Article Text

Download PDFPDF
Usefulness of data augmentation for visual field trend analyses in patients with glaucoma
  1. Ryo Asaoka1,
  2. Hiroshi Murata1,
  3. Masato Matsuura1,
  4. Yuri Fujino1,
  5. Atsuya Miki2,
  6. Masaki Tanito3,
  7. Shiro Mizoue4,5,
  8. Kazuhiko Mori6,
  9. Katsuyoshi Suzuki7,
  10. Takehiro Yamashita8,
  11. Kenji Kashiwagi9,
  12. Nobuyuki Shoji10
  1. 1Department of Ophthalmology, University of Tokyo Graduate School of Medicine, Tokyo, Japan
  2. 2Department of Ophthalmology, Osaka University Graduate School of Medicine, Suita, Japan
  3. 3Department of Ophthalmology, Shimane University Faculty of Medicine, Izumo, Japan
  4. 4Department of Ophthalmology, Minami-matsuyama Hospital, Matsuyama-shi, Japan
  5. 5Department of Ophthalmology, Ehime University School of Medicine, Toon-shi, Japan
  6. 6Department of Ophthalmology, Kyoto Prefectural University of Medicine, Kyoto, Japan
  7. 7Department of Ophthalmology, Yamaguchi University Graduate School of Medicine, Ube, Japan
  8. 8Department of Ophthalmology, Kagoshima University Graduate School of Medical and Dental Sciences, Kagoshima, Japan
  9. 9Department of Ophthalmology, Faculty of Medicine, University of Yamanashi, Chuo, Japan
  10. 10Department of Ophthalmology, Kitasato University, Minato-ku, Japan
  1. Correspondence to Dr Ryo Asaoka, Department of Ophthalmology, University of Tokyo Graduate School of Medicine, Tokyo 113-8655, Japan; ryoasa0120{at}mac.com

Abstract

Aim To investigate the usefulness of data augmentation in visual field (VF) trend analyses in patients with glaucoma.

Method This study included 6380 VFs from 638 eyes of 417 patients with open-angle glaucoma. Various affine transformations were applied to augment the VF data: (1) rotation, (2) scaling, (3) vertical and horizontal shift and (4) a combination of these different transformations. 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, and the root mean squared error (RMSE) was calculated.

Results With PLR, mean RMSE without VF augmentation averaged from 3.95 (VF1–3) to 19.01 (VF1–9) dB. The RMSE was significantly improved by applying the different transformations: (1) rotation (from VF1–3 to VF1–7), (2) scaling (from VF1–3 to VF1–6), (3) vertical and horizontal shifts (from VF1–3 to VF1–4) and (iv) a combination of these (from VF1–3 to VF1–7). Progression rates in VF1–10 had better agreement with those in shorter VF series when a combination of affine transformation was applied. The differences in rates were between 1.9 (VF1–3) and 0.39 (VF1–9) dB if augmentation was used, which was significantly smaller than that observed when augmentation was not applied (from 2.6 with VF1–3 to 0.26 dB with VF1–9).

Conclusion It is useful to apply VF data augmentation techniques when predicting future VF progression in glaucoma using PLR, especially with short VF series.

  • visual field
  • glaucoma
  • progression
  • data augmentation
View Full Text

Statistics from Altmetric.com

Footnotes

  • Contributors Conceived and designed the experiments: HM and RA. Performed the experiments: RA. Analysed the data: RA. Contributed reagents/materials/analysis tools: YF, HM, AM, MT, SM, KM, KS, TY, KK, KH and NS. Wrote the paper: RA.

  • Funding This study was funded by Japan Science and Technology Agency (AIP Acceleration Research), Suzuken Memorial Foundation, the Ministry of Education, Culture, Sports, Science and Technology of Japan (25861618, 26462679 and 18KK0253).

  • Competing interests None declared.

  • Patient consent for publication Not required.

  • Ethics approval The review board of the University of Tokyo, Osaka University Graduate School of Medicine, Shimane University Faculty of Medicine, Matsue Red Cross Hospital, Ehime University Graduate School of Medicine, Kyoto Prefectural University of Medicine, Yamaguchi University Graduate School of Medicine, Kagoshima University Graduate School of Medical and Dental Sciences, and University of Yamanashi Faculty of Medicine, Kitasato University reviewed and approved all protocols.

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Data availability statement No data are available. N/A.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.