Feasibility of simple machine learning approaches to support detection of non-glaucomatous visual fields in future automated glaucoma clinics

摘要

Examples of bilateral field representation for three pituitary fields (top) and glaucomatous fields (bottom). For each pair, the left field represents the left eye and the right fields the right eye. The fields are oriented as the patient would see the world, i.e. temporal loss is shown temporally, superior loss superiorly. Predominantly temporal defects are shown in the three pituitary field representations, whereas the glaucomatous fields show more variation in defect shape, position, severity and laterality