Ординатура / Офтальмология / Учебные материалы / Orthokeratology Principles and Practice 2004

38 ORTHOKERATOLOGY

temporally from the map center. The flat plateau in the center (green) demonstrates that the central cornea postorthokeratology is spherical followed by depression (or relative steepening). Spherical difference elevation maps therefore show the radial difference between a best-fit sphere and the cornea. Ellipsoidal difference height maps are even more advantageous as corneal shape is better described according to ellipses (rather than spheres); in fact, a useful feature would be to view the difference between pretreatment elevation and posttreatment elevation. Ellipsoidal difference elevation maps are available in the Euclid ET-800 topographer (Euclid Systems Corporation).

The surface regularityindexandsurface asymmetry index

Computer algorithms may be derived in order to calculate indices that complement the data from contour maps. Two such indices, the surface regularity index (SRI) and the surface asymmetry index (SAI), have been assessed in two clinical studies by Dingledein et al (1989) and Wilson & Klyce (1991).

The SRI determines the central corneal optical quality; it is a measure of localized surface regu-

Figure 2.22 Spherical difference elevation map of a patient who has undergone orthokeratology treatment with a Contex OK704CYF lens.

larity. The lower the value of this index, the smoother the surface. Thus, for a perfectly smooth surface a theoretical value of zero would be found. However, in practice this is not possible due to instrument errors. Wilson & Klyce (1991) found a high correlation between the SRI and best-corrected spectacle acuity (r = 0.8, P < 0.001) in 31 eyes that met their criteria for inclusion. Clinically, the SRI may be used to differentiate between reduced visual acuity due to factors other than corneal topography.

The SAl determines the asymmetry of the central corneal surface power. The value represents the centrally weighted summation of differences in corneal power 180 0 apart over 128 equally spaced meridians. For a perfectly regular surface, such as a sphere, a theoretical value of zero would be found. Again, due to instrument errors, this is not the case. Dingledein et al (1989) found a reasonable correlation between SAl and best spectacle corrected visual acuity (r = 0.76, P < 0.001) in 39 eyes with keratoconus, compound myopic astigmatism, epikeratophakia, and two corneas from patients with 20 / 20 vision. The differences in values of SAl for normals and those with keratoconus with best-corrected spectacle

acuity of 20 / 20 were statistically significant (P < 0.001). In a more recent study by Wilson & Klyce (1991), a relatively low correlation was found between this index and best-corrected spectacle acuity (r = 0.62, P < 0.005). This discrepancy may be due to the relatively small sample sizes. Clinically, the SAl may be used as a quantitative indicator for monitoring changes in corneal topography. The derivation of these indices is described by Wilson & Klyce (1991). They suggest that the incorporation of these indices would be a useful tool when combined with color-coded maps for the assessment of corneal topography.

Quantitative descriptors of corneal topography

For purposes such as contact lens fitting and surgical modification of the corneal surface, accurate knowledge of the corneal surface parameters is essential. Qualitative assessment is of use only to summarize the enormous amount of data derived from corneal topographic systems. Quantitative descriptors in the form of mathematical functions may be useful to describe the corneal surface whilst allowing the practitioner to retain the concrete information provided from videokeratoscopy. The cornea may be described mathematically in one of three ways: curvature, position, and slope. However, there is no agreement as to the appropriate method of describing corneal shape. Some of these mathematical descriptors are discussed below.

Corneal radius andshape

The radius of curvature of the anterior corneal surface and its refractive index determine the power of that surface. The smaller the radius, the greater the refracting power of that surface. The apical radius of the cornea defines the size of the corneal profile. Guillon et al (1986) showed a variety of corneal shapes within the normal population. Therefore, mathematical descriptors that show a continuous change in curvature would more accurately model the shape of the cornea. It appears that descriptions have taken two separate forms: certain researchers have provided precise descriptions using complex polynomial formulae

CORNEAL TOPOGRAPHY AND ITS MEASUREMENT 39

(Howland et aI1992), whereas others have approximated corneal contour to conic sections (Townsley 1970, 1974, Bibby 1976, Guillon et al 1986). Generally, the results of these studies have shown that modeling the corneal surface in terms of second-order polynomials (conics) is acceptable. The family of second-order polynomial curves used as descriptors of the cornea are known as conics. Mathematically, they are defined as follows:

Equation 2.3

where a and b are the semimajor and semiminor axes respectively. x is the sagittal depth at a chord length of y.

