Video-Based Low-Vision Devices
JENNIE Y. KAGEYAMA, MELISSA W. CHUN, KENT W. SMALL and JOSEPH L. DEMER
Table Of Contents
OPTIMAL IMAGE CHARACTERISTICS FOR READING
COMPUTER VIDEO DISPLAYS AND DIGITAL DEVICES
COSTS AND BENEFITS OF VIDEO VISUAL DEVICES
FUTURE HORIZONS FOR VIDEO-BASED LOW-VISION DEVICES
|Low vision devices, whether optical or video based, facilitate visual function
and help allay limitations due to varying degrees of visual impairment.1,2 Visual impairment results in three forms of loss: loss of central acuity, which
represents resolution; loss of sensitivity to contrast; and
loss of visual field. Visual impairment defined strictly on the basis
of central acuity can be categorized as moderate (20/70 to 20/160), severe (20/200 to 20/400), or profound (worse than 20/400). The
most commonly used visual aids address the problem
of reduced visual acuity and employ magnification to increase image size on the retina to circumvent the patient's
limited visual resolution. However, optical magnifiers, such as loupes, telescopes, and
convex lenses, achieve image enlargement at the expense
of other image characteristics that may be crucial for patients with
low vision, namely, image brightness, contrast, and useful visual field. All
these factors contribute to the impediment for quality viewing.|
In distinction to optical magnifiers, video-based image processing systems need not sacrifice other image properties to achieve magnification. Although beneficial for patients with different degrees of visual impairment, video-based devices are particularly useful for patients with severe to profound visual impairment when conventional low vision devices are determined to be inadequate. Fonda and colleagues3 found that the closed-circuit television (CCTV) was most appropriate when optical devices did not produce sufficient magnification. Video-based systems introduced major advantages for people with low vision. These advantages include higher levels of magnification without the light-losing characteristics of purely optical systems, the ability to manipulate the video output electronically to create an enhanced image with greater contrast, and the availability of a larger field of view than is possible with solely conventional solutions.4 In fact, the CCTV can electronically magnify up to 60× without significant optical aberration. In addition, binocularity is maintained and the posture to read the text on the screen is more natural than the often uncomfortable and specific positioning required by other high-power near-vision aids.5 The primary disadvantage of the CCTV is that the system is not portable. However, with the availability of new technology and the use of head-mounted systems, this limitation can also be overcome.
First suggested in 1959 by Potts and coworkers,6 the use of CCTV systems as a low vision device was technologically feasible and popularized in the early 1970s4 by groups such as Genensky at the Rand Corporation7 and Fonda and colleagues.3 Although CCTVs have always been quite useful for some low vision applications, recent advances in our understanding of visual physiology, as well as in image processing technology, have greatly expanded the potential value of video-based low vision devices. In this chapter, both CCTV technology and the emerging technology of video enhancement are presented, along with an introduction to the various video-based low vision devices currently available for the patient with low vision. Optimal reading characteristics also are discussed.
|A CCTV system consists of a camera, an X-Y positioning platform
under the camera for placement of the text to be scanned, a light source
to illuminate the material, and a video display monitor.3,8 The CCTV user moves the camera or printed text systematically to scan
from left to right and from top to bottom. The X-Y table was a
significant innovation in the design of CCTV systems because it greatly
enhanced the comfort of reading and writing. Magnification is achieved
through use of a zoom lens on the camera, the large size of the video
monitor screen, and reduction of the viewing distance of the patient. An
effective magnification of even 140× is possible,3 although the average magnification most commonly employed is far less.9 Basically, two types of magnification are in operation with the use of
the CCTV: projection magnification and relative distance magnification.5|
Several approaches can be taken to calculate the magnification achieved by a CCTV.10 Total magnification (M) of the CCTV can be calculated from the product of projection magnification (X) and relative distance magnification (Y), M = (X)(Y). X is defined as the print size on the monitor divided by the actual print size of the text (maintaining the same units of measure, e.g., centimeters); Y is the reference distance (25 cm or 40 cm) divided by the working distance in centimeters. Of course, if the reference distance is not designated, it is virtually impossible to determine the total magnification of the system. A cleaner approach in assessing the exact magnification of a CCTV is to determine the equivalent power provided by such a system. This allows the practitioner to obtain the dioptric equivalent of the projection magnification–relative distance magnification interaction. Once the dioptric equivalent is obtained, it is then easy to compare across systems and select systems of equivalent powers for patient evaluation, such as handheld and stand magnifiers, microscopes, or telemicroscopes. Equivalent dioptric power, Deq, equals (X)(Z), where X is the print size on the monitor divided by the actual print size and Z is equal to the dioptric equivalent of the working distance, that is, the reciprocal of the working distance or viewing distance of the patient from the monitor (100/working distance in centimeters). Finally, one can obtain a quick estimate of the amount of projection magnification, disregarding working distance. A piece of graph paper is placed under the camera of the CCTV after the appropriate print size has been established on the screen for a patient's reading needs. Another piece of graph paper is then held up to the monitor, and the regular-size graph squares that fit into the length of one or two magnified graph squares on the screen are counted. By dividing the resultant smaller number into the larger number, the projection magnification can be grossly determined.10
