Aflatoxin contaminated corn is dangerous for domestic animals when used as feed and cause liver cancer when consumed by human beings. Therefore, the ability to detect <i>A. flavus</i> and its toxic metabolite, aflatoxin, is important. The objective of this study is to measure <i>A. flavus</i> growth using hyperspectral technology and develop spectral signatures for <i>A. flavus</i>. Based on the research group's previous experiments using hyperspectral imaging techniques, it has been confirmed that the spectral signature of <i>A. flavus</i> is unique and readily identifiable against any background or surrounding surface and among other fungal strains. This study focused on observing changes in the <i>A. flavus</i> spectral signature over an eight-day growth period. The study used a visible-near-infrared hyperspectral image system for data acquisition. This image system uses focal plane pushbroom scanning for high spatial and high spectral resolution imaging. Procedures previously developed by the research group were used for image calibration and image processing. The results showed that while <i>A. flavus</i> gradually progressed along the experiment timeline, the day-to-day surface reflectance of <i>A. flavus</i> displayed significant difference in discreet regions of the wavelength spectrum. External disturbance due to environmental changes also altered the growth and subsequently changed the reflectance patterns of <i>A. flavus</i>.
We present an automated method to perform accurate, rapid, and objective measurement of the blood oxygen saturation over each segment of the retinal vascular hierarchy from dual-wavelength fundus images. Its speed and automation (2 s per entire image versus 20 s per segment for manual methods) enables detailed level-by-level measurements over wider areas. An automated tracing algorithm is used to estimate vessel centerlines, thickness, directions, and locations of landmarks such as bifurcations and crossover points. The hierarchical structure of the vascular network is recovered from the trace fragments and landmarks by a novel algorithm. Optical densities (OD) are measured from vascular segments using the minimum reflected intensities inside and outside the vessel. The OD ratio (ODR=OD600/OD570) bears an inverse relationship to systemic HbO2 saturation (SO2). The sensitivity for detecting saturation change when breathing air versus pure oxygen was calculated from the measurements made on six subjects and was found to be 0.0226 ODR units, which is in good agreement with previous manual measurements by the dual-wavelength technique, indicating the validity of the automation. A fully automated system for retinal vessel oximetry would prove useful to achieve early assessments of risk for progression of disease conditions associated with oxygen utilization.
The present report evaluated ultraviolet radiation (UVR) effects on the spectral signature of mycotoxin producing fungus <i>Aspergillus flavus</i> (<i>A. flavus</i>). Ultraviolet radiation has long been used to reduce microbe contamination and to inactivate mold spores. In view of the known effects of UVR on microorganisms, and because certain spectral bands in the signature of some fungi may be in the UV range, it is important to know the maximum acceptable limit of UVR exposure that does not significantly alter the fungal spectral signature and affect detection accuracy. A visible-near-infrared (VNIR) hyperspectral imaging system using focal plane pushbroom scanning for high spatial and spectral resolution imaging was utilized to detect any changes. <i>A. flavus</i> cultures were grown for 5 days and imaged after intermittent or continuous UVR treatment. The intermittent group was treated at 1-minute intervals for 10 minutes, and VNIR images were taken after each UVR treatment. The continuous group was irradiated for 10 minutes and imaged before and after treatment. A control sample group did not undergo UVR treatment, but was also imaged at 1-minute intervals for 10 minutes in the same manner as the intermittent group. Before and after UVR treatment, mean fungal sample reflectance was obtained through spatial subset of the image along with standard deviation and pre- and post-treatment reflectance was compared for each sample. Results show significant difference between the reflectances of treated and control <i>A. flavus</i> cultures after 10 min of UV radiation. Aditionally, the results demonstrate that even lethal doses of UVR do not immediately affect the spectral signature of <i>A. flavus</i> cultures suggesting that the excitation UV light source used in the present experiment may be safe to use with the UV hyperspectral imaging system when exposure time falls below 10 min.
USDA and the Institute for Technology Development are currently collaborating on a project using hyperspectral imagery to detect pathogens such as mycotoxin producing molds in grain products. The initial experiments are being implemented on corn kernels. When molds appear on corn, reflectance spectra from the molds and corn are mixed. Therefore, it is important to characterize the corn reflectance, which is the background reflectance in the image. The objective of this study was to qualitatively identify and quantify kernel signatures of several corn genotypes. Four different corn genotypes (genetically distinct corn lines) and four near isogenic corn lines were prepared at the USDA laboratory. The study used a visible-near-infrared hyperspectral imaging system for data acquisition. The imaging system utilizes focal plane pushbroom scanning for high spatial and high spectral resolution imaging. Procedures were developed for optimum image calibration and image processing. It was expected that the results would be useful for reducing the background influence of corn in mold detection and would also be applicable in corn genotype identification, especially among corn lines with different resistance levels to molds.
We describe a non-invasive in vivo hyperspectral imaging technique for visualizing the spatial distribution of retina and optic nerve head (ONH) tissue oxygenation. Real time images of the fundus are acquired with continuous wavelengths (410-918 nm) to generate a data cube consisting of one spectral and two spatial dimensions. Reflected light from the one-dimensional (1-D) area of the sample is first passed through a grating and is then imaged onto a 12-bit silicone charge- coupled device (CCD) detector. A scanner then proceeds to the next 1-D area of the sample. Acquired image frames contain 256 spatial pixels and 256 wavelengths along rows and columns. Image sequences are scanned along the perpendicular spatial dimension using the push-broom method, whereby the spectrograph and camera are translated under constant velocity with respect to the fundus camera image over 6.6 mm of travel. This set of acquired images contains the full reflected light spectrum at each pixel of a two dimensional area of the retina and ONH. The system employs a focal plane scanner (FPS) using a linear actuator to provide motion. An algorithm processes spectral information at each pixel to represent the varying spatial distribution of retina and ONH tissue oxygenation. Imaging data are obtained from ONH tissue at both normal intraocular pressure (IOP) and acutely raised IOP.
Functional imaging of the retina and associated structures may provide information for early assessment of risks of developing retinopathy in diabetic patients. Here we show results of retinal oximetry performed using multi-spectral reflectance imaging techniques to assess hemoglobin (Hb) oxygen saturation (OS) in blood vessels of the inner retina and oxygen utilization at the optic nerve in diabetic patients without retinopathy and early disease during experimental hyperglycemia. Retinal images were obtained through a fundus camera and simultaneously recorded at up to four wavelengths using image-splitting modules coupled to a digital camera. Changes in OS in large retinal vessels, in average OS in disk tissue, and in the reduced state of cytochrome oxidase (CO) at the disk were determined from changes in reflectance associated with the oxidation/reduction states of Hb and CO. Step to high sugar lowered venous oxygen saturation to a degree dependent on disease duration. Moderate increase in sugar produced higher levels of reduced CO in both the disk and surrounding tissue without a detectable change in average tissue OS. Results suggest that regulation of retinal blood supply and oxygen consumption are altered by hyperglycemia and that such functional changes are present before clinical signs of retinopathy.