The family of surfaces and curves that may be derived from a cone (a solid surface produced by the revolution of either of two straight lines meeting at a point about the other) are defined below:

•Conicoids: on rotating conic sections about an axis of symmetry, one produces surfaces known as conicoids.

•Conic sections: these two-dimensional sections are defined according to two factors - the apical radius of curvature and a term known as the eccentricity (e). Baker (1943) derived an equation to describe a conic section:

where x and yare the Cartesian coordinates (the origin is conveniently placed at the corneal apex), ro is the apical radius of curvature, and p is an index of peripheral flattening (the shape factor) and indicates the level of asphericity. For example, p < 0, hyperbola; p = 0, parabola; a< p < 1, oblate (flattening) ellipse; p = 1, sphere; p > 1, prolate (steepening) ellipse. Bennett (1968) has shown a direct relationship between p and e where

The advantage of Baker's and Bennett's notation is that the formula is simplified and it is capable of describing all conic sections. A simple mathematical relationship can be shown to link Equations 2.3 and 2.4, as shown in Equation 2.6.

a a2

40 ORTHOKERATOlOGY

Manufacturers of topography systems have used this information to produce numeric maps (as in the EyeSys topographer). These maps display apical radius, eccentricity, and peripheral sagittal radii of curvature. Known as numeric maps, they provide the orthokeratologist with sufficient information to derive the initial trial lens.

Unfortunately, there may not be sufficient repeatability within instruments and between instruments to derive the same eccentricity value consistently within the same cornea. The reason for this is that, although systems may be accurate for nonbiological surfaces, the Placido ring reflection off the corneal surface may result in difficulty in consistently imaging the rings from the peripheral cornea. As a result, variability in e-values will occur. Another factor that may be responsible for a lack of agreement between different instruments is that some topography systems derive e-values within a specific zone, such as the Keratron (Optikon, Italy).

High-order polynomial descriptors

The corneal surface is a complex shape. In order to describe it mathematically high-order polynomial expressions may be used. A study by Howland et al (1992) attempted to describe corneal shape in precisely this manner. Using the corneal TMS (Computed Anatomy Inc., New York), fourth-order polynomial expressions from the Taylor series were fitted to corneal surface coordinates. From the expressions, the mean corneal curvature (MCC) at any point on the surface was computed and compared with the measured curvature derived from the TMS.

Howland et al (1992) concluded that polynomials captured gross features of curvature but did not resolve sufficient detail. However, from their study they suggested that the use of polynomial fitting could form a classification scheme for describing corneal topography.

The use of corneal topography in orthokeratology

The principal application of corneal topography for eye-care practitioners is contact lens fitting, corneal screening, and refractive surgery. Most

topography systems now have contact lens modules that use information acquired from topography measurement in order to determine the initial lens for rigid lens fitting. Practitioners may also specify the level of apical clearance or even design custom lenses for patients. Szczotska (1997) clinically evaluated the EyeSys Pro-Fit software (version 3.1) in terms of its efficiency in fitting RGP lenses to 22 normal subjects. Comparing the EyeSys fitting to manually fitted patients showed identical success rates (77%) with no subsequent modification of lens fit. However, EyeSys-based fitting reduced chair time by 51.4%. The author concluded that EyeSys-based rigid gas-permeable (RGP) lens fitting improved efficiency in normal eyes.

