (“low-shot”) of training images with the performance of artificial intelligence algorithms for retinal diagnostics?
FindingsThis cross-sectional study found that performance degradation occurred when using traditional algorithms with low numbers of training images. When using only 160 training images per class, traditional approaches had an area under the curve of 0.6585; low-shot methods using contrastive self-supervision outperformed this with an area under the curve of 0.7467.
MeaningThese findings suggest that low-shot deep learning methods show promise for use in artificial intelligence retinal diagnostics and may be beneficial for situations involving much less training data, such as rare retinal diseases or addressing artificial intelligence bias.

TI - Low-Shot Deep Learning of Diabetic Retinopathy With Potential Applications to Address Artificial Intelligence Bias in Retinal Diagnostics and Rare Ophthalmic Diseases
JF - JAMA Ophthalmology
DO - 10.1001/jamaophthalmol.2020.3269
DA - 2020-10-03
UR - https://www.deepdyve.com/lp/american-medical-association/low-shot-deep-learning-of-diabetic-retinopathy-with-potential-OrYnCtAOqN
SP - 1070
EP - 1077
VL - 138
IS - 10
DP - DeepDyve
ER -