Publication: Perception of static and dynamic blur for developing clinical instrumentation in optometry and ophthalmology
Loading...
Identifiers
Publication date
2022-11
Defense date
2022-11-25
Authors
Advisors
Tutors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Blur degrades our retinal images and affects our vision, more and more as we old. Understanding blur is key in the study of many processes in visual perception, from the development of the visual system to the diagnosis and compensation of refractive errors (myopia, hyperopia, astigmatism, and presbyopia). Blur also provides cues for accommodation (focusing near objects) or depth estimation.
Besides refractive errors, other sources of visual blur are high-order geometrical aberrations, chromatic aberrations, and some presbyopia corrections such as multifocality or monovision. In monovision, one eye is corrected for far vision and the other one for near. By design, monovision corrections produce interocular differences in blur, affecting binocular vision.
Optotunable lenses, programmable lenses able to change their optical power very quickly (in the order of milliseconds), provide new opportunities for the study of blur perception, both static and dynamic, and the development of new technologies in optometry and ophthalmology. In fact, recent developments like SimVis technology make use of optotunable lenses driven at high speed to simulate multifocal corrections.
In this thesis, we have used optotunable lenses to find the spatiotemporal limits of defocus perception. Besides, we have proposed and measured for the first time the spatiotemporal defocus sensitivity function, which provides a complete description of defocus perception. We have also developed a model to characterize defocus sensitivity, based on well-established models for contrast sensitivity. We found that the maximum spatiotemporal sensitivity to defocus is around 14 cycles per degree (cpd) and 10 Hertz (Hz), and the upper limits to sensitivity are around 50 cpd and 40 Hz. We found similar results with functional or paralyzed accommodation, suggesting that in this defocus flicker-detection task the presence of varying blur deactivates the accommodation response. These scientific discoveries provide a powerful framework for technologies that make use of temporal changes in defocus.
We have also developed a new subjective refraction method for obtaining the refractive error of an eye, called Direct Subjective Refraction (DSR). This method is based on the use of fast temporal changes in defocus in combination with the longitudinal chromatic aberration of the eye and a bichromatic stimulus made of two monochromatic (red and blue) circles to create flicker and chromatic distortion cues. These cues can guide the patient, without the supervision of the clinician, to obtain their spherical refractive error. We compared it with an unsupervised version of the traditional subjective refraction and also performed the same experiment with paralyzed accommodation, and we found out that accommodation barely influences the results of the DSR. We have demonstrated that the DSR method provides a highly repeatable spherical equivalent (±0.17 D) in less than 1 minute (39 seconds), with barely any supervision of the clinician, and minimizing the impact of accommodation.
Moreover, we have extended the DSR method for estimating the astigmatic component of the refraction using a similar stimulus but with oriented features instead of circles to capture the refractive error in different axes. Several experiments confirmed that the DSR method can capture the amount of astigmatism very precisely and with similar outcomes to the ones obtained with the traditional subjective refraction method.
Finally, we improved the psychophysical procedure of the DSR to estimate the full refractive error (spherical equivalent and astigmatism) and we evolved the on-bench optical setup to a clinical portable device (made of custom highly monochromatic LEDs) to allow measurements in clinical environments. We measured 33 real patients from two clinical sites, demonstrating the viability of the new clinical prototype. Comparing the DSR method with the traditional subjective refraction, we found similar results between methods (-0.34±0.53 D of difference in the spherical equivalent). Although the technology needs more improvements and adaptations to the clinical environment, the initial results are promising and already reveal the potential of the DSR technology.
In monovision, the dominant eye is compensated for far vision, and the non-dominant eye for near vision. The selection of the dominant eye is probably the main issue. In this thesis, we have developed the Eye Dominance Strength (EDS), a new metric based on the perceptual preference of the patient to randomized monovision in one eye or the other, while observing natural images. The EDS metric not only provides a binary result (left eye or right) as conventional tests do but also a quantification of the strength of eye dominance, with high repeatability. The EDS metric might help to prescribe monovision corrections more confidently and reduce the discomfort and the neuroadaptation times.
