July 2019
Volume 60, Issue 9
Open Access
ARVO Annual Meeting Abstract  |   July 2019
Prediction of Causative Genes in Inherited Retinal Disorders From Spectral-domain Optical Coherent Tomography Utilizing Deep Learning Techniques
Author Affiliations & Notes
  • Yu Yokokawa Fujinami
    Laboratory of Visual Physiology, Division of Vision Research, National Institute of Sensory Organs, National Hospital Organization, Tokyo Medical Center, Meguro-ku, TOKYO, Japan
    Graduate School of Health Management, Keio University, Shinjyuku, Tokyo, Japan
  • NIKOLAS PONTIKOS
    Moorfields Eye Hospital, United Kingdom
    Laboratory of Visual Physiology, Division of Vision Research, National Institute of Sensory Organs, National Hospital Organization, Tokyo Medical Center, Meguro-ku, TOKYO, Japan
  • Lizhu Yang
    Laboratory of Visual Physiology, Division of Vision Research, National Institute of Sensory Organs, National Hospital Organization, Tokyo Medical Center, Meguro-ku, TOKYO, Japan
    Department of Ophthalmology, Keio University of Medicine, Japan
  • Kazutoshi Yoshitake
    Division of Molecular and Cellular Biology, National Institute of Sensory Organs, National Tokyo Medical Center, Japan
  • Kazushige Tsunoda
    Laboratory of Visual Physiology, Division of Vision Research, National Institute of Sensory Organs, National Hospital Organization, Tokyo Medical Center, Meguro-ku, TOKYO, Japan
  • Gavin Arno
    UCL Institute of Ophthalmology, London, United Kingdom
    Laboratory of Visual Physiology, Division of Vision Research, National Institute of Sensory Organs, National Hospital Organization, Tokyo Medical Center, Meguro-ku, TOKYO, Japan
  • Takeshi Iwata
    Division of Molecular and Cellular Biology, National Institute of Sensory Organs, National Tokyo Medical Center, Japan
  • Hiroaki Miyata
    Graduate School of Health Management, Keio University, Shinjyuku, Tokyo, Japan
    Department of Health Policy and Management, School of Medicine, Keio University, Japan
  • Kaoru Fujinami
    Laboratory of Visual Physiology, Division of Vision Research, National Institute of Sensory Organs, National Hospital Organization, Tokyo Medical Center, Meguro-ku, TOKYO, Japan
    UCL Institute of Ophthalmology, London, United Kingdom
  • Footnotes
    Commercial Relationships   Yu Fujinami, Acucela Inc. (C), Acucela Inc. (F), Astellas Pharma Inc. (C), Astellas Pharma Inc. (F), Foundation Fighting Blindness (F), Foundation Fighting Blindness Clinical Research Institute (F), Japanese Ophthalmology Society (F), Japan Retinitis Pigmentosa Society (F), Kubota Pharmaceutical Holdings Co., Ltd. (C), NightstaRx Limited. (C), NightstaRx Limited. (F), SANTEN Company Limited. (F); NIKOLAS PONTIKOS, NIHR Moorfields Biomedical Research Centre (F); Lizhu Yang, Astellas Pharma Inc (F); Kazutoshi Yoshitake, None; Kazushige Tsunoda, None; Gavin Arno, None; Takeshi Iwata, None; Hiroaki Miyata, None; Kaoru Fujinami, Acucela Inc. (C), Acucela Inc. (F), Astellas Pharma Inc. (C), Foundation Fighting Blindness, (F), Foundation Fighting Blindness Clinical Research Institute, (F), Japanese Ophthalmology Society, (F), Japan Retinitis Pigmentosa Society. (F), Kubota Pharmaceutical Holdings Co., Ltd. (C), Kubota Pharmaceutical Holdings Co., Ltd. (F), NightstaRx Limited. (C), NightstaRx Limited. (F), SANTEN Company Limited (F)
  • Footnotes
    Support  Grant-in-Aid for Young Scientists of the Ministry of Education, Culture, Sports, Science and Technology, Japan (18K16943).
Investigative Ophthalmology & Visual Science July 2019, Vol.60, 2950. doi:
  • Views
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Yu Yokokawa Fujinami, NIKOLAS PONTIKOS, Lizhu Yang, Kazutoshi Yoshitake, Kazushige Tsunoda, Gavin Arno, Takeshi Iwata, Hiroaki Miyata, Kaoru Fujinami; Prediction of Causative Genes in Inherited Retinal Disorders From Spectral-domain Optical Coherent Tomography Utilizing Deep Learning Techniques. Invest. Ophthalmol. Vis. Sci. 2019;60(9):2950.

      Download citation file:


      © ARVO (1962-2015); The Authors (2016-present)

      ×
  • Supplements
Abstract

Purpose : To illustrate a data-driven deep learning approach to predicting the gene causing the Inherited retinal disorder (IRD) in macular dystrophy caused by ABCA4 and RP1L1 gene aberration in comparison with retinitis pigmentosa caused by EYS gene aberration and normal subjects.

Methods : Seventy-five subjects with molecularly confirmed IRD or no ocular diseases have been ascertained from the database of Japan Eye Genetics Consortium; 10 patients with ABCA4-retinopathy, 20 patients with RP1L1-retinopathy, 28 with EYS-retinopathy, and 17 normal subjects. Horizontal and vertical cross-sectional scans of SD-OCT at the central fovea of both eyes were cropped/adjusted to spatial resolution of 400 pixel/inch with a size of 47.62 × 31.75 mm2 for deep learning.
After preparation of SD-OCT images for four gene categories, subjects were randomly split following a 3:1 ratio into training and test sets. The commercially available deep learning web tool, Medic mind was applied to this four-class classification program. The classification accuracy, sensitivity, and specificity were calculated during the learning process. The process was repeated four times with random assignment of subjects to training and test sets to control for selection bias given the relatively small sample size. For each training/testing process, the classification accuracy was calculated per gene category.

Results : A total of 178 images from 75 subjects were included in this study. The mean training accuracy was 98.5%, (ranged from 90.6% to 100.0%). The mean overall test accuracy was 90.9% (82.0 - 97.6). The test accuracy per gene category was 100% for ABCA4, 78.0%(66.7-87.5) for RP1L1, 89.8%(82.4-100) for EYS, and 93.4%(73.7-100) for normal. Test accuracy of RP1L1 and EYS was not as high relative to the training accuracy which suggests overfitting.

Conclusions : This study highlighted a novel application of deep neural networks in the prediction of the causative gene in IRD retinopathies from SD-OCT, with a mean prediction accuracy of 90%. It is anticipated that deep neural networks will be integrated into general screening to support clinical diagnosis, suggest a causative gene to guide genetic screening, as well as enrich the clinical education of orphan retinal disease.

This abstract was presented at the 2019 ARVO Annual Meeting, held in Vancouver, Canada, April 28 - May 2, 2019.

 

Spectral-domain optical coherence tomographic images of four categories in prediction of causative genes.

Spectral-domain optical coherence tomographic images of four categories in prediction of causative genes.

×
×

This PDF is available to Subscribers Only

Sign in or purchase a subscription to access this content. ×

You must be signed into an individual account to use this feature.

×