No abstract

No abstract

Abstract. Recent developments in Fourier domain-optical coherence tomography ͑Fd-OCT͒ have increased the acquisition speed of current ophthalmic Fd-OCT instruments sufficiently to allow the acquisition of volumetric data sets of human retinas in a clinical setting. The large size and three-dimensional ͑3D͒ nature of these data sets require that intelligent data processing, visualization, and analysis tools are used to take full advantage of the available information. Therefore, we have combined methods from volume visualization, and data analysis in support of better visualization and diagnosis of Fd-OCT retinal volumes. Custom-designed 3D visualization and analysis software is used to view retinal volumes reconstructed from registered B-scans. We use a support vector machine ͑SVM͒ to perform semiautomatic segmentation of retinal layers and structures for subsequent analysis including a comparison of measured layer thicknesses. We have modified the SVM to gracefully handle OCT speckle noise by treating it as a characteristic of the volumetric data. Our software has been tested successfully in clinical settings for its efficacy in assessing 3D retinal structures in healthy as well as diseased cases. Our tool facilitates diagnosis and treatment monitoring of retinal diseases.

Abstract-We have combined methods from volume visualization and data analysis to support better diagnosis and treatment of human retinal diseases. Many diseases can be identified by abnormalities in the thicknesses of various retinal layers captured using optical coherence tomography (OCT). We used a support vector machine (SVM) to perform semi-automatic segmentation of retinal layers for subsequent analysis including a comparison of layer thicknesses to known healthy parameters. We have extended and generalized an older SVM approach to support better performance in a clinical setting through performance enhancements and graceful handling of inherent noise in OCT data by considering statistical characteristics at multiple levels of resolution. The addition of the multi-resolution hierarchy extends the SVM to have "global awareness." A feature, such as a retinal layer, can therefore be modeled within the SVM as a combination of statistical characteristics across all levels; thus capturing high-and low-frequency information. We have compared our semi-automatically generated segmentations to manually segmented layers for verification purposes. Our main goals were to provide a tool that could (i) be used in a clinical setting; (ii) operate on noisy OCT data; and (iii) isolate individual or multiple retinal layers in both healthy and disease cases that contain structural deformities.

No abstract