Stereo imaging in ophthalmology is an important diagnostic modality. Origins of retinal stereo imaging date back to over 100 years ago. Stereo retinal photography and angiography facilitate the diagnostic process by clearly defining retinal structure and elevation.
Stereo imagery has been around for a long time. Long before the resurgence in modern 3D cinema attractions and 3D enabled televisions. The first attempt at reproducing a stereo image was undertaken by Sir Charles Wheatstone in 1838. Wheatstone designed the mirror stereoscope and presented his work on binocular vision to the Royal Society1.
Within months of Wheatstone's presentation in 1838, photography was born. By 1841, Wheatstone collaborated with Fox Talbot & Henry Collen to produce the first stereo photographs. Viewing stereo photographs became a popular parlour pastime in Victorian England after Brewster introduced his lenticular stereoscope at the Great Exhibition of 1851. From this date on, photographers from all over the World produced thousands upon thousands of stereo photographs, mainly of cityscapes and of urban life.
The first retinal photograph was taken in 1886 by Jackman and Webster, using a microscope and an exposure time of 2½ minutes to create the image. By the turn of the 20th Century, rapid progress in retinal photography had been made. Commercially available retinal cameras were being produced and so it was only a matter of time before retinal stereo photography was performed. The first published retinal stereo photographs were taken in 1909 by W Thorner2.
The process for all stereoscopic photography involves exposing two photographs from slightly different viewpoints of the same subject and then presenting these right and left images to the right and left eyes of the observer by means of a stereo viewer. The observer fuses the images together to recreate a sense of depth in one single image.
In simultaneous stereo photography, 2 lenses are used to capture the right and left images during the same exposure. This was popularised using 35mm film in the 1950’s. And today camera manufacturers have developed compact digital stereo cameras. In sequential stereo photography, one single lens is used to capture the left image and the camera is moved laterally to the right to take the right image. The distance between the 2 lenses used in simultaneous photography is typically 64mm. The distance required to move the camera laterally in sequential stereo photography is typically 64mm. This distance is known as the stereo base. This distance of 64mm is typically the distance between both pupils.
Fundus photography is a unique exception to the rule of using the 64mm distance for the stereo base because it incorporates the optical properties of the subject eye into the stereo and parallax equation. At a working distance of about 60mm from the front element of the retinal camera to the subject eye, a lateral movement of only 3-4mm is required to create a stereo effect. During the lateral shift of 3-4mm the image-forming rays fall on opposite slopes of the cornea, increasing parallax and creating a hyper-stereoscopic effect. Cornea-induced parallax, as described by Lee Allen in 19643, negates the use of the 64mm stereo base distance.
Most retinal cameras are capable of producing sequential stereo images. The camera begins in the same position used for single fundus photography. It is shifted, first to the left and then the 3-4mm to the right.
Most software applications that run ophthalmic digital imaging systems will have the ability to tag and format photographs taken in stereo. We routinely take stereo images on all our patients, particularly during fluorescein and ICG angiography. Once the images are saved the software will allow you to view the stereo pair. The right and left images are presented on the viewing monitor. A pair of viewing glasses should be used to merge the two images together (see image above). Some software will allow you to convert the side by side pair into a single anaglyph image. The anaglyph is a variation in the presentation and viewing of stereo images that uses two monochromatic images in complementary colours. The unaided eye perceives two overlapping images. With an appropriate set of viewing filters, the right and left images are separated. The anaglyph method has the advantage in that the red/cyan glasses are relatively cheaper to use (mass produced cardboard viewers are common). Also, when presenting stereo images in a presentation to a large audience, the use of anaglyph images facilitates this imaging modality.
References:
- Wheastone C. Contributions to the Physiology of Vision – Part the First. Philosophical Transactions of the Royal Society of London, Vol. 128, (1838), pp. 371-394
- Thorner W. Die stereoskopische photographie des augenhintesgrundes. KIM Monatsbl Augenheilkd 1909;47:481.
- Allen L. Ocular fundus photography. Am J Ophthalmol 1964;57:13.