Main Article Content
Purpose: The aim of this study was the development of a critical code in order to combine statistics with a workable diagnostic system for open angle glaucoma that could predict improvement or deterioration of the tested visual field of a glaucoma suspect, most likely after the first or second visit for the visual field test.
Methods: The study plan was to apply a set of different filters in order to select the most efficient one that could remove the most of the noise of the test printout, of which probably the greater part of this removed noise could be the component of the learning effect, as it was expected to be at the first or second session. The common mean and median filters were initially used and later on an adapted or Hybrid filter was designed in MatLab© environment and in a similar philosophy to Gardiner’s Predictor filter. Taking into account the details of the study data, an Adaptive or Hybrid filter following the deployment of the optic nerve fibre layers of the retina was tested and selecting different weight depending on the locations of possible glaucoma defects.
Results: Initially, the used mean and median filters, used to remove noise of visual field provided ambitious results. The first filter blurred the edges and the overall appearance looked fuzzy or blurry. The second one calculated the values of the neighbourhood and set these in ascending array. Then selected the median of these values to replace the original one. The result in general looks misleading. Next, applying the Hybrid Adapted filter, the end results illustrated elimination of measured noise in the visual field tests and likely the first visit outcome could predict the third or the fifth visit one.
Conclusion: This is a promising approach to identify and eliminate measurement noise in the visual field tests and to predict, after filtering the first examination outcome, the likely visual field outcome of the third or the fifth visit. The challenge of predicting the progression of open angle glaucoma from the initial visit nowadays is even more than any other the “Holy Grail” of Perimetry.
Ruelle D. Diagnosis of dynamical systems with fluctuating parameters. Proc. R. Soc. London. 1987;A 413:5-8.
Baraniuk R. Signals and systems, the connexion project, Houston, USA; 2003.
Chandrinos A. Aspects of a perimetric learning index - Doctor of Philosophy Thesis, School of Optometry, Cardiff Un., Wales, UK; 2017.
Oppenheim A, Verghese G. Signals, systems and inference, MIT Class Notes, USA; 2010.
Barat V, Borodin Y, Kuzmin A. Intelligent AE signal filtering methods J. Acoustic Emission, , Moscow, 01000, Russia. 2010; 28:(129-131).
Gardiner SK, Fortune B, Demirel S. Signal-to-noise ratios for structural and functional tests in glaucoma. Transl Vis Sci Technol. 2013;2(6):3.
Jornitz MW. Filter construction and design. Adv Biochem Eng Biotechnol. 2006;98: 105-123.
Levy RV, Jornitz MW. Types of filtration. Adv Biochem Eng Biotechnol. 2006;98:1-26.
Senel HG, Peters RA 2nd, Dawant B. Topological median filters. IEEE Trans Image Process. 2002;11(2):89-104.
Green O. Efficient scalable median filtering using histogram-based operations. IEEE Trans Image Process. 2018;27(5):2217-2228.
Zeng B. Optimal median-type filtering under structural constraints. IEEE Trans Image Process. 1995;4(7):921-31.
Chandrinos A. Aspects of automated perimetry * publisher: LAP LAMBERT Academic Publishing; 2020. ISBN: 978-620-2-56290
Betz-Stablein BD, Morgan WH, House PH, Hazelton ML. Spatial modelling of visual field data for assessing glaucoma progression - investigative ophthalmology & visual science. 2013;54:1544-1553.
Gardiner SK, Crabb DP, Fitzke FW, Hitchings RA. Reducing noise in suspected glaucomatous visual fields by using a new spatial filter. Vision Research. 2004;839 – 848.
Gardiner SK, Swanson WH, Demirel S. The effect of limiting the range of perimetric sensitivities on pointwise assessment of visual field progression in glaucoma. Invest Ophthalmol Vis Sci. 2016;1:57(1):288-94.
Nevalainen J, Paetzold J, Papageorgiou E, Sample P, et al. Pamela A. Specification of progression in glaucomatous visual field loss, applying locally condensed stimulus arrangements Graefe's Archive for Clinical and Experimental Ophthalmology. 2009; 247:1659.
Kirwan JF , Hustler A, Bobat H, Toms L, Crabb DP, McNaught A. Portsmouth visual field database: An audit of glaucoma progression – Eye. 2014; 28:974–979.
Fitzke FW, Crabb DP, McNaught AI, et al. Image processing of computerised visual field data.: Br J Ophthalmol. 1995;79:207–12.
Crabb DP, Fitzke FW, McNaught AI, Edgar DF, Hitchings RA. Improving the prediction of visual field progression in glaucoma using spatial processing. Ophthalmology. 1997;104(3):517–524.
Crabb DP, Gardiner SK., Fitzke FW, Hitchings RA. A new spatial filter for visual field data p.31 Perimetry Update 2002/2003 Edited by D. Henson and M. Wall, The Kugler Publications, The Hague, Netherlands; 2003.
Morales E, de Leon JMS, Abdollahi N, Yu F, Nouri-Mahdavi K, Caprioli J. Enhancement of visual field predictions with pointwise exponential regression (PER) and pointwise linear regression (PLR) Translational Vision Science & Technology. 2016;5:12.
Strouthidis N, Scott A, Viswanathan A, Crabb D, Garway-Heath. Monitoring glaucomatous visual field progression: The effect of a novel spatial filter. Invest Ophthalmol Vis Sci. 2007;48:251-257.
Spry P, Johnson C. Identification of progressive glaucomatous visual field loss. Survey of Ophthalmology. 200;(47):158-173.
Garway-Heath DF, Poinoosawmy D, Fitzke FW, Hitchings RA. Mapping the visual field to the optic disc in normal tension glaucoma eyes. Ophthalmology. 2000; 107(10):1809-1815.
Chandrinos A, Tzamouranis DD. A review of learning effect in perimetry ophthalmology research An International Journal. 2020;12(2):23-30.
Chen Z, Hu K, Carpena P, Bernaola-Galvan P, Stanley HE, Ivanov PCh. Effect of nonlinear filters on detrended fluctuation analysis. Phys Rev E Stat Nonlin Soft Matter Phys. 2005;71(1,1):011104.