Boris Gramatikov, Ph.D.
I am a member of the Hopkins faculty since 1996, and am currently Assistant Professor in the Division of Pediatric Ophthalmology and Adult Strabismus at The Wilmer Eye Institute. I am a biomedical engineer with expertise in medical instrumentation and signal processing. My present research interests are in retinal birefringence scanning and automated focus detection, including transition from the laboratory bench to clinical settings. The main goal is to identify and treat children with strabismus (misaligned eyes) or anisometropia (unequal refractive error) before irreversible amblyopia (functional monocular blindness) results. Click here for more on Clinical Technology Development at our Division.
Please click here for my web page at the Wilmer Eye Institute.
I have chaired the Baltimore Chapter of
the IEEE-EMBS (Institute of Electrical and
Electronics Engineers, Engineering
in Medicine and Biology Society), which is a part of the Baltimore Section of the IEEE. I
chaired the Section in 2006, and am presently an ExCom
member and The Section’s Director
for Education Activities and Continuing Education (CEEE). Click here
for a list of past
and planned future CEEE events. I
have also chaired the Baltimore Chapter of the IEEE-EMBS (Engineering in Medicine and Biology Society).
In 2009 I received the The Hartwell Foundation Individual Biomedical Research
Award for work on the Pediatric
Vision Screening Instrument for Early Detection of Amblyopia
(Lazy Eye). Read
the related article in the JHU Gazette…
I and my co-workers recently received a Biomedical
Research Collaboration Award from the Hartwell Foundation for developing
advanced technology for diagnosing retinal abnormalities in infants and
toddlers, in collaboration with Duke University. Read the JHU
announcement. Read more…
Current projects
Retinal birefringence scanning utilizes the optical
polarization properties of the human retina. This method was developed
originally by Dr. David Guyton and Dr. David Hunter ("Automated detection of eye fixation by use
of retinal birefringence scanning", Applied
Optics 38,OT&BO:1273-9, March 1999), and is related to the Haidinger brush phenomenon - a bow-tie or propeller-like
pattern that appears to rotate about the fixation point when a polarizing
filter is placed over the eye and rotated. This phenomenon reflects the retinal
nerve fiber axon arrangement about the fovea and can be utilized in a variety
of useful applications. In our settings, foveal fixation is monitored in
human subjects remotely and continuously by use of a noninvasive retinal scan.
Polarized near-infrared light is imaged onto the retina and scanned in a 3-deg
annulus at frequency f. Reflections are analyzed by differential
polarization detection. The detected signal is predominantly of frequency 2f during
central fixation, and f during paracentral
fixation. Phase shift at f
correlates with the direction of eye displacement. Mathematical modeling of birefringence in retinal birefringence scanning has confirmed the experimental and clinical
results. Potential applications of this technique include screening for
eye disease, eye position monitoring during clinical procedures, and use of eye
fixation to operate devices. Considering that the structure of the fovea is the
basis of the retinal birefringence scan signal, the experience gained from
bringing retinal birefringence scanning to the clinic will also increase our
understanding of foveal structure in health and other forms of disease.
Dr. Gramatikov is involved in the design of a portable Bilateral Retinal
Birefringence Scanner (BRBS) with improved noise performance, to be used as a pediatric vision screener. The new design incorporates binocular foveal birefringence scanning,
and separate channels for binocular focus detection. This combination of focus
and alignment detection should identify over 95% of all children at risk for amblyopia. This work is being performed in the past in
collaboration with AURA (Association of Universities for Research in Astronomy,
Inc.), the Space Telescope Science Institute and the Instrument Development Group (IDG) at the Department of Physics and
Astronomy at the Johns Hopkins University. This project was supported generously by The Hartwell Foundation with the 2009
Individual Biomedical Research Award.
