June 2020
Volume 61, Issue 7
Free
ARVO Annual Meeting Abstract  |   June 2020
Glance: A Visual Analytics Approach for Opening the Black Box to Explain Deep Learning Predictions of Glaucomatous Visual Field Damage from Optical Coherence Tomography Scans
Author Affiliations & Notes
  • Astrid van den Brandt
    Eindhoven University of Technology, Eindhoven, Netherlands
  • Mark Christopher
    Hamilton Glaucoma Center, Shiley Eye Institute, Viterbi Family Department of Ophthalmology, University of California San Diego, California, United States
  • Jasmin Rezapour
    Department of Ophthalmology, University Medical Center Mainz, Mainz, Germany
  • Derek S. Welsbie
    Hamilton Glaucoma Center, Shiley Eye Institute, Viterbi Family Department of Ophthalmology, University of California San Diego, California, United States
  • Andrew S. Camp
    Hamilton Glaucoma Center, Shiley Eye Institute, Viterbi Family Department of Ophthalmology, University of California San Diego, California, United States
  • Sally L. Baxter
    Hamilton Glaucoma Center, Shiley Eye Institute, Viterbi Family Department of Ophthalmology, University of California San Diego, California, United States
  • Jiun Do
    Hamilton Glaucoma Center, Shiley Eye Institute, Viterbi Family Department of Ophthalmology, University of California San Diego, California, United States
  • Sasan Moghimi
    Hamilton Glaucoma Center, Shiley Eye Institute, Viterbi Family Department of Ophthalmology, University of California San Diego, California, United States
  • Akram Belghith
    Hamilton Glaucoma Center, Shiley Eye Institute, Viterbi Family Department of Ophthalmology, University of California San Diego, California, United States
  • Christopher Bowd
    Hamilton Glaucoma Center, Shiley Eye Institute, Viterbi Family Department of Ophthalmology, University of California San Diego, California, United States
  • Michael Henry Goldbaum
    Hamilton Glaucoma Center, Shiley Eye Institute, Viterbi Family Department of Ophthalmology, University of California San Diego, California, United States
  • Robert N Weinreb
    Hamilton Glaucoma Center, Shiley Eye Institute, Viterbi Family Department of Ophthalmology, University of California San Diego, California, United States
  • Michel A. Westenberg
    Eindhoven University of Technology, Eindhoven, Netherlands
  • Chris CP Snijders
    Eindhoven University of Technology, Eindhoven, Netherlands
  • Linda M Zangwill
    Hamilton Glaucoma Center, Shiley Eye Institute, Viterbi Family Department of Ophthalmology, University of California San Diego, California, United States
  • Footnotes
    Commercial Relationships   Astrid van den Brandt, None; Mark Christopher, None; Jasmin Rezapour, None; Derek S. Welsbie, None; Andrew S. Camp, None; Sally L. Baxter, None; Jiun Do, None; Sasan Moghimi, None; Akram Belghith, None; Christopher Bowd, None; Michael Goldbaum, None; Robert Weinreb, Aerie Pharmaceuticals (C), Allergan (C), Bausch&Lomb (C), Bausch&Lomb (F), Carl Zeiss Meditec (F), Centervue (F), Eyenovia (C), Heidelberg Engineering (F), Konan Medical (F), Meditec-Zeiss (P), Optovue (F), Toromedes (P); Michel A. Westenberg, None; Chris Snijders, None; Linda Zangwill, Carl Zeiss Meditec Inc. (F), Carl Zeiss Meditec Inc. (P), Heidelberg Engineering GmbH (F), Heidelberg Engineering GmbH (R), National Eye Institute (F), Optovue Inc. (F), Topcon Medical Systems Inc. (F)
  • Footnotes
    Support  None
Investigative Ophthalmology & Visual Science June 2020, Vol.61, 4527. doi:
  • Views
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Astrid van den Brandt, Mark Christopher, Jasmin Rezapour, Derek S. Welsbie, Andrew S. Camp, Sally L. Baxter, Jiun Do, Sasan Moghimi, Akram Belghith, Christopher Bowd, Michael Henry Goldbaum, Robert N Weinreb, Michel A. Westenberg, Chris CP Snijders, Linda M Zangwill; Glance: A Visual Analytics Approach for Opening the Black Box to Explain Deep Learning Predictions of Glaucomatous Visual Field Damage from Optical Coherence Tomography Scans. Invest. Ophthalmol. Vis. Sci. 2020;61(7):4527.

      Download citation file:


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

      ×
  • Supplements
Abstract

Purpose : To design and evaluate a visual analytics approach for explaining Deep Learning (DL) predictions of visual field (VF) mean deviation (MD) based on Optical Coherence Tomography (OCT ) retinal nerve fiber layer (RNFL) and macula scans.

Methods : We designed a tool (Glance) to provide visual explanations for a DL model that predicts 24-2 and 10-2 VF MD from RNFL and macula OCT scans. Glance’s visual interface was designed through participatory design sessions with ophthalmologists, vision researchers and OCT manufacturers. Three variations of the tool were created: 1. The TSNIT standard RNFL plot and previous actual VF MDs; 2. The TSNIT plot with a DL-derived saliency map of regions utilized by the model to make its predictions; 3. A graphical presentation of previous actual and predicted VF MDs; 4. A TSNIT plot, a saliency map and presentation of actual and predicted past VF MDs. Three clinicians completed a standardized review of 18 eyes using the 4 variations and answered questions related to trust and reliability for each visualization tool.

Results : The visualization tool helped in 4 tasks: (1) assessing reliability of a prediction, (2) understanding why the model made a prediction, (3) alerting to features that are relevant and (4) guiding future scheduling of VF. Glance was helpful for interpreting a prediction and its reliability in 20% of early-to-moderate glaucoma cases and in 4% of advanced cases. Moreover, it helped in identifying predictions that were likely to be inaccurate. Clinicians’ self-reported confidence in their management plan based on the DL predictions was about 30% higher when presented with visualizations that included past predictions and saliency maps compared to having no visual explanations. Clinicians were much less likely to have trust in the predictions if past predictions were inaccurate. The DL model predicted 24-2 VF with mean absolute error (95% CI) of 2.16 dB (2.00 - 2.97). The clinicians’ perceived utility of the tool varied by severity of disease and accuracy of past predictions.

Conclusions : Visualization approaches can assist clinicians to determine interpretability and enhance trust in DL predictions, particularly if information on the accuracy of past predictions is provided. Clinical utility of the visualizations for disease management is dependent on disease severity.

This is a 2020 ARVO Annual Meeting abstract.

×
×

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.

×