June 2017
Volume 58, Issue 8
Open Access
ARVO Annual Meeting Abstract  |   June 2017
Catastrophic Failure in Image-Based Convolutional Neural Network Algorithms for Detecting Diabetic Retinopathy
Author Affiliations & Notes
  • Stephanie Klein Lynch
    Ophthalmology & Visual Sciences, University of Iowa Hospitals & Clinics, Iowa City, Iowa, United States
  • Abhay Shah
    Electrical and Computer Engineering Department, University of Iowa, Iowa City, Iowa, United States
  • James C Folk
    Ophthalmology & Visual Sciences, University of Iowa Hospitals & Clinics, Iowa City, Iowa, United States
  • Xiaodong Wu
    Electrical and Computer Engineering Department, University of Iowa, Iowa City, Iowa, United States
  • Michael David Abramoff
    Ophthalmology & Visual Sciences, University of Iowa Hospitals & Clinics, Iowa City, Iowa, United States
    IDx LLC, Iowa City, Iowa, United States
  • Footnotes
    Commercial Relationships   Stephanie Lynch, None; Abhay Shah, None; James Folk, IDx LLC (I); Xiaodong Wu, None; Michael Abramoff, IDx LLC (C), IDx LLC (I), University of Iowa (P)
  • Footnotes
    Support  NIH Grants R01 EY019112, R01 EY018853; the Department of Veterans Affairs; Alimera, Inc.; The Robert C. Watzke MD Professorship for Retina Research
Investigative Ophthalmology & Visual Science June 2017, Vol.58, 3776. doi:
  • Views
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to Subscribers Only
      Sign In or Create an Account ×
    • Get Citation

      Stephanie Klein Lynch, Abhay Shah, James C Folk, Xiaodong Wu, Michael David Abramoff; Catastrophic Failure in Image-Based Convolutional Neural Network Algorithms for Detecting Diabetic Retinopathy. Invest. Ophthalmol. Vis. Sci. 2017;58(8):3776.

      Download citation file:


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

      ×
  • Supplements
Abstract

Purpose : Convolutional neural networks (CNNs) outperform retinal specialists in detecting diabetic retinopathy (DR). There are two principal CNN designs: 1) image-based deep learning algorithms, such as that developed by Google Inc.,[1] in which a CNN trains based on whole images, and 2) hybrid algorithms, in which multiple, semi-dependent CNNs train based on the appearance of focal lesions [2]. We compared the performance of these two CNN designs on adversarial (confounder) images, in which a small fraction of pixels have been modified.

Methods : Ten (10) DR images were selected from a reference set. The images were subjected to slight pixel modifications through a process called adversarialization. This was performed as follows: the 10 DR images were presented to an image-based CNN, which had already been trained on 500k DR images (AUC 0.99) to high performance. The diagnostic output for the 10 images was re-labeled from ‘DR’ to ‘normal.’ The resulting error was back-propagated into the image through 450 iterations; pixels in each image were updated iteratively at the input layer as εsign(▽xJ(θ,x,y)), where ε is the small learning rate (0.001) and xJ(θ,x,y) is the gradient of the Jacobian. Ten (10) adversarial images resulted. These were input into the hybrid algorithm, trained on a set of 5 million image components (AUC 0.98), to determine if DR would be detected.

Results : The difference between the adversarial images and original DR images averaged 0.5-1.3 pixel values (0.12%-0.51%) [Fig. 1]. Clinicians and the hybrid algorithm identified the adversarial images correctly as ‘DR,’ despite the fact that the image-based system classified all of them as normal.

Conclusions : Although both image-based and hybrid systems perform equivalently on validated datasets, image-based systems may fail catastrophically when confronted with adversarial images. They are sensitive to extremely small changes, potentially leading to false negatives. Hybrid algorithms based on multiple semi-dependent CNNs may offer a more robust option for clinical screening.

1. Gulshan V et al. Development and Validation of a Deep Learning Algorithm for Detection of Diabetic Retinopathy in Retinal Fundus Photographs. JAMA. 2016 Nov 29.
2. Abràmoff MD et al. Improved Automated Detection of Diabetic Retinopathy on a Publicly Available Dataset Through Integration of Deep Learning. IOVS. 2016 Oct;57(13):5200-5206.

This is an abstract that was submitted for the 2017 ARVO Annual Meeting, held in Baltimore, MD, May 7-11, 2017.

 

×
×

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.

×