Stand-mounted CCTV systems are commonly configured with television receivers, video monitors, or computer monitors. They provide high-contrast, gray-scale, and inverse video display modes with adjustments for levels of contrast and brightness. The characteristics of enhancement may need to vary with the spatial frequency content of the television picture and the progression of the user's ocular condition.5
Contrast polarities are shown in Figure 1, whereas different contrast intensities are shown in Figure 2. Video displays can achieve contrast of up to 96%.11 Because video image contrast is electronically controllable at levels below the maximum achievable level in a given device, all video displays can be easily adjusted to produce any desired level of contrast intensity less than the maximum. This means that display contrast can usually be set to a level greater than that of the original subject matter before the camera. More importantly, contrast polarity can be easily reversed by an inverting amplifier, usually incorporated in video cameras and in better-quality monitors. These features of contrast enhancement and polarity reversal are unique to video-based visual devices.
Handheld systems typically consist of a camera and a control box with a cable to be connected to a TV or video monitor. The user holds the camera and moves it across a page, which often requires sufficient hand-eye coordination. Contrast intensity and polarity can also be controlled on these systems.
Studies have confirmed the overall effectiveness of CCTVs in improving the reading ability of individuals with all levels of visual impairment.12 Sloan,13 in her case report on CCTVs, identified the following advantages of such a system:
However, a common theme stressed in the literature and by practitioners is that the effective use of CCTVs is seldom achieved through mere trial and error; rather, both a comprehensive low vision examination and training in use of the system are critical to success.
The digital revolution has begun to change the design of CCTVs and will continue to do so at an accelerated pace. Images can be enhanced and modified in a multitude of ways, and systems can be endowed with more memory and storage capabilities. These devices may become more popular and useful for orientation and mobility purposes as well.4
|OPTIMAL IMAGE CHARACTERISTICS FOR READING|
|In our complex society, the ability to read is crucial to everyday activity
and to nearly all forms of employment. For this reason, reading performance
is usually taken as a benchmark for visual function.|
An inherent limitation of video technology is that the amount of information displayed on the screen is defined by the camera's field of view. Obviously, when characters are made larger, fewer can be simultaneously displayed on a screen. For both normally sighted readers and those with low vision, Legge and associates14,15 found that reading rates are generally maximized when four or more characters are simultaneously displayed. Display of a window of fewer than four characters slows reading, whereas presentation of more than four characters limits the available magnification for a given video screen size. There are exceptions to this rule, however, and certain patients may continue to exhibit increases in reading rate when a window of up to 20 characters is employed.16
The reading rate of normal subjects declines markedly for contrast below 10%, as letters appear to fade into the background (see Figure 2). Most readers with low vision exhibit a sensitivity to image contrast that resembles that of normal subjects reading faded text. In the majority of patients with low vision, the minimum image contrast required for best reading speed averages 3.9 times that required by normally sighted subjects.11 Contrast is such an important parameter in CCTV reading performance that most patients with low vision prefer to use all of the contrast available from a CCTV display. In fact, contrast sensitivity of individual patients has been found to be a better predictor of CCTV reading performance than Snellen acuity tested at distance.17 The ability to detect threshold variations in contrast, in both the normally sighted person and the patient with low vision, is generally a function of the spatial frequency content of the image. Stripes, called gratings, having sinusoidally varying brightness are usually used to test contrast sensitivity at defined frequencies. Spatial frequency is defined as the number of cycles of alternation of the bright and dim stripes per degree of visual angle. Normal subjects exhibit maximal sensitivity to contrast for spatial frequencies of 2 to 4 cycles per degree,18 although spatial frequencies of as much as 25 to 30 cycles per degree can be detected when contrast is very high. The upper limit of spatial frequency detection corresponds roughly to threshold visual acuity for high-contrast Snellen letters. The reduced acuity of patients with low vision is reflected in a lower limit of maximum spatial frequency detection, but in these patients, the frequency of peak contrast sensitivity is usually reduced as well.17 As noted previously, the absolute contrast sensitivity of patients with low vision is also reduced. This is seen in Figure 3, which plots contrast sensitivity (the reciprocal of contrast threshold) as it varies with spatial frequency. The shapes of