Screening Attentional-related Diseases based on Correlation between Salience and Gaze

Screening Attentional-related Diseases based on Correlation between Salience and Gaze. Tseng, P., Cameron, I. G.<nbsp>M., Munoz, D. P., & Itti, L. In Proc. Vision Science Society Annual Meeting (VSS09), May, 2009.
abstract bibtex

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.

@inproceedings{ Tseng_etal09vss,
  author = {P. Tseng and I. G. M. Cameron and D. P. Munoz and L. Itti},
  title = {Screening Attentional-related Diseases based on Correlation
                  between Salience and Gaze},
  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. },
  booktitle = {Proc. Vision Science Society Annual Meeting (VSS09)},
  year = {2009},
  month = {May},
  type = {bu;td;mod;psy;med},
  review = {abs/conf}
}

Downloads: 0

{"_id":{"_str":"5298a1a19eb585cc2600092e"},"__v":0,"authorIDs":[],"author_short":["Tseng, P.","Cameron, I.<nbsp>G.<nbsp>M.","Munoz, D.<nbsp>P.","Itti, L."],"bibbaseid":"tseng-cameron-munoz-itti-screeningattentionalrelateddiseasesbasedoncorrelationbetweensalienceandgaze-2009","bibdata":{"html":"<div class=\"bibbase_paper\"> \n\n\n<span class=\"bibbase_paper_titleauthoryear\">\n\t<span class=\"bibbase_paper_title\"><a name=\"Tseng_etal09vss\"> </a>Screening Attentional-related Diseases based on Correlation between Salience and Gaze.</span>\n\t<span class=\"bibbase_paper_author\">\nTseng, P.; Cameron, I. G.<nbsp>M.; Munoz, D. P.; and Itti, L.</span>\n\t\n</span>\n\n\n\nIn\n<i>Proc. Vision Science Society Annual Meeting (VSS09)</i>, May 2009.\n\n\n\n\n\n<br class=\"bibbase_paper_content\"/>\n\n<span class=\"bibbase_paper_content\">\n \n \n \n <a href=\"javascript:showBib('Tseng_etal09vss')\"\n class=\"bibbase link\">\n \n\t\n\t\n\t\n BibTeX\n <i class=\"fa fa-caret-down\"></i></a>\n \n \n  \n <a class=\"bibbase_abstract_link bibbase link\"\n href=\"javascript:showAbstract('Tseng_etal09vss')\">\n Abstract\n <i class=\"fa fa-caret-down\"></i></a>\n \n \n \n\n \n \n \n</span>\n\n<div class=\"well well-small bibbase\" id=\"bib_Tseng_etal09vss\"\n style=\"display:none\">\n <pre>@inproceedings{ Tseng_etal09vss,\n author = {P. Tseng and I. G. M. Cameron and D. P. Munoz and L. Itti},\n title = {Screening Attentional-related Diseases based on Correlation\n between Salience and Gaze},\n abstract = {Several studies have shown that eye movements and certain\n complex visual functions are influenced by diseases\n such as Parkinson's Disease (PD) and Attention\n Deficit Hyperactivity Disorder (ADHD). Here we\n examine how bottom-up (stimulus-driven) attentional\n selection mechanisms may differ between patient and\n control populations, and we take advantage of the\n difference to develop classifiers to differentiate\n patients from controls. We tracked gaze of four\n groups of observers (15 control children, aged 7-14;\n 6 ADHD children, aged 9-15; 12 control elderly, aged\n 66-79; and 9 PD elderly, aged 53-68) while they\n freely viewed MTV-style videos. These stimuli are\n composed of short (2-4 seconds), unrelated clips of\n natural scenes to reduce top-down (contextual)\n expectations and emphasize bottom-up influences on\n gaze allocations at the scene change. We used a\n saliency model to compute bottom-up saliency maps\n for every video frame. Saliency maps can be computed\n from a full set of features (color, intensity,\n orientation, flicker, motion) or from individual\n features. Support-vector-machine classifiers (with\n Radial-Basis Function Kernel) were built for each\n feature contributing the saliency map and for the\n combination of them. Leave-one-out was used to train\n and test the classifiers. Two classification\n experiments were performed: (1) between ADHD and\n control children; (2) between PD and control\n elderly. Saliency maps computed with all features\n can well differentiate patients and control\n populations (correctness: experiment 1 - 100%;\n experiment 2 - 95.24%). Additionally, saliency maps\n computed from any one feature performed nearly as\n well (both experiments' results are 0-5%\n worse). Moreover, 0-250 ms after scene change is the\n most discriminative period for the\n classification. This study demonstrates that the\n bottom-up mechanism is greatly influenced by PD and\n ADHD, and the difference can serve as a probable\n diagnosis tool for clinical applications. },\n booktitle = {Proc. Vision Science Society Annual Meeting (VSS09)},\n year = {2009},\n month = {May},\n type = {bu;td;mod;psy;med},\n review = {abs/conf}\n}</pre>\n</div>\n\n\n<div class=\"well well-small bibbase\" id=\"abstract_Tseng_etal09vss\"\n style=\"display:none\">\n 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.