Corneal topographical analysis has proved to be invaluable and essential to the practice of orthokeratology. As with refractive surgery, practitioners involved with any procedure introducing a change to the corneal shape must have access to a topographical device. The primary aim of using such a device is twofold: firstly, to be able to record baseline data that will enable the practitioner to choose a suitable trial lens and secondly, so that any change in corneal shape can be accurately documented. The use of difference maps has greatly enhanced the understanding of lensinduced changes that occur in orthokeratology.

Difference maps and application of topography to orthokeratology

Figure 2.23A shows an example of a difference map. All difference maps show three maps: the top left (in Fig. 2.23A) is the pretreatment sagittal topography plot, the lower left the posttreatment sagittal plot, and the right map shows the point- by-point difference in sagittal radius from preto posttreatment. The increments on the numeric scale on the right can be set to diopters or in millimeters depending on practitioner preference. It shows the change in corneal curvature from preto posttreatment.

As well as being an objective measure of the change induced during orthokeratology, difference maps can also inform the practitioner of the effect of the fit of the contact lens on the corneal surface. Figure 2.23A shows a difference map

where the relative inferior steepening implies that the fitted lens displaced upward and hence flattened the superior cornea whilst secondarily inducing inferior steepening. Lenses inducing such a change in topography generally have a sag that is less than the sag of the cornea at an equivalent chord diameter (i.e., the lens is too flat). Conversely, Figure 2.238 shows an ideal response to orthokeratology where the lens has remained centered and induced a smooth central area of flattening without inducing corneal distortion. Note that the change in refractive error is of the order of 2.00 D. Figure 2.23C shows an area of

central steepening; lenses inducing this type of topographic change are usually too steep.

As will be discussed in Chapter 3, lens binding following overnight wear is common. However, in nearly all cases lenses tend to become mobile after a period of eye opening. Unfortunately, lenses may occasionally bind and induce corneal distortion. The topographic hallmark of binding-induced distortion is a sharp ring of localized steepening (Fig. 2.24). This area of steepening (usually red in color) corresponds to epithelial indentation at the site of lens binding. In most cases, binding occurs in the superior area of the cornea.

The modem approach to orthokeratology lens fitting is based on matching the corneal sag to the sag of the contact lens for reverse geometry lenses. Current programs used to perform such calculations require the use of the apical radius of the cornea and the corneal eccentricity. Almost all modem topography systems display these parameters. However, the practitioner must be aware

whether the mean (global) corneal eccentricity is derived or whether the eccentricity over a specific zone or meridian is derived, as this can result in significant discrepancies with respect to the final predicted lens.

Currently, there are no publications directly relating to the accuracy of corneal topographic systems in predicting accurate orthokeratology

lenses. However, the Keratron topographer has the facility for the practitioner to perform sagittalbased fitting on tricurve lenses. All that needs to be entered into the contact lens module of the software is the lens / corneal clearance at the back optic zone diameter (BOZO), the tear reservoir, and the lens edge in order to produce a required tear layer profile (Fig. 2.25A). The computer then derives the contact lens curves that produce the required tear layer profile and subsequently the predicted sodium fluorescein pattern is displayed (Fig. 2.25B). It must be stated that, although in principle this appears to be a logical way of designing a lens, the practitioner's choice of tear reservoir height and back optic zone radius (BOZR) would not then be determined in a scientific manner. The Keratron therefore has no advantages over other topographic systems provided one uses a computer program to design the lenses (as is generally the case).

Which topographer is most suitable?

