Schroedinger Eigenmaps for the analysis of biomedical data

Supervised learning and dimension reduction techniques for quantification of retinal fluid in optical coherence tomography images

PurposeThe purpose of the present study is to develop fast automated quantification of retinal fluid in optical coherence tomography (OCT) image sets.MethodsWe developed an image analysis pipeline tailored towards OCT images that consists of five steps for binary retinal fluid segmentation. The method is based on feature extraction, pre-segmention, dimension reduction procedures, and supervised learning tools.ResultsFluid identification using our pipeline was tested on two separate patient groups: one associated to neovascular age-related macular degeneration, the other showing diabetic macular edema. For training and evaluation purposes, retinal fluid was annotated manually in each cross-section by human expert graders of the Vienna Reading Center. Compared with the manual annotations, our pipeline yields good quantification, visually and in numbers.ConclusionsBy demonstrating good automated retinal fluid quantification, our pipeline appears useful to expert graders within their current grading processes. Owing to dimension reduction, the actual learning part is fast and requires only few training samples. Hence, it is well-suited for integration into actual manufacturer's devices, further improving segmentation by its use in daily clinical life.

Automated microaneurysm detection in diabetic retinopathy using curvelet transform

Microaneurysms (MAs) are known to be the early signs of diabetic retinopathy (DR). An automated MA detection system based on curvelet transform is proposed for color fundus image analysis. Candidates of MA were extracted in two parallel steps. In step one, blood vessels were removed from preprocessed green band image and preliminary MA candidates were selected by local thresholding technique. In step two, based on statistical features, the image background was estimated. The results from the two steps allowed us to identify preliminary MA candidates which were also present in the image foreground. A collection set of features was fed to a rule-based classifier to divide the candidates into MAs and non-MAs. The proposed system was tested with Retinopathy Online Challenge database. The automated system detected 162 MAs out of 336, thus achieved a sensitivity of 48.21% with 65 false positives per image. Counting MA is a means to measure the progression of DR. Hence, the proposed system may be deployed to monitor the progression of DR at early stage in population studies.

Automatic Image Registration of Multi-Modal Remotely Sensed Data with Global Shearlet Features

Automatic image registration is the process of aligning two or more images of approximately the same scene with minimal human assistance. Wavelet-based automatic registration methods are standard, but sometimes are not robust to the choice of initial conditions. That is, if the images to be registered are too far apart relative to the initial guess of the algorithm, the registration algorithm does not converge or has poor accuracy, and is thus not robust. These problems occur because wavelet techniques primarily identify isotropic textural features and are less effective at identifying linear and curvilinear edge features. We integrate the recently developed mathematical construction of shearlets, which is more effective at identifying sparse anisotropic edges, with an existing automatic wavelet-based registration algorithm. Our shearlet features algorithm produces more distinct features than wavelet features algorithms; the separation of edges from textures is even stronger than with wavelets. Our algorithm computes shearlet and wavelet features for the images to be registered, then performs least squares minimization on these features to compute a registration transformation. Our algorithm is two-staged and multiresolution in nature. First, a cascade of shearlet features is used to provide a robust, though approximate, registration. This is then refined by registering with a cascade of wavelet features. Experiments across a variety of image classes show an improved robustness to initial conditions, when compared to wavelet features alone.

Modeling Photo-Bleaching Kinetics to Create High Resolution Maps of Rod Rhodopsin in the Human Retina

We introduce and describe a novel non-invasive in-vivo method for mapping local rod rhodopsin distribution in the human retina over a 30-degree field. Our approach is based on analyzing the brightening of detected lipofuscin autofluorescence within small pixel clusters in registered imaging sequences taken with a commercial 488nm confocal scanning laser ophthalmoscope (cSLO) over a 1 minute period. We modeled the kinetics of rhodopsin bleaching by applying variational optimization techniques from applied mathematics. The physical model and the numerical analysis with its implementation are outlined in detail. This new technique enables the creation of spatial maps of the retinal rhodopsin and retinal pigment epithelium (RPE) bisretinoid distribution with an ≈ 50μm resolution.

Nonlinear gene cluster analysis with labeling for microarray gene expression data in organ development

Background

The gene networks underlying closure of the optic fissure during vertebrate eye development are not well-understood. We use a novel clustering method based on nonlinear dimension reduction with data labeling to analyze microarray data from laser capture microdissected (LCM) cells at the site and developmental stages (days 10.5 to 12.5) of optic fissure closure.

Results

Our nonlinear methods created clusters of genes that mapped onto more specific biological processes and functions related to eye development as defined by Gene Ontology at lower false discovery rates than conventional linear cluster algorithms. Our new methods build on the advantages of LCM to isolate pure phenotypic populations within complex tissues in order to identify systems biology relationships among critical gene products expressed at lower copy number.

Conclusions

The combination of LCM of embryonic organs, gene expression microarrays, and nonlinear dimension reduction with labeling is a potentially useful approach to extract subtle spatial and temporal co-variations within the gene regulatory networks that specify mammalian organogenesis and organ function. Our results motivate further analysis of nonlinear dimension reduction with labeling within other microarray data sets from LCM dissected tissues or other cell specific samples to determine the more general utility of our method for uncovering more specific biological functional relationships.