Objective measurement of cortical and subcapsular opacification in retroillumination photographs

Computer-aided assessment of diagnostic images for epidemiological research

Background

Diagnostic images are often assessed for clinical outcomes using subjective methods, which are limited by the skill of the reviewer. Computer-aided diagnosis (CAD) algorithms that assist reviewers in their decisions concerning outcomes have been developed to increase sensitivity and specificity in the clinical setting. However, these systems have not been well utilized in research settings to improve the measurement of clinical endpoints. Reductions in bias through their use could have important implications for etiologic research.

Methods

Using the example of cortical cataract detection, we developed an algorithm for assisting a reviewer in evaluating digital images for the presence and severity of lesions. Available image processing and statistical methods that were easily implementable were used as the basis for the CAD algorithm. The performance of the system was compared to the subjective assessment of five reviewers using 60 simulated images. Cortical cataract severity scores from 0 to 16 were assigned to the images by the reviewers and the CAD system, with each image assessed twice to obtain a measure of variability. Image characteristics that affected reviewer bias were also assessed by systematically varying the appearance of the simulated images.

Results

The algorithm yielded severity scores with smaller bias on images where cataract severity was mild to moderate (approximately ≤ 6/16^ths). On high severity images, the bias of the CAD system exceeded that of the reviewers. The variability of the CAD system was zero on repeated images but ranged from 0.48 to 1.22 for the reviewers. The direction and magnitude of the bias exhibited by the reviewers was a function of the number of cataract opacities, the shape and the contrast of the lesions in the simulated images.

Conclusion

CAD systems are feasible to implement with available software and can be valuable when medical images contain exposure or outcome information for etiologic research. Our results indicate that such systems have the potential to decrease bias and discriminate very small changes in disease severity. Simulated images are a tool that can be used to assess performance of a CAD system when a gold standard is not available.

Wavefront analysis of higher-order aberrations in patients with cataract

PURPOSE

To determine local refractive changes and higher-order aberrations in patients with nuclear or cortical cataract.

SETTING

Osaka University Medical School, Osaka, Japan.

METHODS

Wavefront analysis of both ocular and corneal aberrations was performed with the Hartmann-Shack aberrometer in 2 patients, a 22-year-old woman with bilateral developmental nuclear cataract and a 68-year-old woman with mild bilateral cortical cataract.

RESULTS

Case 1 showed a delay in the wavefront that caused a myopic shift in the central pupillary area in both eyes, associated with the nuclear cataract. The spherical-like aberration (right eye, 36%; left eye, 21%) was greater than the coma-like aberration in both eyes. Case 2 showed an advancement of the wavefront that caused a hyperopic shift, especially in the lower temporal pupillary area, that was associated with the cortical cataract. The coma-like aberration (right eye, 63%; left eye, 52%) was greater than the spherical-like aberration in both eyes. The polarity of the third-order spherical aberration was negative in Case 1 and positive in Case 2. Corneal higher-order aberrations were small and had a different distribution than ocular higher-order aberrations in both patients.

CONCLUSIONS

The Hartmann-Shack aberrometer was useful in detecting local refractive changes and higher-order aberrations in patients with mild cataract. The polarity and the absolute value of ocular higher-order aberrations may be useful parameters to characterize eyes with cataract.

[Late effects of radiation on the eye and ocular adnexa]

Late ocular effects after irradiation of the eyes and ocular adnexa are reviewed. Ocular and orbital injuries occur more often after irradiation of central nervous system neoplasms or treatment of paranasal sinus malignancies. The complete loss of vision is the major complication which is multifactorial. This paper describes the radiation effects and grading of ocular adnexa, lens, retina, optic nerves and orbit. The tolerance doses and treatment are described.