contrast sensitivity functions of different patients may vary even though their Snellen acuities are the same. It is tempting to think that there may be a way to manipulate the characteristics of the image to compensate for deficiencies in the contrast sensitivity functions of individual patients in order to improve their reading performance. One simple way to do this is through magnification, which has the effect of reducing the spatial frequency content of the image by the amount of the magnification factor. Another method of image enhancement proposed by Lawton18 employs computerized spatial frequency compensation filters to enhance the midrange spatial frequencies of images to compensate for specific deficits in the contrast sensitivity functions of individual patients with low vision. Although this method reportedly improves reading speed and reduces the amount of magnification required for reading, these claims have not been subjected to clinical confirmation.
Contrast polarity has no significant effect on reading performance of normally sighted persons.19 Black characters on a white background are as easily read by normal subjects as white characters on a black background. However, the overall brightness of these two types of presentations is markedly different. The page with the black background has a lower average brightness and thus less ability to produce glare than the white page. Rubin and Legge11 found that contrast polarity has a significant effect on reading rate in a minority of patients with low vision, namely, in patients with “cloudy” ocular media. Patients with cloudy media typically read white characters on a black background faster than black text on a white background. The authors attributed this difference to the scattering of background light by the ocular media to produce “veiling glare” that reduces effective image contrast at the retina. This hypothesis is further supported by experiments in which black characters are presented against a white background within a window surrounding the characters.20 Reading rate was reduced for larger windows of white background in a manner consistent with veiling glare. Patients with macular degeneration had no significant preference for contrast polarity, whereas patients with reduced central vision due to retinitis pigmentosa often preferred white characters on a black background.21 Because many of the patients with retinitis pigmentosa also had posterior subcapsular cataract, this preference may also be due to glare effects from cloudy media, although cystoid macular edema may also have played a role.
Among the many ways that CCTV systems enhance visual images is their ability to provide images in different colors. Under photopic conditions, the color of displayed characters has no significant effect on the reading rates of normal subjects, except for characters near the threshold of resolution.22 Even for patients with low vision, image color only occasionally has a significant effect on reading. When there is an effect, reading performance is most likely to be impaired for red and, to a lesser extent, blue characters.22 Reading performance is best with green or white characters. There may, however, be an overall subjective color or contrast polarity preference depending on the individual, regardless of ocular condition.
Studies have shown that varying screen colors on CCTV systems does not affect reading performance when screen luminance and contrast settings are equivalent. To investigate whether screen color is an important variable in the prescription of CCTV systems, Jacobs23 measured the visual performance of CCTV users on white, green, and amber screens. There was no obvious link between the low vision condition and preference or aversion to those colors. Luminance transmittance appeared to be a greater factor in reading ability and maximizing luminance contrast optimized reading performance. On the basis of these findings, it may be concluded that a low-vision clinic does not need more than one CCTV screen color for its visual assessments. When the screen color can be chosen, a white screen may be more versatile.
When CCTV low vision devices are employed, patients control the rate and pattern of scanning of the text according to their individual needs. Normal subjects generally read static, or nonmoving, text more rapidly than text that is constantly moving across the screen. For patients with low vision, however, reading rates for drifting text tend to be modestly higher than for static text. In a study performed by Goodrich and colleagues24 it was determined that reading durations with the CCTV are significantly greater than with optical aids. A possible advantage of the CCTV is that it is sufficiently flexible to compensate for substantial changes in the patient's visual efficiency.25
The optimal characteristics of a CCTV low vision aid can be summarized as follows: Magnification should be variable but should at least include an effective value of 30×. At the chosen level of magnification, screen size should be large enough to display at least four characters, preferably as many as ten characters. This implies that larger screen sizes will nearly always be advantageous over smaller ones, enabling higher magnification. Contrast intensity should be as high as technically possible, and ambient illumination or filters should be adjusted to reduce glare from the video screen so as to maintain display contrast. Contrast polarity using a black background assists in eliminating glare. White characters are probably optimal for readability and contrast, but green is also acceptable. The X-Y positioning capabilities of the CCTV system should permit the low vision reader to smoothly scan the text to be read. Finally, the mechanical arrangement of the monitor and camera should enable the patient to maintain a comfortable posture to facilitate prolonged reading.
