摘要:
Systems and methods of analyzing an optical coherence tomography image of a retina are discussed. A 2-dimensional slice of the image can be aligned to produce an approximately horizontal image of the retina and an edge map based at least in part on the aligned slice. Also, at least one global representation can be determined based on a (multi-scale) spatial division, such as multi-scale spatial pyramid, on the slice and/or edge map. Creating the local features is based on the specified cell structure of the global representation. The local features can be constructed based on local binary pattern (LBP)-based features. Additionally, a slice can be categorized into one or more categories via one or more classifiers (e.g., support vector machines). Each category can be associated with at least one ocular pathology, and classifying can be based on the constructed global descriptors, which can include the LBP-based local descriptors.
摘要:
Systems and methods of analyzing an optical coherence tomography image of a retina are discussed. A 2-dimensional slice of the image can be aligned to produce an approximately horizontal image of the retina and an edge map based at least in part on the aligned slice. Also, at least one global representation can be determined based on a (multi-scale) spatial division, such as multi-scale spatial pyramid, on the slice and/or edge map. Creating the local features is based on the specified cell structure of the global representation. The local features can be constructed based on local binary pattern (LBP)-based features. Additionally, a slice can be categorized into one or more categories via one or more classifiers (e.g., support vector machines). Each category can be associated with at least one ocular pathology, and classifying can be based on the constructed global descriptors, which can include the LBP-based local descriptors.
摘要:
A system. The system includes a computing device configured for communication with an imaging system and with a display device. The computing device includes a contour modeling module. The contour modeling module is configured for superimposing reference anchors on a cross-sectional image generated from 3D image data, for generating a line which connects the reference anchors, for sampling the 3D image data in a variable thickness plane defined by the connecting line, and for generating a contour-modeled C-mode image from the sampled 3D image data.
摘要:
A scan location matching (SLM) method identifies conventional time domain optical coherence tomography (TD-OCT) circle scan locations within three-dimensional spectral domain OCT scan volumes. A technique uses both the SLM algorithm and a mathematical retinal nerve fiber bundle distribution (RNFBD) model, which is a simplified version of the anatomical retinal axon bundle distribution pattern, to normalize TD-OCT thickness measurements for the retinal nerve fiber layer (RNFL) of an off-centered TD-OCT circle scan to a virtual location, centered on the optic nerve head. The RNFBD model eliminates scan-to-scan RNFL thickness measurement variation caused by manual placement of TD-OCT circle scan.
摘要:
A system. The system includes a computing device configured for communication with an imaging system and with a display device. The computing device includes a contour modeling module. The contour modeling module is configured for superimposing reference anchors on a cross-sectional image generated from 3D image data, for generating a line which connects the reference anchors, for sampling the 3D image data in a variable thickness plane defined by the connecting line, and for generating a contour-modeled C-mode image from the sampled 3D image data.
摘要:
Advances in optical coherence tomography (OCT) have prompted a transition from time domain OCT, providing 2D OCT images, to spectral domain OCT, which has a 3D imaging capability. Yet conventional technology offers little toward the goal of inter-device compatibility between extant 2D OCT images and newer 3D OCT images for the same or comparable subjects, as in the context of ongoing monitoring the quantitative status of a patient's eyes. The inventive methodology is particularly useful to identify the scan location of tissue in a 2D OCT image within the 3D OCT volumetric data, thereby allowing clinicians to image a patient via 3D OCT, based on available 2D OCT images, with minimal inter-device variation.
摘要:
A scan location matching (SLM) method identifies conventional time domain optical coherence tomography (TD-OCT) circle scan locations within three-dimensional spectral domain OCT scan volumes. A technique uses both the SLM algorithm and a mathematical retinal nerve fiber bundle distribution (RNFBD) model, which is a simplified version of the anatomical retinal axon bundle distribution pattern, to normalize TD-OCT thickness measurements for the retinal nerve fiber layer (RNFL) of an off-centered TD-OCT circle scan to a virtual location, centered on the optic nerve head. The RNFBD model eliminates scan-to-scan RNFL thickness measurement variation caused by manual placement of TD-OCT circle scan.
摘要:
Advances in optical coherence tomography (OCT) have prompted a transition from time domain OCT, providing 2D OCT images, to spectral domain OCT, which has a 3D imaging capability. Yet conventional technology offers little toward the goal of inter-device compatibility between extant 2D OCT images and newer 3D OCT images for the same or comparable subjects, as in the context of ongoing monitoring the quantitative status of a patient's eyes. The inventive methodology is particularly useful to identify the scan location of tissue in a 2D OCT image within the 3D volumetric data, thereby allowing clinicians to image a patient via 3D OCT, based on available 2D OCT images, with minimal inter-device variation.