Determining Progression in Glaucoma Using Visual Fields
The standardized visual field assessment, which measures visual function in 76 locations of the central visual area, is an important diagnostic tool in the treatment of the eye disease glaucoma. It helps determine whether the disease is stable or progressing towards blindness, with important implications for treatment. Automatic techniques to classify patients based on this assessment have had limited success, primarily due to the high variability of individual visual field measurements.
The purpose of this paper is to describe the problem of visual field classification to the data mining community, and assess the success of data mining techniques on it. Preliminary results show that machine learning methods rival existing techniques for predicting whether glaucoma is progressing—though we have not yet been able to demonstrate improvements that are statistically significant. It is likely that further improvement is possible, and we encourage others to work on this important practical data mining problem.
Unable to display preview. Download preview PDF.
- 1.P. Armitage and G. Berry. Statistical Methods in Medical Research. Blackwell Scientific Pulications, Oxford, third edition, 1994.Google Scholar
- 2.M.K. Birch, P.K. Wishart, and N.P. O’Donnell. Determining progressive visual field loss in serial humphrey visual fields. Ophthomology, 102(8):1227–1235, 1995.Google Scholar
- 3.L. Brigatti, K. Nouri-Mahdavi, M. Weitzman, and J. Caprioli. Automatic detection of glaucomatous visual field progression with neural networks. Archives of Ophthalmol., 115:725–728, 1997.Google Scholar
- 5.B.C. Chauhan and C.A. Johnson. Test-retest variability of frequency-doubling perimetry and conventional perimetry in glaucoma patients and normal subjects. Investigative Ophthomology and Vision Science, 40(3):648–656, 1999.Google Scholar
- 6.B.C. Chauhan, S.M. Drance, and G.R. Douglas. The use of visual field indices in detecting changes in the visual field in glaucoma. Investigative Ophthomology and Vision Science, 31(3):512–520, 1990.Google Scholar
- 8.D.P. Crabb, F.W. Fitzke, A.I. McNaught, D.F. Edgar, and R.A. Hitchings. Improving the prediction of visual field progression in glaucoma using spatial processing. Ophthomology, 104(3):517–524, 1997.Google Scholar
- 9.M. Fingeret and T.L. Lewis. Primary care of the glaucomas. Mc Graw Hill, New York, second edition, 2001.Google Scholar
- 10.J.H. Friedman. Greedy function approximation: A gradient boosting machine. Technical report, Department of Statistics, Stanford University, CA, 1999.Google Scholar
- 14.S. Keerthi, S. Shevade, C. Bhattacharyya, and K. Murthy. Improvements to platt’s SMO algorithm for SVM classifier design. Technical report, Dept. of CSA, Banglore, India, 1999.Google Scholar
- 16.T. Leitman, J. Eng, J. Katz, and H.A. Quigley. Neural networks for visual field analysis: how do they compare with other algorithms. J Glaucoma, 8:77–80, 1999.Google Scholar
- 18.S. Mandava, M. Zulauf, T. Zeyen, and J. Caprioli. An evaluation of clusters in the glaucomatous visual field. American J. of Ophthalmol., 116(6):684–691, 1993.Google Scholar
- 19.R.K. Morgan, W.J. Feuer, and D.R. Anderson. Statpac 2 glaucoma change probability. Archive of Ophthalmol., 109:1690–1692, 1991.Google Scholar
- 20.K. Nouri-Mahdavi, L. Brigatti, M. Weitzman, and J. Caprioli. Comparison of methods to detect visual field progression in glaucoma. Ophthomology, 104(8):1228–1236, 1997.Google Scholar
- 21.J. Platt. Fast training of support vector machines using sequential minimal optimization. In Advances in Kernel Methods-Support Vector Learning. MIT Press, Cambridge, MA, 1998.Google Scholar
- 22.S.D. Smith, J. Katz, and H.A. Quigly. Analysis of progressive change in automated visual fields in glaucoma. Investigative Ophthomology and Vision Science, 37(7):1419–1428, 1996.Google Scholar
- 23.P.G.D. Spry, A.B. Bates, C.A. Johnson, and B.C. Chauhan. Simulation of longitudinal threshold visual field data. Investigative Ophthomology and Vision Science, 41(8):2192–2200, 2000.Google Scholar
- 25.C.J. Wild and G.A.F. Weber. Introduction to probability and statistics. Department of Statistics, University of Auckland, New Zealand, 1995.Google Scholar
- 26.J.M. Wild, M.K. Hussey, J.G. Flanagan, and G.E. Trope. Pointwise topographical and longitudinal modeling of the visual field in glaucoma. Investigative Ophthomology and Vision Science, 34(6):1907–1916, 1993.Google Scholar
- 27.J.M. Wild, N. Hutchings, M.K. Hussey, J.G. Flanagan, and G.E. Trope. Pointwise univariate linear regression of perimetric sensitivity against follow-up time in glaucoma. Ophthomology, 104(5):808–815, 1997.Google Scholar
- 28.Ian H. Witten and Eibe Frank. Data Mining: Practical Machine Learning Tools and Techniques with Java Implementations. Morgan Kaufmann, San Francisco, CA, 2000.Google Scholar