A transfer learning-assisted glaucoma detection system was build first with training images from the fundus images database of Taipei Veterans General Hospital,then fine-tuned to improve its applicability on fundus images from different datasets. This system aims to assist clinic-based glaucoma screening to increase the diagnostic rate of glaucoma by using existing healthcare facility.

Glaucoma is the leading cause of irreversible blindness, but is underdiagnosed worldwide. The diagnosis of glaucoma is further complicated by inter-physician discordance, especially in the interpretation of fundus photographs. A transfer learning-assisted glaucoma detection system with a diagnostic accuracy of 90 may facilitate clinic-based glaucoma screening.

For each fundus image, this system provides a probability value representing the likelihoods of glaucoma as a reference for clinician. Clinician may arrange further evaluation for eyes at risk of glaucoma. This approach has great potential to identify glaucoma at an early stage in a cost-effective way to reduce the socioeconomic burden of glaucoma blindness.