Q&A: Using natural images for the evaluation of distance image quality in IOLs

As premium intraocular lens (IOL) technology continues to evolve—from traditional monofocals to extended depth-of-focus and full visual range designs—so too must the methods used to evaluate their optical performance. Modulation transfer function (MTF) has long served as the benchmark for bench-based IOL assessment, but questions persist about how well single-frequency MTF metrics translate to the complex visual demands of real-world vision. A study presented at the 2026 Association for Research in Vision and Ophthalmology (ARVO) Annual Meeting in Denver takes a novel step forward, introducing natural image analysis as a complementary tool for evaluating distance image quality across 4 distinct IOL designs.

Aixa Alarcon, PhD, and co-authors Zhouping Lyu, Mark Jenkins, Sara El Aissati, Henk Weeber, and Carmen Canovas—researchers affiliated with Johnson & Johnson Vision—present findings that validate the area under the MTF curve (MTFa) as the strongest predictor of natural image quality, while also revealing how dysphotopsia effects and residual astigmatism interact with optical performance in ways that standard MTF reporting may not fully capture. In a post-conference interview with Modern Retina, Alarcon discussed the clinical implications of these findings for surgeons counseling patients on premium IOL selection.

Editor’s note: This interview was conducted via email and was lightly edited for style and clarity.

Why has MTF been the gold standard for evaluating IOL image quality, and what limitations led your team to explore natural image assessment as a complementary approach?

Aixa Alarcon, PhD: MTF is a standard measure of image quality that is widely used in the evaluation of lenses and optical systems. However, MTF is not a singular metric and can be measured under many different conditions. In the context of IOLs, it is important to measure MTF in a setting that is relevant to the clinical performance of a lens. In particular, clinically relevant conditions for an IOL means measuring using an eye model that incorporates the average chromatic and spherical aberration of the human cornea as well as using white light in the tests. Under such conditions, MTF metrics are well correlated with patients’ visual outcomes.^1,2 Other measurements conditions, such as measurements in green light, can be used to evaluate the quality of the lens for example during the manufacturing process but they can be deceptive and are not as well correlated with the visual performance of the patient.^3,4

MTF provides information of the contrast as a function of the spatial frequency, i.e. the contrast for specific levels of details of the image. However, our vision is a combination of multiple spatial frequencies and even when looking at a VA chart, low and high spatial frequencies contribute to the image. The goal of this study was to evaluate how differences in MTF transform into differences in natural images that generally contain a wider spatial frequency content. To illustrate this difference, Figure 1 shows the image of the moon collected for the new FVR of J&J and the trifocal IOL of another manufacturer as well as the MTF difference. Additionally, we wanted to evaluate which metrics based on MTF (e.g. the MTF at 50 c/mm or the area under the MTF) would provide a better correlation with the quality of a natural image.

How closely do your physical eye models approximate real-world optical conditions, and what are the trade-offs when moving from bench optics to models that mimic human eye physiology?

Aixa Alarcon, PhD: For this evaluation we used clinically relevant conditions, i.e, we measure in an eye model with the average chromatic and spherical aberration of the human cornea and used white light in the measurements. As reported in previous scientific publications^3,4, this is extremely important to be able to predict the clinical performance of the IOLs, especially when comparing lenses with different optical designs and materials with different chromatic dispersion. The relevance of these conditions has been also included in the latest version of the ISO standard (ISO 11979-2.2024 Annex C).This newest version of the standard describes these conditions as the ones to be used to perform the preclinical assessment of the potential benefits and risks associated with new IOL designs.

For clinicians more familiar with MTF curves and Snellen acuity, what does the cross-correlation coefficient actually measure, and why did you select it as your benchmark metric?

Aixa Alarcon, PhD: We selected the cross-correlation coefficient as a quantitative estimation of the overall quality of a natural image. This metric has been previously used in other studies to evaluate the quality of USAF targets⁵ and to predict clinical defocus curves². The value of the cross correlation is that it provides a single value for the overall quality of an image, accounting for the wide spatial frequency content, different levels of contrast and orientations. This is not the case for the MTF or for visual acuity.The MTF provides the contrast per spatial frequency while visual acuity relies on high contrast images.

Why does integrating the area under the MTF curve outperform sampling at a single spatial frequency as a predictor of real-world optical performance?

Aixa Alarcon, PhD: In a previous study² we found that the area under the MTF, that is a metric that accounts for multiple spatial frequencies, is highly correlated with the visual acuity (R² = 0.95) and that it outperformed the MTF at a single spatial frequency for predicting clinical defocus curves. We wanted to evaluate whether the area under the MTF could also predict the overall quality of more complex images with different levels of contrast and a wider spatial frequency content. The results of this study showed that, as found previously, the area under the MTF provided a better correlation with the quality of the image than the MTF.

You tested four IOL designs and found the rEDF and FVR delivered the best distance image quality—what optical characteristics drive that advantage, and how should surgeons apply those findings when counseling patients on lens selection?

