This paper presents an image processing technique for automatically categorize age-related macular degeneration
(AMD) phenotypes from retinal images. Ultimately, an automated approach will be much more precise and consistent in
phenotyping of retinal diseases, such as AMD. We have applied the automated phenotyping to retina images from a
cohort of mono- and dizygotic twins. The application of this technology will allow one to perform more quantitative
studies that will lead to a better understanding of the genetic and environmental factors associated with diseases such as
AMD. A method for classifying retinal images based on features derived from the application of amplitude-modulation
frequency-modulation (AM-FM) methods is presented. Retinal images from identical and fraternal twins who presented
with AMD were processed to determine whether AM-FM could be used to differentiate between the two types of twins.
Results of the automatic classifier agreed with the findings of other researchers in explaining the variation of the disease
between the related twins. AM-FM features classified 72% of the twins correctly. Visual grading found that genetics
could explain between 46% and 71% of the variance.
The purpose of this study is to show that there exists a spectral characteristic that differentiates normal macular tissue
from various types of genetic-based macular diseases. This paper demonstrates statistically that hyperspectral images of
macular and other retinal tissue can be used to spectrally differentiate different forms of age-related macular
degeneration. A hyperspectral fundus imaging device has been developed and tested for the purpose of collecting
hyperspectral images of the human retina. A methodology based on partial least squares and ANOVA has been applied
to determine the hyperspectral representation of individual spectral characteristics of retinal features. Each discrete
tissue type in the retina has an identifiable spectral shape or signature which, when combined with spatial context, aids
in detection of pathological features. Variations in the amount and distribution of various ocular pigments or the
inclusion of additional biochemical substances will allow detection of pathological conditions prior to traditional
histological presentation. Fundus imaging cameras are ubiquitous and are one of the most common imaging modalities
used in documenting a patient's retinal state for diagnosis, e.g. remotely, or for monitoring the progression of an ocular
disease. The added diagnostic information obtained with only a minor retro-fit of a specialized spectral camera will lead
to new diagnostic information to the clinical ophthalmologist or eye-care specialist.
AirSentinel® is a new low cost, compact ultraviolet-based light induced fluorescence (UV-LIF) bio-aerosol threat detection trigger. Earlier UV-LIF triggers, for example, FLAPS, BARTS, BAWS, Bioni, and BioLert, have used UV laser sources to induce fluorescence of biological aerosols. Two recent developments from the DARPA MTO SUVOS program, BAST and TAC-BIO, use UV LEDs for the same purpose, thereby broadening the term UV-LIF to mean laser or LED induced autofluorescence. All of these earlier triggers interrogate aerosols on a particle-by-particle basis on- the-fly. The major trade-off for these instruments is cost, size, and complexity versus counting efficiency (probability of detection) with the lower size end of the respirable range being most difficult to detect. AirSentinel® employs a different approach to UV-LIF detection: aerosol concentration by collection on a surface, surface interrogation, and surface rejuvenation prior to repeated concentration and interrogation cycles. Aerosol particle concentration via impaction on a surface addresses the issue of small particle counting efficiency since the fluorescence from the sum of the particles is the sum of the fluorescence signals from the collected particles, typically hundreds or thousands in number. Surface interrogation for a LIF signal is accomplished by illumination with a 280 nm and/or a 365 nm LED. As expected, test results show better relative detection performance using 280 nm excitation LEDs for bio-toxin simulants and somewhat better performance at 365 nm for standard Bacillus globigii spore targets. AirSentinel® beta technology is currently in long term testing in a number of public and other government buildings.
The Receiver Operating Characteristic curve (ROC) has long been used in medical applications to compare screening and diagnostic methods. As the threshold used by any screening or diagnostic method is changed, the operating characteristics of the method, such as the number of true positive and false negative determinations changes as well. The ROC curve is one way to characterize the changes in order to compare different methods. This definition, however, is difficult to apply to chemical and biological sensors detecting the release of a toxic agent given that there is more than one ROC curve. There is a continuum of ROC curves corresponding to a continuum of release levels. A new definition of ROC curves has been adopted for chemical and biological sensors which will reduce the continuum of curves to a single curve. This paper presents a methodology to estimate ROC curves using this new definition.
An integrated radar and ultrasound sensor, capable of remotely detecting and imaging concealed weapons, is being developed. A modified frequency-agile, mine-detection radar is intended to specify with high probability of detection at ranges of 1 to 10 m which individuals in a moving crowd may be concealing metallic or nonmetallic weapons. Within about 1 to 5 m, the active ultrasound sensor is intended to enable a user to identify a concealed weapon on a moving person with low false-detection rate, achieved through a real-time centimeter-resolution image of the weapon. The goal for sensor fusion is to have the radar acquire concealed weapons at long ranges and seamlessly hand over tracking data to the ultrasound sensor for high-resolution imaging on a video monitor. We have demonstrated centimeter-resolution ultrasound images of metallic and non-metallic weapons concealed on a human at ranges over 1 m. Processing of the ultrasound images includes filters for noise, frequency, brightness, and contrast. A frequency-agile radar has been developed by JAYCOR under the U.S. Army Advanced Mine Detection Radar Program. The signature of an armed person, detected by this radar, differs appreciably from that of the same person unarmed.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.