Most orthokeratologists would agree that it is essential to have a topographic device when embarking on orthokeratology treatment. There are numerous devices available in the market place, so which one is most appropriate? Ideally, one requires a device that is accurate and repeatable. Dave et al (1998a) showed that the EyeSys (model II) topographer exhibited progressively

poorer repeatability in the peripheral cornea. Furthermore, the variability of repeatability was also dependent on the corneal meridian such that greatest repeatability was found in the temporal meridian and poorest repeatability in the superior and nasal corneal meridians. The authors suggested that this meridional dependence was due to the ocular adnexa (lids, tear film, etc.) and not simply a factor associated with the videokeratoscope. In another study by Dave et al (l998b), the accuracy of the EyeSys corneal topographer was assessed using various aspheric (P-values of 0.8 and 0.5) and spherical surfaces. They found that the EyeSys topographer had a high level of accuracy and repeatability. However, the accuracy reduced with increasingly flattening surfaces, i.e., as the P-values decreased (Tables 2.5 and 2.6). However, as expected, the repeatability was exceptionally high and constant for all surfaces

Table 2.5 Accuracy of measuring aspheric surfaces using the EyeSys topographer (after Dave et al

1998b)

44 ORTHOKERATOLOGY

Table 2.6 Repeatability of the EyeSys topographer for spherical and aspheric surfaces (after Dave et al

1998b)

Table 2.7 Accuracy and precision (precision values in brackets) of four topographers in measuring six artificial surfaces (afterTang et al 2000)

TMS, Topographic Modeling System.

measured, thus confirming that the biological structure of the cornea and adnexa reduce repeatability of topographers in a clinical environment.

Another requirement for topographic devices is that the device will not exhibit any bias towards any type of surface structure, i.e., it should be able to measure a normal nontreated eye with the same accuracy as any eye that has undergone treatment. In an interesting study by Tang et al (2000), the accuracy (root mean square error) and precision (maximum standard error) of four topographers using six test surfaces were determined. The surfaces comprised a sphere, asphere, multicurve, and three bicurve surfaces. The results are summarized in Table 2.7.

The Medmont exhibited greatest accuracy and precision in measuring the aspheric surface. This surface represented the "normal" cornea. The TMS showed greatest accuracy in measuring bicurve surfaces (symbolizing the profile of the cornea postrefractive surgery and orthokeratol- ogy) followed by the Medmont and Keratron;

however, the Medmont had the greatest precision in measuring bicurve surfaces. This study indicates that the algorithms and design of different topographers result in bias towards different surfaces. For orthokeratology it is most important to determine pretreatment corneal shape (or sag) and therefore the practitioner must choose a topographer that has the highest accuracy and precision in measuring the "normal" cornea.

Orthokeratology fitting is considerably different from fitting conventional geometry rigid lenses. The apical clearance in conventional lenses is of the order of 15-20 urn: however, orthokeratology lenses are fitted to have an apical clearance of the order of 2-6 urn (mode of 4 urn). Therefore, in order to fit a lens within 4 urn one requires a precision of 2 urn, Hough & Edwards (1999) utilized a well-known formula used to derive sample size for clinical trials (see Equation 2.7). Knowing the standard error (precision) of fitting required and the repeatability of the topographer, one can derive the number of readings required to achieve this defined precision.

Equation 2.7

where (J is the standard deviation of repeatability of the topographer and e the standard error or precision required.

Cho et al (2002) utilized this formula to determine the number of measurements required for four topographers (Dicon CT200, Humphrey ATLAS 991, Medmont E300, Orbscan II) for orthokeratology fitting with a precision of 3 ....m. The authors found that the Dicon required 64 repeat readings, Humphrey 12, Medmont two, and Orbscan 552. Practitioners can themselves determine the number of repeat readings required by their topographer by deriving the standard deviation of repeatability for elevation over a 9 mm chord and inputting the value into Equation 2.7.

SUMMARY

Eye-care practitioners have always been aware of the need to understand the topographical nature of the corneal surface. Within the last decade, researchers in this field have made a number of

significant improvements to older systems. Data are now stored as digitized images on computer disks to enable easy access and to eliminate errors such as shrinkage and distortion that occurred with older photographic storage media. Accurate localization of the target rings (using techniques which allow the reflected Placido rings to be located accurately) from digitized images has also aided the increase in accuracy compared to instruments where ring images were located manually. The improvement of target design (by having a greater number of rings and by modifying the spatial arrangement of the rings) has permitted greater detailed analysis of the cornea. Representation of the data has now evolved into detailed mathematical descriptors and schematic color-coded topographical maps. Modern topography devices use sophisticated algorithms that no longer bias corneal curvature to spherical surfaces, thereby improving their accuracy. However, with our greater understanding, we have also discovered limitations of current display options such as sagittal and tangential

