Contextual Augmentation Based on Metric-Guided Features for Ocular Axial Length Prediction

Ocular axial length (AL) measurement is important in ophthalmology because it should be considered prior to operations, such as strabismus surgery or cataract surgery, and the automation of AL measurement with easily obtained retinal fundus images has been studied. However, the performance of deep learning methods inevitably depends on distribution of the data set used, and the lack of data is an issue that needs to be addressed. In this study, we propose a framework for generating pairs of fundus images and their corresponding ALs to improve the AL inference. The generator's encoder was trained independently using metric learning based on the AL information. A random vector and zero padding were incorporated into the generator to increase data creation flexibility, after which AL information was inserted as conditional information. We verified the effectiveness of this framework by evaluating the performance of AL inference models after training them on a combined data set comprising privately collected actual data and data generated by the proposed method. Compared to using only the actual data set, the mean absolute error and standard deviation of the proposed method decreased from 10.23 and 2.56 to 3.96 and 0.23, respectively, even with a smaller number of layers in the AL prediction models.

Automated summary

Ocular axial length (AL) measurement is important in ophthalmology and automation of AL measurement using retinal fundus images has been studied. In this study, a framework for generating pairs of fundus images and their corresponding ALs was proposed to improve AL inference. The effectiveness of the framework was verified by evaluating the performance of AL inference models trained on a combined dataset of actual data and generated data.