July 2019
Volume 60, Issue 9
Free
ARVO Annual Meeting Abstract  |   July 2019
Effect of Image Compression and Number of Fields on a Deep Learning System for Detection of Diabetic Retinopathy
Author Affiliations & Notes
  • Michelle Y.T. Yip
    Singapore Eye Research Institute, Singapore, Singapore
    Duke-NUS Medical School, Singapore
  • Gilbert Lim
    National University of Singapore, School Of Computing, Singapore
  • Valentina Bellemo
    Singapore Eye Research Institute, Singapore, Singapore
  • Yuchen Xie
    Singapore Eye Research Institute, Singapore, Singapore
  • Xin Qi Lee
    Singapore Eye Research Institute, Singapore, Singapore
  • Quang Nguyen
    Singapore Eye Research Institute, Singapore, Singapore
  • Haslina Hamzah
    Singapore National Eye Centre, Singapore
  • Jinyi Ho
    Singapore National Eye Centre, Singapore
  • Charumathi Sabanayagam
    Singapore Eye Research Institute, Singapore, Singapore
  • Carol Yim-lui Cheung
    The Chinese University of Hong Kong, Hong Kong
  • Gavin Siew Wei Tan
    Singapore Eye Research Institute, Singapore, Singapore
    Singapore National Eye Centre, Singapore
  • Wynne Hsu
    National University of Singapore, School Of Computing, Singapore
  • Mong Li Lee
    National University of Singapore, School Of Computing, Singapore
  • Tien Yin Wong
    Singapore Eye Research Institute, Singapore, Singapore
    Singapore National Eye Centre, Singapore
  • Daniel SW Ting
    Singapore Eye Research Institute, Singapore, Singapore
    Singapore National Eye Centre, Singapore
  • Footnotes
    Commercial Relationships   Michelle Yip, None; Gilbert Lim, Automated Retinal Image Analysis Software for Referable Diabetic Retinopathy, Glaucoma Suspect and Age-Related Macular Degeneration. 10201706186V (Singapore) (P); Valentina Bellemo, None; Yuchen Xie, None; Xin Qi Lee, None; Quang Nguyen, None; Haslina Hamzah, None; Jinyi Ho, None; Charumathi Sabanayagam, None; Carol Cheung, None; Gavin Tan, None; Wynne Hsu, Automated Retinal Image Analysis Software for Referable Diabetic Retinopathy, Glaucoma Suspect and Age-Related Macular Degeneration. 10201706186V (Singapore) (P); Mong Li Lee, Automated Retinal Image Analysis Software for Referable Diabetic Retinopathy, Glaucoma Suspect and Age-Related Macular Degeneration. 10201706186V (Singapore) (P); Tien Wong, Automated Retinal Image Analysis Software for Referable Diabetic Retinopathy, Glaucoma Suspect and Age-Related Macular Degeneration. 10201706186V (Singapore) (P); Daniel Ting, Automated Retinal Image Analysis Software for Referable Diabetic Retinopathy, Glaucoma Suspect and Age-Related Macular Degeneration. 10201706186V (Singapore) (P)
  • Footnotes
    Support  None
Investigative Ophthalmology & Visual Science July 2019, Vol.60, 1438. doi:
  • Views
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Michelle Y.T. Yip, Gilbert Lim, Valentina Bellemo, Yuchen Xie, Xin Qi Lee, Quang Nguyen, Haslina Hamzah, Jinyi Ho, Charumathi Sabanayagam, Carol Yim-lui Cheung, Gavin Siew Wei Tan, Wynne Hsu, Mong Li Lee, Tien Yin Wong, Daniel SW Ting; Effect of Image Compression and Number of Fields on a Deep Learning System for Detection of Diabetic Retinopathy. Invest. Ophthalmol. Vis. Sci. 2019;60(9):1438.

      Download citation file:


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

      ×
  • Supplements
Abstract

Purpose : Many deep learning systems(DLS), trained with different methods, reported excellent diagnostic ability to detect diabetic retinopathy(DR). As different screening programs around the world produce images of varying specifications such as image size and field of view, it is important to evaluate the generalizability of a DLS to accommodate for these variations. This study aims to investigate the impact of the image compression and field of view on the diagnostic performance of a previously validated DLS.

Methods : The DLS was trained with 2-field(45° macula(MC) and optic-disc centred(OD)) retinal images from the Singapore Integrated Diabetic Retinopathy Program(SiDRP) taken from 2010-13. It was trained to detect referable DR, defined as moderate non-proliferative DR or worse, diabetic macular edema or ungradable. To examine image compression, we resized raw retinal images collected from SiDRP between 2014-15, from 350 kilobytes(KB) to 300, 250, 200 and 150 KB using OpenCV’s JPEG compression engine(n=71,896 images, 35,948 eyes, 14,880 subjects). To explore fields of view, we evaluated the DLS’ performance on 1-field(45° MC) images from SiDRP 2014-15 compared to 2-field images. The outcome measures were area under the curve(AUC), sensitivity and specificity.

Results : AUC of the DLS remained consistently above 0.9 despite reduction in image size and there is no significant difference until the image downsized to 250KB (p<0.05). Sensitivity and specificity is maintained above 80% until the threshold of 250KB (90.0% & 81.1% respectively) where further downsizing led to specificity to fall below 80%. Heatmap analysis showed that increase in false positives were due to lower image quality leading the DLS to focus on fake aberrancies. Providing the DLS with 1-field images preserves the performance above 0.85 (AUC 0.866, Sensitivity 91.7%, Specificity 91.9%) although a fall in AUC and specificity occurs (p<0.05).

Conclusions : Reducing image size and number of fields do not significantly affect the clinical diagnostic performance of the DLS, provided that image size is maintained to a minimum of 250KB and the single-field provided is a MC image.

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

 

Table 1. Performance of the DLS with different image sizes (top) and different fields of view (bottom).

Table 1. Performance of the DLS with different image sizes (top) and different fields of view (bottom).

 

Figure 1. Heatmap analysis of a normal retinal image misidentified to be diseased at a lower resolution due to the creation of fake aberrancies.

Figure 1. Heatmap analysis of a normal retinal image misidentified to be diseased at a lower resolution due to the creation of fake aberrancies.

×
×

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.

×