Retinal
birefringence scanning combined with optical coherence tomography. Today, when a retinal condition is suspected in an
infant or toddler, the challenge to the ophthalmologist is to enable a child to
achieve and maintain ocular fixation during examination. To obtain a precise
retinal examination, the choices are placing a child under general anesthesia
(with many risks, such as pneumonia and potential death),
perform a limited examination, or worse, to wait until the child grows older to
identify what is wrong. For adults, where fixation and cooperation are usually
not an issue, a technology known as optical coherence tomography (OCT) is used
to provide high speed, 3-dimensional, and magnified cross-sectional images of
the retina (e.g., images depicting macula). This standard of care for
outpatient diagnosis and management of retinal disease in adults is
well-tolerated in older children. Unfortunately, the size of a conventional OCT
device is too large for deployment in an infant care setting. Because of The Hartwell Foundation
support to Dr. Toth, a state-of-the-art handheld OCT
imaging device has now been shown to be valuable in the assessment of premature
and newborn infants, but the apparatus must be held less than one inch from the
eye to achieve proper ocular alignment. Most neonates can be adequately
immobilized for retinal imaging, although not perfectly so. Infants, active
toddlers and young children will not cooperate for such imaging and thus a more
suitable version of the hand-held OCT instrument is desperately needed. Of
concern is also the quality of data, acquired during only short-lasting
episodes of central fixation (when the child’s eyes are looking in the right
direction). It is distressing that treatable retinal conditions in young
children are frequently missed because of this situation. To overcome the
limitations of current technology, Drs. Toth at Duke
and Gramatikov at Johns Hopkins were supported with
a 2012 Biomedical Research
Collaboration Award by The Hartwell Foundation collaboration.
They and their
collaborators are developing a swept-source optical coherence tomography (SSOCT)
retinal imaging system that is 10 times faster than current high speed OCT
systems, works from a comfortable viewing distance, and based upon retinal
birefringence scanning (RBS) technology offered by Drs. Gramatikov, Guyton and Irsch, will automatically detect central fixation.
A second generation Pediatric Vision Screener
using RBS has already been developed with funding from The Hartwell Foundation. The new system will have the
unsuspecting child view a cartoon on a small LCD computer screen, with no
distracting apparatus between the child and the screen. The examiner will be
able to retain optical alignment of the SSOCT device with the child’s eye while
invisible polarized light automatically detects when the eye is looking at
specific targets on the computer screen. Our goal will be a clinically suitable
imaging method for diagnosing retinal diseases and monitoring the progression
or response to ocular therapy without a need for sedation or anesthesia. We
believe that SSOCT will have the power to transform the current ability to
determine levels of retinal disease in infants and children and likelihood of
disease progression.
Determination of ocular defocus using the double-pass blur image of a
point source of light. The double-pass blur image of a point source is used
to determine the quality of focus of the human eye. This project is based on
earlier work by Dr. David Guyton, Dr. David Hunter and Nainesh
Gandhi. An apparatus that can determine a threshold of defocus was devised
using a circle/annulus (bulls-eye) photodiode that can distinguish focused from
defocused light. A monochromatic near-infrared light source (laser diode) is
employed. The light reflected from the fundus of the
eye is projected by a beamsplitter onto the bulls-eye
detector, which is optically conjugate to the laser diode. The photodetector signal obtained is a function of the amount
of defocus caused by the double-pass point spread reflected from the retina.
The signal is affected by refractive error, incomplete accommodation, media
opacities, and abnormalities of retinal reflection. One of the major risk
factors for amblyopia is refractive error, or, more
accurately, the degree of defocus the eye experiences. The defocus is perhaps
best quantified by measuring the size of the retinal blur circle. Such defocus
detection can be combined with eye-fixation monitoring to be used as a
screening tool to detect risk factors for amblyopia.
More about me
The Pediatric Vision Screener – early prototype
(2000-2001)
Current work on the Pediatric Vision Screener
Using computer modeling, optical technologies, modern electronic hardware, novel signal processing and optimization methods, we are presently redesigning and reconstructing the Pediatric Vision Screener, with the goal of making it a clinically relevant, market-oriented and competitive product.
Some spin-off products: a vision scanner using non-moving parts, a biometric (security) scanner using the birefringent properties of the retina, and others (please see publications).
More about me:
This page was last updated on Jan 21, 2013