Automated eye tracking system calibration using artificial neural networks

The electro-oculogram (EOG) continues to be widely used to record eye movements especially in clinical settings. However, an efficient and accurate means of converting these recordings into eye position is lacking. An artificial neural network (ANN) that maps two-dimensional (2D) eye movement recordings into 2D eye positions can enhance the utility of such recordings. Multi-layer perceptrons (MLPs) with non-linear activation functions and trained with back propagation proved to be capable of calibrating simulated EOG data to a mean accuracy of 0.33 degrees . Linear perceptrons (LPs) were only nearly half as accurate. For five subjects, the mean accuracy provided by the MLPs was 1.09 degrees of visual angle ( degrees ) for EOG data, and 0.98 degrees for an infrared eye tracker. MLPs enabled calibration of 2D saccadic EOG to an accuracy not significantly different from that obtained with the infrared tracker. Using initial weights trained on another person reduced MLP training time, reaching convergence in as little as 20 iterations.

[1]  A. J. Van Opstal,et al.  Skewness of saccadic velocity profiles: A unifying parameter for normal and slow saccades , 1987, Vision Research.

[2]  O. Gambini,et al.  Smooth pursuit and saccadic eye movements and Wisconsin card sorting test performance in obsessive- compulsive disorder , 1993, Psychiatry Research.

[3]  Elizabeth A. Krupinski,et al.  Recording and analyzing eye-position data using a microcomputer workstation , 1992 .

[4]  W. Heide,et al.  Target anticipation and impairment of smooth pursuit eye movements in schizophrenia , 1998, Experimental Brain Research.

[5]  D. Fender,et al.  The interplay of drifts and flicks in binocular fixation. , 1969, Vision research.

[6]  J. Stern.  Theoretical and applied aspects of eye movement research A. G. Gale and F. Johnson, (Elsevier Science Publishers B.V., Amsterdam, 1984) pp. xiii + 565, Dfl. 185 , 1985, Biological Psychology.

[7]  L. Young,et al.  Survey of eye movement recording methods , 1975 .

[8]  R. Leigh,et al.  The neurology of eye movements , 1984 .

[9]  S Matthysse,et al.  Eye tracking dysfunction and schizophrenia: a critical perspective. , 1993, Schizophrenia bulletin.

[10]  L. Stark,et al.  The main sequence, a tool for studying human eye movements , 1975 .

[11]  J. V. Gisbergen,et al.  A parametric analysis of human saccades in different experimental paradigms , 1987, Vision Research.

[12]  Robert J. Schalkoff,et al.  Artificial neural networks , 1997 .

[13]  P. Venables,et al.  A Manual of Psychophysiological Methods , 1967 .

[14]  R. W. Ditchburn,et al.  Involuntary eye movements during fixation , 1953, The Journal of physiology.

[15]  R. Carpenter,et al.  Movements of the Eyes , 1978 .

[16]  Guillaume S. Masson,et al.  A look into the black box : eye movements as a probe of visual motion processing , 1998 .

[17]  F H Previc,et al.  Discriminability of random-dot stereograms in three-dimensional space. , 1995, The International journal of neuroscience.

[18]  Thierry d'Amato,et al.  Saccadic eye movements in schizophrenic patients , 1998, Psychiatry Research.

[19]  F. Previc,et al.  Visual search asymmetries in three-dimensional space , 1993, Vision Research.

[20]  J. P. H. Reulen,et al.  Precise recording of eye movement: the IRIS technique Part 1 , 2006, Medical and Biological Engineering and Computing.

[21]  R. Klein,et al.  What are human express saccades? , 1993, Perception & psychophysics.

[22]  Jihong Lee.  A Novel Design Method for Multilayer Feedforward Neural Networks , 1994, Neural Computation.

[23]  C Kennard,et al.  Oculomotor abnormalities in schizophrenia , 1998, Neurology.

[24]  R. Baloh,et al.  Quantitative measurement of saccade amplitude, duration, and velocity , 1975, Neurology.

[25]  D. Lykken,et al.  Two-year retest stability of eye tracking performance and a comparison of electro-oculographic and infrared recording techniques: evidence of EEG in the electro-oculogram. , 1981, Psychophysiology.