REFERENCES

maps as computed curvature measurements are dependent on fixation and corneal symmetry. Investigators have shown that Gaussian curvature, elevation maps, and mean curvature maps permit independence of fixation axis. Currently, the Orbscan system allows practitioners to view these types of curvature maps, enabling greater localization of corneal abnormalities.

There are numerous applications for the use of topographical systems in areas such as contact lens fitting (Hodd & Ruston 1993, McCarey et aI1993), diagnosis and monitoring of keratoconus (Maguire & Bourne 1989, Maguire & Lowry 1991, Wilson et al 1991), monitoring of corneal shape after refractive surgery (Maguire et aI1987b; McDonnell et al 1989),corneal grafting, and postoperative management of cataract patients. Practitioners involved in orthokeratology must make use of moderngeneration videokeratoscopes so that changes in corneal shape can be accurately documented and the necessary prescreening performed. The development of topography systems will undoubtedly be in the area of computer software.

Andersen J, Koch-Jensen P,Osterby 0 (1993a) Corneal topography: image processing and numerical analysis of keratoscopy. Acta Ophthalmologica 71: 151-159

Andersen J, Koch-Jensen P, Osterby 0 (1993b) Corneal topography: photokeratoscopy including the central region. Acta Ophthalmologica 71: 145-150

Antalis J J, Lembach R G, Carney L G (1993)A comparison of the TMS-1 and the Corneal Analysis System for the evaluation of abnormal corneas. CLAO Journal 19: 58-63

Applegate R A (1992)Optical and clinical issues in the measurement of corneal topography. Ophthalmic and Visual Optical Technical Digest

Arffa R C, Warnicki J W, Rehkopf P G (1989)Corneal topography using rasterstereography. Refractive Corneal Surgery 5: 414-417

Baker T Y (1943) Ray tracing through non-spherical surfaces. Proceedings of the Physics Society 55: 361-364

Barsky BA, Klein S A, Garcia D D (1997)Gaussian power with cylinder vector field representation for corneal topography maps. Optometry and Vision Science 74:917-925

Belin M W, Zloty P (1993)Accuracy of the PAR corneal topography system with spatial alignment. CLAO Journal 19: 64-68

Belin M W, Litoff D, Strods S J, Winn S S, Smith R S (1992) The PAR technology corneal topography system. Refractive Corneal Surgery 8: 88-96

Bennett A G (1968) Part I Aspherical contact lens surfaces. Ophthalmic Optician: 1037-1040

Bibby M M (1976) Computer assisted photokeratoscopy and contact lens design. Optician 171(4423):37-44

Bland J M, Altman D G (1986) Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 1: 307-310

Bogan S J, Waring I G 0, Ibrahim 0, Drews C, Curtis L (1990) Classification of normal corneal topography based on computer-assisted videokeratography. Archives of Ophthalmology 108: 945-949

Bonnet R, Cochet D (1962) New method of topographical ophthalmometry - its theoretical and clinical applications. American Journal of Optometry 39: 227

Cho P, Lam A, Mountford J, Ng L (2002) The performance of four different corneal topographers on normal human corneas and its impact on orthokeratology lens fitting. Optometry and Vision Science 79(3): 175-183

Dave T N (1995) Evaluation of videokeratoscopy. PhD thesis. Aston University, Birmingham, UK

Dave T N, Ruston D M, Fowler C W (1998a) Evaluation of the EyeSys model II computerized videokeratoscope. Part I: clinical assessment. Optometry and Vision and Science 75:647-655

Dave T N, Ruston D M, Fowler C W (1998b) Evaluation of the EyeSys model II computerized videokeratoscope. Part II: the repeatability and accuracy in measuring convex aspheric surfaces. Optometry and Vision Science 75: 656-662

