This paper discusses some of the physics that underlie IOL MTF tests and makes suggestions for performing meaningful MTF test of both monofocal and multifocal IOLs. The purpose of this paper is not to define a test protocol,but to initiate a focused discussion between those parties who have a common interest in adopting a suitable protocol.

In this study,which also provides an assessment of machine performance, IOL optimization of aspheric surfaces, using the conic constant factor, has been selected. Such surfaces are easily machined by most software-driven lathes, such as DAC 2- and 4-axis lathes, which were utilized for this study. Measuring tilt and decentration performances on the optimized aspheric lenses, it is shown that 'best shape' IOLs scan be produced using a broad range of lens designs. This promises to increase manufacturer and surgeon confidence in maximizing patients' visual acuity with better contrast sensitivity using aspheric designs.

In this paper we will discuss the state of the art in regard to measuring the modulation transfer function of intraocular lenses. We will also discuss many of the subtle yet important issues of testing in aqueous versus air, the need for an 'artificial cornea,' and other critical test-related issues.

Until recently, the manufacture of toric ophthalmic lenses relied largely upon expensive, manual techniques for generation and polishing. Recent gains in computer numeric control (CNC) technology and tooling enable lens designers to employ single- point diamond, fly-cutting methods in the production of torics. Fly-cutting methods continue to improve, significantly expanding lens design possibilities while lowering production costs. Advantages of CNC fly cutting include precise control of surface geometry, rapid production with high throughput, and high-quality lens surface finishes requiring minimal polishing. As accessibility and affordability increase within the ophthalmic market, torics promise to dramatically expand lens design choices available to consumers.

An understanding is required of the range of image quality reduction that can be tolerated by human subjects. The modulation transfer function (MTF) of the phakic eye provides a reasonable reference for nominal optical performance. We propose MTFs of a 1/4D and 1/2D defocused eye as minimum acceptability criteria. Weber's law can be used to assess whether differences from nominal values are likely to be detected. These criteria apply to assessment of distant object imaging. To predict levels of clinical measures based on in-vitro MTFs, a simple model of human threshold recognition/detection is discussed. This Expected Visual Outcome model provides a tool for estimating the clinical efficacy of a particular optical surface design.

Machining raw material by a lathing operation is the most common method for making the optical part of a polymethyl-methacrylate (PMMA) IOL. Spherical surfaces have been used for many years to achieve this goal using the radioscope for in-process control. For a few years, advanced control techniques such as Talysurf or Fizeau interferometer have been used to achieve improved surface quality and ideal lathe setup. Machining an optimized conic surface would appear to be more difficult to control. CAD drawings of a conic surface of an optimized IOL versus a spherical surface show a difference in vault of a few microns, which can be seen easily on interferometer display fringes. It is then easier to control the asphericity of a lathed surface by achieving tolerance in fringes when the dome is within 1 to 4 fringes of a perfectly spherical surface. This method has been validated with several thousand lenses using the USAF target method as a routine control and has been shown to produce increased accuracy of lens power and resolution efficiency.

Current trends in and intraocular lens design suggest ever- increasing demand for aspheric lens geometries - multisurface and/or toric surfaces - in a variety of new materials. As computer numeric controls (CNC) lathes and mills continue to evolve with he ophthalmic market, engineering hurdles present themselves to designers: Can hardware based upon single-point diamond turning accommodate the demands of software-driven designs? What are the limits of CNC resolution and repeatability in high-throughput production? What are the controlling factors in lathed, polish-free surface production? Emerging technologies in the lathed biomedical optics field are discussed along with their limitations, including refined diamond tooling, vibrational control, automation, and advanced motion control systems.

Optical characteristics of the multifocal intraocular lens (IOL) are different from regular monofocal IOLs. We have developed a test method and procedure to measure completed optical performance of the AMO Array^TM multifocal IOL and to control its optical quality. The measured parameters include effective focal length, aqueous diopter power, add power, depth-of-focus, resolution efficiency, contrast modulation for far and near image, subjective quality factor variance , and through focus response. Automated Inspection Station was developed to perform the measurements. In addition, surface characteristics of the multifocal IOL are measured by a special custom-built interferometer (Wavy Optical Tester).

Simultaneous vision multifocal ophthalmic lenses in contact lenses or intraocular lenses present a challenge in design and testing. The SNR represents one criterion that can aid the design in evaluating such systems. The definition of SNR as applied to such systems and an example of its use are presented here.

