A metric-based image-formation model explains the improvement in subjective refraction using temporal defocus waves.

Area of Science:

• Ophthalmology
• Computational Optics
• Visual Science

Background:

• Direct subjective refraction (DSR) is an emerging technique for refractive error assessment.
• DSR utilizes temporal defocus changes and a flicker minimization task for accuracy.
• Existing clinical methods for refractive error measurement can be improved.

Purpose of the Study:

• To develop computational models for the DSR method of refractive error measurement.
• To simulate DSR using rapid optical power changes and bichromatic stimuli.
• To provide a framework for enhancing the clinical application of DSR.

Main Methods:

• Simulated retinal images using the eye's polychromatic point spread function.
• Defined an image quality (IQ) metric based on spatial frequencies.
• Modeled blur minimization (BM), monochromatic flicker minimization (MFM), and DSR tasks.
• Quantified image similarity for flicker-based tasks.
• Analyzed through-focus peak width across a ±3-D range, varying pupil size and spherical aberration.

Main Results:

• Through-focus 90% peak widths were 0.48 D (BM), 0.16 D (MFM), and 0.19 D (DSR).
• Results align with previous experimental findings.
• Peak width increased with smaller pupils and spherical aberration in BM and MFM.
• DSR model showed a relatively constant peak width across conditions.

Conclusions:

• The developed computational models successfully explain prior experimental observations.
• Models support the superior repeatability of DSR compared to traditional refraction methods.
• These models offer a pathway to refine and improve DSR clinical practice.