46 ORTHOKERATOLOGY

Dingledein SA, Klyce S D, Wilson S E (1989)Quantitative descriptors of corneal shape derived from computerassisted analysis of photokeratographs. Refractive Corneal Surgery 5: 372-378

Doss J D, Hutson R L, Rowsley J J, Brown D R (1981) Method for calculation of corneal profile and power distribution. Archives of Ophthalmology 99: 1261-1265

Edmund C (1985) The central-peripheral radius of the normal corneal curvature. A photokeratoscopic study. Acta Ophthalmologica 63: 670-677

Edmund C (1987) Location of the corneal apex and its influence in the stability of the central corneal curvature. A photokeratoscopy study. American Journal of Optometry and Physiological Optics 64: 846-852

El Hage S G (1971)Suggested new methods for photokeratoscopy: a comparison for their validities. Part I. American Journal of Optometry and Physiological Optics 48: 897-912

EI Hage S G (1972)Differential equation for the use of the diffused ring keratoscope. American Journal of Optometry and Physiological Optics 49: 422--436

Fincham E F (1953) New photokeratoscope utilizing a hemispherical object surface. Medical and Biological Illustration 3: 87

Gormley D J, Gersten M, Koplin R S, Lubkin V (1988) Corneal modelling. Cornea 7: 30-35

Guillon M, Lyndon D 1', Wilson C (1986)Corneal topography: a clinical model. Ophthalmology and Physiological Optics 6: 47-56

Gullstrand A, Ludlum P (1966) Photographicophthalmometric and clinical investigations of corneal refraction. American Journal of Optometry 43: 43

Hannush S B, Crawford S L, Waring G 0, Gemmill M C, Lynn M J, Nizam A (1989) Accuracy and precision of keratometry, photokeratoscopy, and corneal modeling on calibrated steel balls. Archives of Ophthalmology 107: 1235-1239

Hannush N B, Crawford S L, Waring G 0, Gemmill M C Lynn M J, Nizam A (1990) Reproducibility of normal corneal power measurements with a keratometer, photokeratoscope and video imaging system. Archives of Ophthalmology 108: 539-544

Hodd N F B, Ruston D M (1993) The EyeSys corneal analysis system: its value in contact lens practice. Optometry Today 33(17): 12-22

Hough T. Edwards K (1999) The reproducibility of videokeratoscope measurements as applied to the human cornea. Contact Lens and Anterior Eye 22: 91-99

Howland H C, Glasser A, Applegate R (1992) Polynomial approximations of corneal surfaces and corneal curvature topography. Ophthalmic and Visual Optics Technical Digest 3: 34-36

Hubbe R E (1994)The effect of poor fixation on computerassisted topographic corneal analysis. Ophthalmology 101:1745-1748

[eandervin M, Barr J (1998)Comparison of repeat videokeratography: repeatability and accuracy. Optometry and Vision Science 75(9):663-669

Klein S A (1993) Improvements for video-keratography. Ophthalmic and Visual Optics Technical Digest

Klein S A (1997) Axial curvature and the skew ray error in corneal topography. Optometry and Vision Science 74: 931-944

Klyce S D (1984) Computer-assisted corneal topography. Investigations in Ophthalmology and Visual Science 25: 1426-1435

Knoll H A (1961) Corneal contours in the general population as revealed by the photokeratoscope. American Journal of Optometry and Archives of the American Academy of Optometry 38: 389-397

Knoll H A, Stimson R, Weeks C L (1957) New photokeratoscope utilizing a hemispherical object surface. Journal of the Optical Society of America 47: 221

Koch D, Foulks G N, Moran C T, Wakil J S (1989) The corneal EyeSys system: accuracy analysis and reproducibility of first-generation prototype. Refractive Corneal Surgery 5: 424--429