We have also studied the influence of monovision corrections in dynamic visual scenarios. We have discovered a new version of a 100-year-old optical illusion, called the Pulfrich effect, which produces misperceptions of the depth of moving objects due to interocular differences in the processing speed between eyes. This speed difference is interpreted by the brain as binocular disparity which arises a depth illusion. Classically, the Pulfrich effect was described with interocular differences in luminance: the dimmer image is processed slower than the brighter image. We discovered that interocular blur differences, like those produced in monovision corrections, produce the opposite effect: the blurrier image is processed faster than the sharper image. We report that blurring an eye with 1.50 D while keeping the other focused can cause delays in the processing speed as high as 3.7 milliseconds, with a remarkable impact on the depth illusion size. For that reason, we called the illusion the Reverse Pulfrich effect. The Reverse Pulfrich effect can affect public safety as the depth illusions, for example, while driving, can entail severe problems for monovision wearers.
In this thesis, we have also developed the anti-Pulfrich monovision corrections, which take advantage of the opposite sign of the Classic and Reverse Pulfrich effects, to null the effect caused by monovision by reducing the light of the blurring lens with a tint. We have demonstrated the efficacy of anti-Pulfrich monovision corrections delivered by contact lenses in young volunteers. Additionally, we demonstrated that interocular differences in retinal magnifications do not cause any difference in the processing speed and therefore do not produce the Pulfrich effect.
The classic version of the Pulfrich effect has been reported to occur in some pathologies, such as cataracts, or optic neuritis, producing a spontaneous Pufrich effect. In this thesis, we have reported for the first time a case of spontaneous Reverse Pulfrich effect in a patient adapted to surgical monovision after cataract surgery. The spontaneous Pulfrich effect measured was as high as 4.82 ms and caused severe binocular symptoms in the patient. After removal of the surgical monovision correction due to strong visual impairment, we measured a readaptation process with a timeframe of weeks.
We have also evaluated the effect of overall luminance level (from 0.4 to 12.8 cd/m2) for different pupil sizes (2, 4, and 6 mm) in the different versions of the Pulfrich effect. We found that reducing the overall light level increases the delay caused by interocular luminance differences (Classic Pulfrich effect, confirming results from the literature) and increases the delay with interocular blur differences (Reverse Pulfrich effect, first time reported). The similarity of the results in the Reverse Pulfrich effect for 4 and 6 mm pupil sizes suggests that high-order aberrations play a role in the delay caused by differential blur. The different increasing ratio of both the Classic and Pulfrich effects has implications for the development of potential anti-Pulfrich monovision corrections, which may need to modify their characteristics depending on the overall light level.
Finally, we have developed a portable setup based on an autostereoscopic technology using lenticular lenses to measure the prevalence of the Classic and Reverse Pulfrich effects on a young population. Although both effects are present in this sample (93% of them showed both effects), the effect sizes only elicit a considerable Pulfrich effect in half of the subjects measured. The framework developed opens the possibility for fast measurements of the Pulfrich effect in clinical environments. Furthermore, we also developed another portable setup for measuring stereoacuity based on a parallax barrier tablet. We have shown that stereoacuity can be precisely measured through SimVis Gekko, allowing fast and accurate predictions for different optical corrections.
In summary, this thesis has covered different aspects of vision related to blur perception, covering from their theoretical description to their direct clinical application. First, a new subjective refraction method for measuring the refractive error of an eye based on quick blur changes was developed and validated, providing fast and accurate measurements with high potential for clinical implementation. Second, a new metric for selecting the best eye for monovision corrections was designed and tested providing a measurement of the strength of eye dominance. Third, a new optical illusion caused by differential ocular blur, with important clinical implications, was discovered. Fourth, a new optical correction to compensate for the optical illusion previously discovered, the anti-Pulfrich monovision correction, was developed. Finally, two new portable devices based on autostereoscopic techniques were developed with the potential to measure different aspects of binocular vision, stereoacuity and the Pulfrich effect, in clinical environments. The outcomes of this thesis have advanced the understanding of blur perception and the application of that knowledge to the development of clinical instrumentation in optometry and ophthalmology.
Description
Mención Internacional en el título de doctor
Tesis por compendio de publicaciones.
Tesis por compendio de publicaciones.
Keywords
Direct Subjective Refraction, DSR, Eye Dominance Strength, EDS, Anti-Pulfrich corrections, Defocus perception, Optotunable lenses