|COMPUTER VIDEO DISPLAYS AND DIGITAL DEVICES|
|The increasing pervasiveness of computers in society has been paralleled
by a trend toward use of adaptive computers and computer-based
video displays by patients with low vision.26 Computer-based video displays share all of the advantages of the
CCTV devices discussed previously but in addition, present the opportunity
to manipulate additional features.|
Another opportunity for flexibility in character presentation with computer video displays is in the spacing between characters. This may be of value to patients with central scotomas, who are forced to read using peripheral retinal loci where crowding effects may be significant. Although character spacing has no effect on acuity, the time required for recognition of characters increases when multiple, closely spaced characters are presented to peripheral retina.27 The standard spacing between characters is 20% of character width for most printed text. In the peripheral visual field, this spacing has been found to increase the time required for recognition by 22% to 65% of the minimum recognition time for uncrowded characters. Since the optimum letter spacing may vary from patient to patient, adjustable character spacing would be a desirable feature of computer video displays. This feature is currently available with existing software packages for personal computers. Such software includes various magnification and contrast enhancing capabilities.
Most personal computer hardware and software include built-in options that permit a moderate (usually twofold) enlargement of characters on the screen. Other computers, such as the Macintosh, can display font sizes up to 70 point or more directly on the screen. (Conventional character displays are 10 or 12 point.) However, the larger characters may not be compatible with certain widely used software, such as spreadsheets, which become impractical when they are so extensive that they fill more than several times the area of the display screen. Modern laser and ink-jet printers generally have the desirable capability of producing very high contrast print with characters in a wide range of sizes useful for patients with low vision, although again this may not be compatible with certain applications, such as spreadsheets and graphics. Special software is available to enlarge computer video displays. There not only are programs that magnify text and icons, but also products that use voice-output programs that read the text to the user. Because new hardware and software products are constantly being released, it is best to contact local retail distributors and computer users groups for additional information.
A very effective, albeit costly, method of improving the readability of a computer video display is to increase the size, or diagonal measurement, of the monitor. Screen magnifiers that are optical magnifiers, can be placed directly over video display screens. These are inexpensive and versatile but provide magnification limited to only about 2×.28 Nonmagnifying filter screens can be placed over video displays to reduce glare and improve effective contrast. Of course, spectacle telemicroscopes, allowing close focus, or loupes may be used in conjunction with video displays for additional magnification benefits. Handheld magnifiers may also be used as adjuncts.
Previously, most video-based systems were mainly employed for reading. However, with advancements in technology as well as increased patient necessities and demands, several digital devices for viewing far and intermediate distances have been developed, along with more sophisticated reading devices to facilitate reading tasks.
A digital handheld monocular telescope has recently been introduced that combines high magnification at long, mid, and near ranges with contrast enhancement and CCTV configuration capabilities (Fig. 4). Shaped like a handheld video camera with ergonomic design, the device has the ability to perform in low ambient light, a benefit over conventional monocular telescopes, which tend to require more light, depending on the ocular condition (Betacom, Glendale, AZ).
A new version of the conventional stand magnifier has also been developed (Fig. 5). It consists of a 3× (7.6 D) illuminated light-emitting diode stand magnifier, which when attached to any television set, displays digital images on the screen with approximately 9× to 17× magnification. The construction of this device allows the option of using it as an optical stand magnifier or digital device (Eschenbach Optik of America, Inc., Ridgefield, CT).
Revolutionary camera designs enable people with low vision to perform daily tasks more effectively. To illustrate, there is now a compact camera that turns a full 180 degrees to magnify images in any position up to 30× with autofocus capability of distant, intermediate, and near ranges. The system can be made portable with a pair of lightweight glasses that display the magnified image within the battery-operated system. Another device, a portable handheld digital magnifier, enables the use of a similar pair of lightweight glasses with built-in monitor for reading text (Fig. 6) (Enhanced Vision Systems, Huntington Beach, CA).
