Abstract: A method of obtaining one or more components from an image may include normalizing and pre-processing the image to obtain a processed image. Features may be extracted from the processed image. Neural-network-based regression may then be performed on the set of extracted features to predict the one or more components. These techniques may be applied, for example, to the problem of extracting and removing bone components from radiographic images, which may be thoracic (lung) images.

Abstract: A technique for registering multiple image representations relating to a common region of interest may involve producing multiple displacements from the image representations for a key point. A final displacement for the key point may be determined from correlations of the multiple displacements.

Abstract: A method of obtaining one or more components from an image may include normalizing and pre-processing the image to obtain a processed image. Features may be extracted from the processed image. Neural-network-based regression may then be performed on the set of extracted features to predict the one or more components. These techniques may be applied, for example, to the problem of extracting and removing bone components from radiographic images, which may be thoracic (lung) images.

Abstract: Retrieval of archival data from an application such as, for example, a CAD system, to aid in the interpretation of current data using the same application is disclosed. The data may comprise of medical imagery. The archival data may be used to transform a current set of images to be more consistent with a reference set of images to facilitate comparison between the sets of images. The archival data may also be used to influence markings and assessment of the current set of images. Additionally, the archival data may be used in conjugation with prior computer aided-detection markings or user identified regions to aid in the assessment of regions of interest identified in the current examination.

Abstract: A method and system for enhancing the display of computer-aided detection marks on a medical image without obscuring underlying details are disclosed. A medical image is obtained and processed by a computer-aided detection (CAD) algorithm. The image is displayed or printed with its corresponding CAD-detected regions superimposed on the image. In one embodiment, the characteristics of the CAD-detected region will be enhanced in order to emphasis the suspicious image features. In another embodiment, a region surrounding the CAD-detection regions will be brighten with enhanced contrast by brightening the pixels associated with that region as compared to the rest of the image so that region surrounding the CAD-detected region clearly stands out. When viewed on a display, the user can toggle the CAD-detected region enhancements on or off. Further, the CAD-detected regions can be cropped from the medical image to further improve the visibility of image attributes.