The researchers conducted this research to investigate the usefulness of data augmentation in the VF trend analyses in patients with glaucoma.
This study included a total of 6380 VFs from 638 eyes of 417 patients with open-angle glaucoma. Using pointwise linear regression (PLR), the total deviation (TD) values of a patient’s 10th VF were predicted using TD values from shorter VF series (from first to third VFs (VF1–3) to first to ninth VFs (VF1–9)) with and without VF data augmentation.
Mean RMSE without VF augmentation averaged from 3.95 (VF1–3) to 19.01 (VF1–9) dB with PLR. The RMSE was significantly improved by applying the different transformations. Progression rates in VF1–10 had better agreement with those in shorter VF series when researchers applied a variety of affine transformations. The differences in rates were between 1.9 (VF1–3) and 0.39 (VF1–9) dB if augmentation was used, significantly smaller than that observed when boost was not applied.
The study concluded that it is useful to apply VF data augmentation techniques while predicting future VF progression in glaucoma using PLR, especially with short VF series.
Reference: https://bjo.bmj.com/content/104/12/1697