For distance viewing, a head-mounted camera system is available with an autofocus magnification range of up to 25×. The system has been updated to provide a more lightweight and compact design with a portable battery pack. The zoom feature and autofocus capability assists in discerning details with greater ease (Fig. 7) (Enhanced Vision Systems, Huntington Beach, CA). The head-mounted design enables hands-free use. The autofocus feature allows a wide range of viewing distances. Contrast enhancement and reversal permit individuals to select optimal images for different tasks. Automatic video gain control and maintenance of constant screen luminance are functions of particular benefit to patients with adaptation/glare difficulties. Finally, the system allows for binocular viewing and the possibility for a wider field of view. Disadvantages include weight, appearance, cost, complexity of operation, and potential problems of motion sickness and claustrophobia. When viewing through a magnifying system, retinal image motion is magnified, and the visual vestibulo-ocular reflex gain must be equal to the magnification to stabilize the retinal image. The systems are unsuitable for patients with serious hand/head tremors or patients who are physically unable to operate the controls or support the headset.29
A mini, desktop-style, portable video magnifier is available for reading and writing and weighs less than 3 pounds. Much like the conventional CCTV, this lightweight system has a magnification range of 4× to 50× and connects to most television sets and computer monitors (Freedom Vision, Mountain View, CA).
|COSTS AND BENEFITS OF VIDEO VISUAL DEVICES|
|Although video-based visual devices are the most versatile and effective
of the highly magnifying low vision devices, they are also the
most costly. Particularly when coupled with personal computers, video-based
devices may be uniquely effective in restoring employment
or educational potential for the patient. The greater costs of these
aids must be weighed against their benefits.8|
The rate of success in the use of video-based visual devices was high even for the devices available in the mid 1970s9; there is some evidence that improved technology has provided additional benefit. Even with technically efficient devices, initial training in their proper use is essential, and additional training can provide further improvement in reading speed and duration.26,30 Although reading comprehension is probably not a good measure of the effectiveness of a visual device, Legge and coworkers31 demonstrated that the comprehension of low vision subjects reading with a video display is equivalent to that of normally sighted subjects when reading speed is the same relative to the subject's maximum rate.
|FUTURE HORIZONS FOR VIDEO-BASED LOW-VISION DEVICES|
|The new generation of adaptive devices has incorporated the advancements
of digital technology. Whereas older video-based low vision devices
were heavy, bulky, and mechanically awkward, recent developments
in solid state video technology are already providing substantial practical
benefits to the patient with low vision. Besides the obvious advantages
of reduced weight and bulk, these portable displays feature
low power consumption. New portable video-based devices are constantly
being researched, including a digital device so compact that it
may be worn around the neck and carried in the palm of one's hand. Given
the rapid progress in video display technology, further refinements
in this area undoubtedly will occur and will probably be associated
with decreasing prices. With technically innovative advancements, the
ability to improve low vision devices has dramatically increased
and the potential to improve a low vision patient's ability to function
more efficiently in daily activities has increased as well.|
Interfaces between CCTV displays and personal computers are also rapidly evolving. These may be of particular benefit to patients who use visual devices for reading printed text during concurrent work using a personal computer. A device is already available that splits the video screen to allow part of it to display the output of the CCTV camera and the remainder to display enlarged computer characters (Telesensory Corporation, Sunnyvale, CA; International Business Machines, Armonk, NY).
Enormous potential exists for the development of software for patients with low vision to enhance readability of computer video display screens and perform complex digital image processing using personal computers. It seems likely that the continued decline in the cost of computer technology as well as advancements in design, resulting in lighter, smaller products and display images of improved quality, will enable practitioners to more commonly prescribe these new low vision devices and will provide further options for prescribing. Video magnification technology has the potential to offer considerable improvements to a patient's visual functioning and his or her ability to perform activities of daily living, as compared with conventional low vision aids. Moreover, this technology may offer solutions to some of the issues raised by users of conventional devices. Clearly, the benefits of video-based and digital devices are significant, optimizing functionality and, undoubtedly, enhancing quality of life for low vision patients.
The authors have no financial interest in any of the low vision products mentioned in this chapter.