摘要:
A system and method for managing medical image data items. A large image manager is used to store medical image data items external to a core PACS. Representative objects are generated and provided to the PACS to be incorporated into the core PACS workflows. The representative objects can be generated by extracted pixel data and metadata from the medical image data items. The representative objects are generated in a format compatible with the protocol used by the PACS workflows. Requests sent to the PACS for medical images can be re-routed to the large image manager to provide the requested images.
摘要:
A pipe-line method for multi-label segmentation of anatomic structures in a medical image using a convolutional neural network trained with a weighted loss function takes into account under—representation of at least one anatomical structure in a ground-truth mask relative to other anatomical structures. Different architectures for the convolutional neural network are described.
摘要:
A method and system to correct for alignment errors between assumed and actual geometric parameters of an acquisition geometry during image reconstruction in a chest tomosynthesis application includes receiving at least 2 raw projection images acquired on at least 2 different positions in a known acquisition geometry, determining an actual geometric parameter value by determining the minimum of a redundant planes cost function which is calculated for a varying range of the geometric parameter values, and which is determined by: a) at least one plane which intersects an X-ray source trajectory with at least two points, b) an intersection between the planes and a detector surface for which points the source positions are determined, and c) for which the parameters determining the intersection (λ, μ, l) are used for the construction of the cost function, applying the calculated actual geometric parameter value of the acquisition geometry for the image reconstruction of the plurality of images, characterized in that, a weighting function is applied to the plurality of acquired images prior to calculating the cost function.
摘要:
A method for training an artificial neural network on an additional untrained segmentation task prevents the loss of previously acquired segmentation skills on originally trained segmentation tasks.
摘要:
A short axis in a 3 dimensional image of a lesion is determined starting from voxels defining the long axis and voxels in the plane of the long axis. Voxels within the plane of the long axis are projected perpendicularly onto the long axis and receive an identifier indicative of the region on the long axis onto which they are projected. Distances between points (projected sub-voxels) in pairs of points within the same range and within adjacent ranges are evaluated in order to determine the longest distance.
摘要:
A method for reshaping the characteristic exposure response and dosimetry of a direct radiography system into a specified exposure response profile includes pixel-wise converting image data according to a response transfer model which is derived from an x-ray generator's post exposure parameters data associated with image signals obtained at various exposure levels during system calibration and from a few extra dose measurements and their corresponding post exposure data also gathered during system calibration under reference exposure conditions.
摘要:
A method, apparatus, and system for reliably identifying a specific part of a spine in an image of a human or animal body, includes the steps of determining one or more parts of the spine in the image, determining one or more discriminative parameters for each of the one or more parts of the spine in the image, the discriminative parameters relating to at least one anatomical property of each of the one or more parts of the spine, classifying the discriminative parameters of the one or more parts of the spine in the image, and identifying a specific part of the spine based on the classification of the discriminative parameters of the one or more parts of the spine in the image. An identified vertebra, in particular the T12 vertebra and/or its associated intervertebral discs, can be used advantageously as a starting point of powerful automatic spine labeling algorithms.
摘要:
A method includes selecting, in a radiographic exposure unit, a direct radiographic panel as an active panel for a forthcoming radiographic exposure, activating a gravity sensor installed on the direct radiographic panel, and activating by the activated gravity sensor a processor installed on the direct radiographic panel. As a result of a communication from the processor with a radiographic work station over a network, the activated direct radiographic panel is retained as the active direct radiographic panel for the forthcoming radiographic exposure.
摘要:
A process for purifying a compound of formula 1, includes the following steps: a) adding an acid to an aqueous solution of the compound of formula 1, including salts and hydrates thereof so as to obtain a slurry having a pH≦3; and b) filtering the slurry and at least one time washing the obtained precipitate with a liquid comprising water; and c) dissolving the precipitate obtained in step b) in water to obtain an aqueous solution; and d) filtering of the solution obtained in step c) over a nanofiltration membrane having a Molecular Weight Cut Off in the range from 150 to 500 and wherein optionally, between step c) and step d) the pH of the aqueous solution is adjusted to a pH value in the pH range as specified by the manufacturer of the nanofiltration membrane. A process for preparing a gadolinium complex of the purified compound of formula 1 is also disclosed. This gadolinium complex can be used for making a pharmaceutical composition as a contrast agent for magnetic resonance imaging.
摘要:
A method, apparatus, and system for reliably identifying a specific part of a spine in an image of a human or animal body, includes the steps of determining one or more parts of the spine in the image, determining one or more discriminative parameters for each of the one or more parts of the spine in the image, the discriminative parameters relating to at least one anatomical property of each of the one or more parts of the spine, classifying the discriminative parameters of the one or more parts of the spine in the image, and identifying a specific part of the spine based on the classification of the discriminative parameters of the one or more parts of the spine in the image. An identified vertebra, in particular the T12 vertebra and/or its associated intervertebral discs, can be used advantageously as a starting point of powerful automatic spine labeling algorithms.