Glaucoma is neurodegenerative disease characterized by distinctive changes in the optic nerve head and visual field. Without treatment, glaucoma can lead to permanent blindness. Therefore, monitoring glaucoma progression is important to detect uncontrolled disease and the possible need for therapy advancement. In this context, three-dimensional (3-D) spectral domain optical coherence tomography (SD-OCT) has been commonly used in the diagnosis and management of glaucoma patients. We present a new framework for detection of glaucoma progression using 3-D SD-OCT images. In contrast to previous works that use the retinal nerve fiber layer thickness measurement provided by commercially available instruments, we consider the whole 3-D volume for change detection. To account for the spatial voxel dependency, we propose the use of the Markov random field (MRF) model as a prior for the change detection map. In order to improve the robustness of the proposed approach, a nonlocal strategy was adopted to define the MRF energy function. To accommodate the presence of false-positive detection, we used a fuzzy logic approach to classify a 3-D SD-OCT image into a “non-progressing” or “progressing” glaucoma class. We compared the diagnostic performance of the proposed framework to the existing methods of progression detection.
Confocal microscopes (CM) are routinely used for building 3-D images of microscopic structures. Nonideal imaging conditions in a white-light CM introduce additive noise and blur. The optical section images need to be restored prior to quantitative analysis. We present an adaptive noise filtering technique using Karhunen–Loéve expansion (KLE) by the method of snapshots and a ringing metric to quantify the ringing artifacts introduced in the images restored at various iterations of iterative Lucy–Richardson deconvolution algorithm. The KLE provides a set of basis functions that comprise the optimal linear basis for an ensemble of empirical observations. We show that most of the noise in the scene can be removed by reconstructing the images using the KLE basis vector with the largest eigenvalue. The prefiltering scheme presented is faster and does not require prior knowledge about image noise. Optical sections processed using the KLE prefilter can be restored using a simple inverse restoration algorithm; thus, the methodology is suitable for real-time image restoration applications. The KLE image prefilter outperforms the temporal-average prefilter in restoring CM optical sections. The ringing metric developed uses simple binary morphological operations to quantify the ringing artifacts and confirms with the visual observation of ringing artifacts in the restored images.
Purpose: To develop a multi-spectral method to measure oxygen saturation of the retina in the human eye.
Methods: Five Cynomolgus monkeys with normal eyes were anesthetized with intramuscular ketamine/xylazine and intravenous pentobarbital. Multi-spectral fundus imaging was performed in five monkeys with a commercial fundus camera equipped with a liquid crystal tuned filter in the illumination light path and a 16-bit digital camera. Recording parameters were controlled with software written specifically for the application. Seven images at successively longer oxygen-sensing wavelengths were recorded within 4 seconds. Individual images for each wavelength were captured in less than 100 msec of flash illumination. Slightly misaligned images of separate wavelengths due to slight eye motion were registered and corrected by translational and rotational image registration prior to analysis. Numerical values of relative oxygen saturation of retinal arteries and veins and the underlying tissue in between the artery/vein pairs were evaluated by an algorithm previously described, but which is now corrected for blood volume from averaged pixels (n > 1000). Color saturation maps were constructed by applying the algorithm at each image pixel using a Matlab script.
Results: Both the numerical values of relative oxygen saturation and the saturation maps correspond to the physiological condition, that is, in a normal retina, the artery is more saturated than the tissue and the tissue is more saturated than the vein. With the multi-spectral fundus camera and proper registration of the multi-wavelength images, we were able to determine oxygen saturation in the primate retinal structures on a tolerable time scale which is applicable to human subjects.
Conclusions: Seven wavelength multi-spectral imagery can be used to measure oxygen saturation in retinal artery, vein, and tissue (microcirculation). This technique is safe and can be used to monitor oxygen uptake in humans.
This work is original and is not under consideration for publication elsewhere.
We present a fractal measure based pattern classification algorithm for automatic feature extraction and identification of fungus associated with an infection of the cornea of the eye. A white-light confocal microscope image
of suspected fungus exhibited locally linear and branching structures. The pixel intensity variation across the
width of a fungal element was gaussian. Linear features were extracted using a set of 2D directional matched
gaussian-filters. Portions of fungus profiles that were not in the same focal plane appeared relatively blurred. We
use gaussian filters of standard deviation slightly larger than the width of a fungus to reduce discontinuities. Cell
nuclei of cornea and nerves also exhibited locally linear structure. Cell nuclei were excluded by their relatively
shorter lengths. Nerves in the cornea exhibited less branching compared with the fungus. Fractal dimensions
of the locally linear features were computed using a box-counting method. A set of corneal images with fungal
infection was used to generate class-conditional fractal measure distributions of fungus and nerves. The a priori
class-conditional densities were built using an adaptive-mixtures method to reflect the true nature of the feature
distributions and improve the classification accuracy. A maximum-likelihood classifier was used to classify
the linear features extracted from test corneal images as 'normal' or 'with fungal infiltrates', using the a priori
fractal measure distributions. We demonstrate the algorithm on the corneal images with culture-positive fungal
infiltrates. The algorithm is fully automatic and will help diagnose fungal keratitis by generating a diagnostic
mask of locations of the fungal infiltrates.
We present a noise filtering technique using Karhunen-Loeve expansion by the method of snapshots (KLS) using a small ensemble of 3 images. The KLS provides a set of basis functions which comprise the optimal linear basis for the description of an ensemble of empirical observations. The KLS basis is computed using the eigenvectors of the covariance matrix R of the ensemble of images. The significance of each of the basis functions is determined by the magnitude of the corresponding eigenvalues of R, the largest being the most significant. Since all the three images in the ensemble represent the same scene and are registered, the KLS basis construct using the eigenvectors of R with the least eigenvalues typically represent the non-significant and uncommon features in the ensemble. We show that most of the noise in the scene can be removed by reconstructing the image using the KLS basis function constructed using the eigenvector of R with the largest eigenvalue. R is 3x3 symmetric positive definite matrix and hence has a full set of orthogonal eigenvectors. The KLS filtering scheme described here is faster and does not require prior knowledge about the image noise. We show the performance of the proposed method on the images of random cotton fibers acquired using white-light confocal microscope (WLCM) and compare the performance with a median filter. Also, we show that a simple inverse-filter deconvolution algorithm provides an impressive image restoration by pre-filtering the images using the proposed KLS filtering technique.