May 2008
Volume 8, Issue 6
Free
Vision Sciences Society Annual Meeting Abstract  |   May 2008
Differentiating patients from controls based on correlation between salience and gaze
Author Affiliations
  • Po-He Tseng
    Department of Computer Science, University of Southern California
  • Ian G. M. Cameron
    Centre for Neuroscience Studies and Department of Physiology, Queen's University
  • Doug Munoz
    Centre for Neuroscience Studies and Department of Physiology, Queen's University
  • Laurent Itti
    Department of Computer Science, University of Southern California, and Neuroscience Program, University of Southern California
Journal of Vision May 2008, Vol.8, 1091. doi:10.1167/8.6.1091
  • Views
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Po-He Tseng, Ian G. M. Cameron, Doug Munoz, Laurent Itti; Differentiating patients from controls based on correlation between salience and gaze. Journal of Vision 2008;8(6):1091. doi: 10.1167/8.6.1091.

      Download citation file:


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

      ×
  • Supplements
Abstract

Several studies have shown that eye movements and certain complex visual functions are influenced by diseases such as Parkinson's Disease (PD) and Attention Deficit Hyperactivity Disorder (ADHD). Here we examine how bottom-up (stimulus-driven) attentional selection mechanisms may differ between patient and control populations, and we take advantage of the difference to develop classifiers to differentiate patients from controls. We tracked gaze of four groups of observers (15 control children, aged 7–14; 6 ADHD children, aged 9–15; 12 control elderly, aged 66–79; and 9 PD elderly, aged 53–68) while they freely viewed MTV-style videos. These stimuli are composed of short (2–4 seconds), unrelated clips of natural scenes to reduce top-down (contextual) expectations and emphasize bottom-up influences on gaze allocations at the scene change. We used a saliency model to compute bottom-up saliency maps for every video frame. Saliency maps can be computed from a full set of features (color, intensity, orientation, flicker, motion) or from individual features. Support-vector-machine classifiers (with Radial-Basis Function Kernel) were built for each feature contributing the saliency map and for the combination of them. Leave-one-out was used to train and test the classifiers. Two classification experiments were performed: (1) between ADHD and control children; (2) between PD and control elderly. Saliency maps computed with all features can well differentiate patients and control populations (correctness: experiment 1 – 100%; experiment 2 – 95.24%). Additionally, saliency maps computed from any one feature performed nearly as well (both experiments' results are 0–5% worse). Moreover, 0–250 ms after scene change is the most discriminative period for the classification. This study demonstrates that the bottom-up mechanism is greatly influenced by PD and ADHD, and the difference can serve as a probable diagnosis tool for clinical applications.

Tseng, P.-H. Cameron, I. G. M. Munoz, D. Itti, L. (2008). Differentiating patients from controls based on correlation between salience and gaze [Abstract]. Journal of Vision, 8(6):1091, 1091a, http://journalofvision.org/8/6/1091/, doi:10.1167/8.6.1091. [CrossRef]
Footnotes
 This work was supported by grants from The National Science Foundation and the Human Frontier Science Program.
×
×

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.

×