Aixa Alarcon, PhD: Higher distance image quality means higher quality vision for the patient, that is, means good distance visual acuity and high contrast sensitivity. But a higher distance image quality has another additional benefit. It has been found that the combination of high quality distance vision and the extended depth of focus could drive the tolerance to refractive errors of an intraocular lens.⁶ The study of Black et al.⁶ showed that the rEDF provided high tolerance to small amounts of postoperative refractive errors, at the level of an enhanced aspheric monofocal IOL designed to slightly extend depth of focus. Excellent distance vision and contrast sensitivity, high patient satisfaction and a comparable dysphotopsia profile were demonstrated with both IOLs in the presence of post-operative refractive errors.

Lower gamma values weakened the MTFa-to-CCC correlation—what does that tell us about the role of dynamic range in IOL testing, and are there clinical scenarios where that variability would matter for patients?

Aixa Alarcon, PhD: Lower gamma values show the effect of dysphotopsias. Halos, starbursts and glare are phenomena that are associated with designs that provide range of vision and affect patients mostly in low light conditions in the presence of a glare source (such as the light of a car driving at night). The use of different gamma values to analyze the images illustrates the effect of these phenomena. The weakness of the correlation between MTF and the CCC in low gamma images shows that the effect of dysphotopsias is suppressed when image quality is evaluated using only the MTF.

This shows the value of the three components that determine distance image quality: visual acuity, contrast sensitivity and dysphotopsias. This also explains the high patient satisfaction obtained with a lens like the TECNIS PureSee IOL delivering a monofocal-like quality of distance vision in combination with the increased range of vision of an extended depth of focus IOL.^7,8

Astigmatism up to 0.75D preserved IOL performance rankings, but beyond that threshold differences began to blur—what are the implications for surgeons managing residual or surgically induced astigmatism in premium IOL patients?

Aixa Alarcon, PhD: Residual astigmatism has a negative effect on the vision of the patient. Therefore, surgeons should target the reduction of astigmatism to the minimum level, independently of the IOL design.

However, not all IOLs deliver the same tolerance⁹ and it is known that, for example, multifocal lenses are less tolerant to astigmatism than monofocal IOLs.¹⁰ The latest innovations introduced by J&J, the rEDF and FVR, are designed to improve that tolerance by implementing unique optical designs that increase the range of vision maintaining excellent distance.

What are the next steps in this research?

Aixa Alarcon, PhD: As far as we know, this is the first time that natural images are used to evaluate the image quality of IOLs. It highlights the relevance of image contrast and the impact that it has on complex images. We would like to expand this evaluation to more conditions and images with different contrast and spatial frequency content.

References:

1. Knorz, M. C. (1994). A theoretical model to predict contrast sensitivity with bifocal intraocular lenses. German Journal of Ophthalmology, 3(3), 189-194.

2. Alarcon, A., Canovas, C., Rosen, R., Weeber, H., Tsai, L., Hileman, K., & Piers, P. (2016). Preclinical metrics to predict through-focus visual acuity for pseudophakic patients. Biomedical optics express, 7(5), 1877-1888.

3. Weeber, H. A., Cánovas, C., Alarcón, A., & Piers, P. A. (2016). Laboratory-measured MTF of IOLs and clinical performance. Journal of Refractive Surgery, 32(3), 211-212.

4. Piers, P. A., Chang, D. H., Alarcón, A., & Cánovas, C. (2017). Clinically relevant interpretations of optical bench measurement of intraocular lenses. Journal of Refractive Surgery, 33(1), 64-64.

5. Kim, M. J., Zheleznyak, L., MacRae, S., Tchah, H., & Yoon, G. (2011). Objective evaluation of through-focus optical performance of presbyopia-correcting intraocular lenses using an optical bench system. Journal of Cataract & Refractive Surgery, 37(7), 1305-1312.

6. Black, D. A., Bala, C., Alarcon, A., & Vilupuru, S. (2024). Tolerance to refractive error with a new extended depth of focus intraocular lens. Eye, 38(Suppl 1), 15-20.

7. Corbett, D., Black, D., Roberts, T. V., Cronin, B., Gunn, D., Bala, C., ... & Vilupuru, S. (2024). Quality of vision clinical outcomes for a new fully-refractive extended depth of focus Intraocular Lens. Eye, 38(Suppl 1), 9-14.

8. Auffarth, G. U., Negoescu, A. T., Kremser, F., Chychko, L., Vogormian, L., Khoramnia, R., & Yildirim, T. M. (2026). Clinical Performance After Implantation of a New Purely Refractive Extended Depth of Field Intraocular Lens in Cataract Patients. Clinical Ophthalmology, 583059.

9. Carones, F. (2017). Residual astigmatism threshold and patient satisfaction with bifocal, trifocal and extended range of vision intraocular lenses (IOLs). Open J Ophthalmol, 7(1), 1-7.

10. Hayashi, K., Hayashi, H., Nakao, F., & Hayashi, F. (2000). Influence of astigmatism on multifocal and monofocal intraocular lenses. American journal of ophthalmology, 130(4), 477-482.