Visual acuity (clearness of vision) usually is measured by an eye doctor using an eye chart. It measures the smallest letters that can be reliably identified by the patient at a specified distance. The traditional test requires the patient to look and report which letters they see.
This invention provides an automated system to estimate visual acuity based on objective measurements of the eye optics and wavefront aberrations (WAs). WAs are represented in a particular form known as the Zernike polynomials. A typical measurement on the eye of a patient will consist of a list of about 16 numbers that constitute the coefficients of the polynomials. The WFAMetric is an algorithm that converts the list of numbers into an estimate of the visual acuity of the patient. If changes are planned to the WA of the patient through surgery or optical aids, the predicted change in acuity can be calculated.
The wavefront aberrations are a collection of different sorts of optical defects, including the familiar defocus and astigmatism that are corrected by eyeglasses, but also more complex “higher-order” aberrations such as coma, spherical aberration, and others. The WA provides a comprehensive description of the optics of the eye, and thus determines the acuity. Until recently, a practical method of computing this relationship did not exist.
This innovation simulates the observer performing the acuity task with an eye possessing a particular set of WA. When a letter from an eye chart test is presented, a digital image of the letter is at first distorted by the specified WA, and noise is added to mimic the noisiness of the visual system. From previous research, the appropriate noise level to match human performance has been determined. Next, an attempt is made to match the blurred noisy image to similarly blurred candidate letter images, and to select the closest match. This is repeated for many trials at many letter sizes, thereby determining the smallest letter that can be reliably identified — the visual acuity. This simulation may seem complex and cumbersome, but it has been streamlined and simplified at all the key steps. The entire process is now robust, accurate, simple, and fast, with results being returned in typically a few seconds.
This work was done by Andrew B. Watson and Albert J. Ahumada Jr. of Ames Research Center.
NASA invites companies to inquire about licensing possibilities for this technology for commercial applications. Contact the Ames Technology Partnerships Office at 1-855-NASA-BIZ (1-855-6272-249). Refer to ARC-16331-1.