The special equipment requirements and limitations for bifocal intraocular lens MTF measurement using a CCD detector are discussed. Proper MTF measurement requires that at least 90% of the energy passing through the lens aperture be imaged within the sampled area of the sensor. Spectral filtering of optical density greater than 2.5 sufficiently eliminates out-of-band radiance. Well-corrected relay optics of numerical aperture at least twice that of the lens under test and a calibrated image scale are important for accurate results. Significant modulation losses at spatial frequencies well below the Nyquist sampling occur due to pixel geometry, blooming, and charge transfer inefficiency. for a bifocal intraocular lens, where the overall blur pattern may be as large as 0.5 mm, these factors can limit the ability to accurately compute modulation at spatial frequencies greater than 100 cycles/mm. Correction of equipment-induced modulation losses can be realized through improved data processing. Use of MTF measurement for quality control of the Storz TRUE VISTA ^TM bifocal intraocular lens is discussed. Results agree with theoretical expectations to within 10% modulation. Sensitivity to defects such as astigmatism and zone defects are demonstrated at low spatial frequencies.

A model eye wet-cell was used to evaluate the effects of tilt and decentration on the optical performance of three different designs of multifocal intraocular lenses. Tilt and decentration have the greatest effect on diffractive lenses shrinking the depth of focus, decreasing resolution, and increasing power. Tilt or decentration of aspheric refractive lenses significantly decreases image quality. Tilt and decentration have little effect on spherical refractive lenses.

In this paper some practical observations related to the fabrication of multifocal IOLs are presented from the viewpoint of a diffractive optics design and fabrication group whose experience lies mostly outside the area of ophthalmic optics.

The traditional parameters used for describing lens performance, such as PSF of MTF, fail to take into account the contribution of the human visual system to image quality. The proposed criterion assumes that the eye responds through accommodation and achieves a 'best focus' situation. The best focus is defined as that focal length where the maximal amount of radiation falls within a certain depth-of-focus interval. The depth range is determined by the eye's ability to detect power changes, which is around 0.25 diopters. The distribution of focal power over the lens aperture is measured with a moire deflectometer. Convolving the distribution with a smearing function of width equal to the dioptic depth parameter produces a power profile as perceived by the visual system. The maximum of this seared distribution defines the measured power of the lens, and the value of the distribution at the maximum, normalized by the total area under the curve, is the quality figure of the lens. The new parameter, giving the encircled energy with the focal depth range, may be refined by taking into account other features of lens fabrication and human response. In particular, the distribution is weighted according to the radial distance from the lens center. With high- power lenses with their inherent spherical aberration, less weight is given to more distant portions of the lens. The measurement of lens quality is demonstrated with a large number of soft contact lenses measured in solution. These lenses exhibit a wide range of distortions and imperfections.

A 2.5 X magnification system consisting of a two-zone intraocular implant and a spectacle was developed, tested, and clinically tried by fifty patients with cataract ad age-related macular degeneration. Optical bench testing results and clinical data confirmed that the field of view of the system was 2.6 times wider than an equivalent external telescope. The study also demonstrated that the implant itself was clinically equivalent to a standard monofocal intraocular lens for cataract. The clinical study indicated that higher magnification without compromising the compactness and optical quality was needed as the disease progressed. Also, a sound vision rehabilitation process is important to provide patients the full benefits of the system.

To assess the images formed by a lens, we have to know first its focal length and its transfer function, characteristic of the quality of the theoretical focused image. At this time, aberrant lenses cannot be perfectly characterized. Several optical benches have been built in order to measure focal length and MTF of these lenses, but repeatability was very poor. Corneal and Supelec then conceived of a new method based mainly on calculations instead of mechanical device adjustments. A preliminary study proved the feasibility of a new method, which we call the 'speckle' method. Several tests have shown that power and imaging quality were included into the two-dimensional signal measured by a CCD. Mathematical theory was then conceived to assess power and imaging quality with good repeatability. We will present here the theoretical basis of the speckle method, then present the practical bench. Results will be given after a second applied research program, which will be completed by the end of 1994.

Contrast sensitivity has emerged as an accepted, comprehensive tool which measures visual function. More capable than Snellen acuity, contrast sensitivity provides early detection of eye disease and the measurement of visual capability and performance-'2 Sine—wave gratings have been recognized as the gold standard for contrast sensitivity for over 29 years in the scientific, clinical and applied communities. Low contrast letter acuity has emerged the last 10 years as an alternative to gratings for contrast testing. The advantages and disadvantages of each approach to contrast testing have been discussed before.3 Here is offered a set of criteria and a methodology for evaluating any contrast test, whether composed of gratings, letters or any other target.