Koch D D, Wakil J S, Samuelson S W, Haft E A (1992) Comparison of the accuracy and reproducibility of the keratometer and the EyeSys corneal analysis system model I. Journal of Cataract and Refractive Surgery 18: 342-347

Ludlum W M, Wittenburg S, Rosenthale J, Harris G (1967) Photographic analysis of the ocular dioptric components. American Journal of Optometry and Archives of the American Academy of Optometry 44: 276-296

Maguire LJ, Bourne W M (1989) Corneal topography of early keratoconus. American Journal of Ophthalmology 108: 107-112

Maguire L J, Lowry J C (1991) Identifying progression of subclinical keratoconus by serial topography analysis. American Journal of Ophthalmology 112:41--45

Maguire L J, Singer D E, Klyce S D (1987a) Graphic presentations of computer-analyzed keratoscope photographs. Archives of Ophthalmology 105: 223-230

Maguire L J, Klyce S D, Singer D E, McDonald M B, Kaufman HE (1987b) Corneal topography in myopic patients undergoing epikeratophakia. American Journal of Ophthalmology 103: 404--416

Mandell R B (1992) The enigma of the corneal contour. CLAO Journal 18: 267-273

Mandell R B, St Helen R (1965) Position and curvature of the corneal apex. American Journal of Optometry 46: 25-29

Mandell R B, St Helen R (1968) Stability of the corneal contour. American Journal of Optometry and Archives of the American Academy of Optometry 12: 797-806

McCarey B E, Amos C F,Taub L R (1993)Surface topography of soft contact lenses for neutralizing astigmatism. CLAO Journal 19: 114

McDonnell P J, McClusky D J, Garbus J J (1989)Corneal topography and fluctuating visual acuity after radial keratotomy. Ophthalmology 96: 665-670

McMahon T T, Anderson R J, Joslin C E, Rosas G A (2001) Precision of three topography instruments in keratoconus subjects. Optometry and Vision Science 78(8): 599-604

Miller D C J (1980) A proposed new division of corneal division of corneal functions. New York, Raven Press

Nieves J E, Applegate R A (1992) Alignment errors and working distance directly influence the accuracy of corneal topography measurements. ARVO abstracts. Investigations in Ophthalmology and Visual Science

Schultze R L (1998) Accuracy of corneal elevation with four corneal topography systems. Journal of Refractive Surgery 14(2): 100-104

Stone J (1962) The validity of some existing methods of measuring corneal contour compared with suggested new methods. British Journal of Physical Optics 19: 205-230

Szczotska L B (1997) Clinical evaluation of a topopographic based contact lens fitting software. Optometry and Vision Science 74: 14-19

Tang W, Collins M J, Carney L, Davis B (2000) The accuracy and precision performance of four videokeratoscopes in measuring test surfaces. Optometry and Vision Science 77:483-491

Townsley M (1967) New equipment and methods for determining the contour of the human cornea. Contacto

11:72

CORNEAL TOPOGRAPHY AND ITS MEASUREMENT 47

Townsley M (1970) New knowledge of the corneal contour. Contacto 14: 38-44

Townsley M (1974) The graphic presentation of corneal contours. Contacto 18: 24-32

Tsilimbaris M K, Vlachonikolis I G, Siganos D, Makridakas G, Pallikaris I G (1991) Comparison of keratometric readings as obtained by [aval ophthalmometer and corneal analysis system (EyeSys). Refractive Corneal Surgery 7: 368-373

Wang J, Rice D A, Klyce S D (1989) A new reconstruction algorithm for improvement of corneal topographical corneal analysis. Refractive Corneal Surgery 5: 379-387

Waring G 0 (1989) Making sense of keratospeak. Proposed conventional terminology for corneal topography. Refractive Corneal Surgery 5: 362-367

Wilson S E, Klyce S D (1991) Quantitative descriptors of corneal topography. Archives of Ophthalmology 109: 349-353

Wilson S E, Lin D T C Klyce S D (1991) Corneal topography of keratoconus. Cornea 10: 2-8