A longitudinal study of cortical cataracts using retroillumination photographs

The National Eye Institute (NEI) computer planimetry system has been shown to have good reproducibility in assessing the size of cortical cataracts from retroilluminatin photographs. In this study, we determined the usefulness of this system in monitoring cortical cataract changes over time. Using the Neitz Kawara retroillumination camera, retroillumination photographs of cortical cataracts in 81 eyes were obtained every six months for an average of 31 months. As previously described, the cortical cataract outlines were traced in a masked fashion onto transparent plastic overlays, and the tracings then digitized into a Macintosh Quadra computer using a computer scanner. Cortical cataract area was then determined using a specially developed software program. For each eye, the rate of cortical area change was determined by the slope of the regression line fitted to the follow-up measurements. Cataract progression was classified to be significant if the slope exceeded a critical value. Of the 81 eyes, 24 (30%) had significant cataract progression, while 57 (70%) did not progress. This study presents data on progression of the area of cortical cataracts and suggests the usefulness of the NEI computer planimetry system for monitoring such changes from retroillumination photographs. The slope-based test can also play a useful part in longitudinal studies with irregular time intervals and variable number of visits.

Late effects of radiation on the eye and ocular adnexa

A clinically useful classification system is suggested that can be used in prospective trials to evaluate the effects of radiation on the visual system. We review radiation-induced pathophysiological and clinical changes of the various ocular structures as well as dose-response data and management of ocular complications. The rationale for the classification scheme chosen is also discussed.

Quantitative analysis of retroillumination images

We have developed a semi-automated image processing system for analysis and evaluation of retroillumination images. This paper describes methods used to compensate for illumination variations in the images, separation of data into cataractous and non-cataractous portions, how quantitative measurements are made and how they assess the pathological condition. In addition to the traditional area measurement, this system computes the net integral of density and several measurements involving the location of the opacity in relation to the pupillary margin. The computer measures of area, integral of density, area centrality, weighted area, density centrality and weighted density provide more data than previously described systems. Data produced by this interactive and automated system can be used in studies of posterior subcapsular and cortical cataracts, and to study the effect of these opacities on vision.

Measurement error in assessing the size of cortical cataracts from retroillumination photographs

This study describes a new method of quantifying anteriorly located cortical cataracts using retroillumination photographs and computer planimetry. Cortical cataracts were graded clinically and then photographed using the Neitz retroillumination camera twice by each of 2 photographers. The cataract outlines were traced onto a transparent overlay, and computer planimetry was performed using a Scan Maker 600ZS, a MAC II Computer and specially developed software. We estimated the measurement error of the method and its associated effect on sample size estimates for clinical studies. We calculated that the variability in this technique would contribute about 21 additional subjects to overall sample size estimates in studies comparing the mean areas of cortical opacities. In many studies this would be a small addition to total sample size requirements. This technique provides clinically useful measurements of the size of a cortical opacity as seen on a retroillumination photograph. This may be useful for future clinical studies on natural progression of cortical cataracts as well as for clinical trials of anticataract drugs.

Reproducibility study of posterior subcapsular opacities on the NEI retroillumination image analysis system

We developed a semi-automated retroillumination image analysis system which combines speed, ease of operation and interactive analysis. The system measures cataract area and integral of cataract density (ID). For system reproducibility evaluation, 20 eyes with posterior subcapsular opacities were captured twice by two photographers. Variability was estimated under a random effects analysis of variance model. Measurement errors for area and for ID were each small contributors to total variability (the sum of variability between study eyes plus measurement error), being 0.4% and 0.1% respectively. The largest contributor to area measurement error was image analysis variability (97%). For ID measurement error, the variability in images (44%) and in image analysis (46%) were major contributors. The reproducibility is comparable to previously described retroillumination analysis systems. This easy to use system may therefore be useful in clinical research studies including possible clinical trials of anti-cataract drugs.

Analysis of retro-illumination photographs for use in longitudinal studies of cataract

Retro-illumination photography has potential for the monitoring of cortical and posterior subcapsular cataract formation. Variations in the individual illumination results have limited accurate quantification of opacities within such images. We present a new image analysis technique which minimises the effect of uneven and varying retro-illumination. The new technique has been tested for variation between photographers and over a short time period. We believe it is of value in lens assessment in combination with a semi-quantitative grading system.