Summary: The Knighton laboratory is interested in bringing basic physical principles to bear on the problems of clinical imaging of the tissues of the eye.

Robert W. Knighton, Ph.D.
Research Professor of Ophthalmology
Secondary Appointment in Biomedical Engineering

Search for published research articles by this doctor in the National Library of Science.

Current Research Summary: Since the invention of the ophthalmoscope by Helmholtz in 1850, clinical diagnosis of ocular disease has included observation of the fundus of the eye. Fundus drawings and fundus photographs have provided permanent records of these observations but, with the exception of spatial relations, usually have not provided quantitative information. Recent advances in electro-optics have led to a number of new methods for fundus imaging that use electronic detection of light. Electronic detection provides an inherently quantitative measure of the light that forms the image, opening another dimension for obtaining information. That this new dimension will be useful is apparent from current clinical practice; ophthalmoscopists describe the appearance of abnormal fundus features using a variety of terms (e.g., "waxy pallor", "cherry red", "beaten metal", "cellophane") that clearly refer to optical properties (e.g., spectral and directional reflectance) of the feature observed. Quantitative imaging promises rapid, objective, accurate and reproducible measurement of these properties and allows data to be compared between individuals (for discriminating disease from normal) and over time (for detecting progression or resolution of disease). Furthermore, quantitative measurements can be described by mathematical models that explicitly incorporate anatomical and physical properties of the fundus tissue. These models form a valuable link between the image itself and the underlying tissue properties, a link that greatly enhances image interpretation.

The Quantitative Ophthalmic Imaging Group at Bascom Palmer Eye Institute pursues both basic studies aimed at providing a comprehensive quantitative understanding of the optical properties of fundus tissue and applied studies aimed at understanding and improving specific clinical technologies. Our overall approach to the twin goals of basic understanding and clinical improvement has been to form a multidisciplinary team of basic scientists and clinicians who have interacted successfully to bring a clinically relevant focus to the laboratory work and technical innovation to the clinical studies.

Over the past several years we have focused on determining the optical properties of the retinal nerve fiber layer (RNFL) with the aim of improving glaucoma diagnosis and management. Recognizable damage to the RNFL occurs early in glaucoma, often preceding by years the detectable loss of visual sensitivity as measured with visual fields. Our approach to understanding the optical properties of the RNFL has been to employ quantitative measurements and mathematical modeling that iteratively refine our description of the properties and restrict the possible candidates for the anatomic mechanisms. This approach has frequently required the development of innovative apparatus, measurements and analyses.

We have shown that light reflected by the RNFL arises from light scattering by cylinders, that at least two different cylindrical mechanisms are involved, with the reflectance at short wavelengths dominated by thin cylinders and at long wavelengths by thicker cylinders, and that microtubules within ganglion cell axons play a role. Our current model treats the RNFL as a thick birefringent slab containing a parallel array of scattering cylinders with the diameter distribution necessary to produce the observed spectra. The highly directional nature of reflectance from cylinders may have significant consequences for clinical measurements.

Scanning laser polarimetry (SLP) is a commercially available technology for RNFL assessment that detects RNFL birefringence. The SLP instrument incorporates a "corneal compensator" to cancel corneal birefringence, a potentially confounding variable. Previous versions used fixed corneal compensation (FCC), i.e., the same fixed retarder was used for all patients. Using a corneal polarimeter of our own design, we showed that corneal birefringence varies widely in a population and is highly correlated with SLP-FCC measurements. We developed a clear understanding of SLP-FCC that led to SLP-VCC (variable corneal compensation), a technology that uses corneal compensator settings individualized for the eye under test. The result is now an accurate image of the retardance of the RNFL that should increase the sensitivity of SLP for glaucomatous damage.

Optical coherence tomography (OCT) is a commercially available technology that uses low coherence light to produce a cross-sectional image of the retina with 10 micrometer resolution. The RNFL appears as a brightly reflecting layer in an OCT image that allows precise measurement of RNFL thickness, a measurement that can be used for glaucoma diagnosis and management. Additionally, OCT images of the macula reveal the anatomic details of a variety of macular pathologies, including macular holes, subretinal and intraretinal fluid, and neovascular membranes. We have begun to investigate the reflectance properties of all fundus tissues that are imaged by OCT in order to improve the quantitative potential of this technology.