\n</div>\n\n\n</div>\n","downloads":0,"bibbaseid":"tseng-cameron-munoz-itti-screeningattentionalrelateddiseasesbasedoncorrelationbetweensalienceandgaze-2009","role":"author","year":"2009","type":"bu;td;mod;psy;med","title":"Screening Attentional-related Diseases based on Correlation between Salience and Gaze","review":"abs/conf","month":"May","key":"Tseng_etal09vss","id":"Tseng_etal09vss","booktitle":"Proc. Vision Science Society Annual Meeting (VSS09)","bibtype":"inproceedings","bibtex":"@inproceedings{ Tseng_etal09vss,\n author = {P. Tseng and I. G. M. Cameron and D. P. Munoz and L. Itti},\n title = {Screening Attentional-related Diseases based on Correlation\n between Salience and Gaze},\n abstract = {Several studies have shown that eye movements and certain\n complex visual functions are influenced by diseases\n such as Parkinson's Disease (PD) and Attention\n Deficit Hyperactivity Disorder (ADHD). Here we\n examine how bottom-up (stimulus-driven) attentional\n selection mechanisms may differ between patient and\n control populations, and we take advantage of the\n difference to develop classifiers to differentiate\n patients from controls. We tracked gaze of four\n groups of observers (15 control children, aged 7-14;\n 6 ADHD children, aged 9-15; 12 control elderly, aged\n 66-79; and 9 PD elderly, aged 53-68) while they\n freely viewed MTV-style videos. These stimuli are\n composed of short (2-4 seconds), unrelated clips of\n natural scenes to reduce top-down (contextual)\n expectations and emphasize bottom-up influences on\n gaze allocations at the scene change. We used a\n saliency model to compute bottom-up saliency maps\n for every video frame. Saliency maps can be computed\n from a full set of features (color, intensity,\n orientation, flicker, motion) or from individual\n features. Support-vector-machine classifiers (with\n Radial-Basis Function Kernel) were built for each\n feature contributing the saliency map and for the\n combination of them. Leave-one-out was used to train\n and test the classifiers. Two classification\n experiments were performed: (1) between ADHD and\n control children; (2) between PD and control\n elderly. Saliency maps computed with all features\n can well differentiate patients and control\n populations (correctness: experiment 1 - 100%;\n experiment 2 - 95.24%). Additionally, saliency maps\n computed from any one feature performed nearly as\n well (both experiments' results are 0-5%\n worse). Moreover, 0-250 ms after scene change is the\n most discriminative period for the\n classification. This study demonstrates that the\n bottom-up mechanism is greatly influenced by PD and\n ADHD, and the difference can serve as a probable\n diagnosis tool for clinical applications. },\n booktitle = {Proc. Vision Science Society Annual Meeting (VSS09)},\n year = {2009},\n month = {May},\n type = {bu;td;mod;psy;med},\n review = {abs/conf}\n}","author_short":["Tseng, P.","Cameron, I.<nbsp>G.<nbsp>M.","Munoz, D.<nbsp>P.","Itti, L."],"author":["Tseng, P.","Cameron, I. G. M.","Munoz, D. P.","Itti, L."],"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."},"bibtype":"inproceedings","biburl":"http://ilab.usc.edu/publications/src/ilab.bib","downloads":0,"search_terms":["screening","attentional","related","diseases","based","correlation","between","salience","gaze","tseng","cameron","munoz","itti"],"title":"Screening Attentional-related Diseases based on Correlation between Salience and Gaze","year":2009,"dataSources":["wedBDxEpNXNCLZ2sZ"]}