Statistical analysis and design in ophthalmology: Toward optimizing your data

Automatic detection of glaucoma via fundus imaging and artificial intelligence: A review

“…We use the term "image classification" to refer to the automated process of determining the category to which a given fundus image belongs e.g., healthy, or glaucomatous group (binary classification); or healthy, suspected glaucoma or glaucoma group (multi-class classification). This process is also referred to as image discrimination ( 23 ) or disease prediction. To achieve the classification, AI can apply a threshold to the estimated probability of glaucoma, e.g., if the image's estimated probability is higher than the threshold, the image is classified as glaucoma.…”