摘要:
The invention relates to off-center detector 3D X-ray or proton radiography reconstruction. Redundancy weighting with a steep weighting function around the iso-axis typically leads to artifacts in the reconstruction, for example, if inconsistencies between two nominal redundant projections occur, e.g. due to slightly incorrect detector calibration or scatter correction, etc. With the present invention, an approach is presented for overcoming or mitigating these problems.
摘要:
Disclosed herein is a medical system (100, 300, 400) comprising a local memory (118) storing local machine executable instructions (130) and a local computational system (110). Execution of the machine executable instructions further causes the computational system to: receive (200) a feature vector (132, 134, 136) descriptive of medical image data (410), wherein the feature vector is configured to be input into a decoder neural network (142), wherein the decoder neural network is configured to output an approximation of the medical image data (146) when receiving at least a part of the feature vector as input, wherein the feature vector comprises a ranking (702) assigning an importance to elements of the feature vector; and sequentially (202) transmit portions (132, 134, 136) of the feature vector to a remote computational system (120) via a network connection (114), wherein the portions of the features vector with a higher importance are transmitted first.
摘要:
A system and related method for determining position and composition of a clot in a patient. The system (SYS) comprises and input interface (IF) for receiving input imagery of a region of interest comprising contrasted input imagery and further input imagery contrasted lower than the contrasted input imagery. A location determiner (LD) of the system processes the contrasted imagery to determine a location of the clot. A composition determiner (CD) of the system determines a clot composition based on the further input imagery at the determined location. An indication of said clot location and clot composition is provided via output interface (OF).
摘要:
System (S-SYS) and related method for determining a material density map for a target material. The system (S-SYS) may receive spectral data representable in a two-dimensional data space. The spectral data may include measurements acquired by a spectral imaging apparatus of an object in a three-dimensional image domain of the spectral imaging apparatus. The system determines clusters in the data space, one indicative of the target material, the target material cluster, and clusters indicative auxiliary materials, the auxiliary material clusters. The system may determine a mutual geometrical constellation of the clusters. The system determines the material density map based on the geometrical constellation so determined.
摘要:
The present invention relates to a system (1) for adaptive segmentation. The system (1) comprises a configurator (10), which is configured to determine an adapted angular range (AR) with respect to an operation mode of the system (1) and which is configured to determine a segmentation parameter (SP) based on the adapted angular range (AR). Further, the system comprises an imaging sensor (20), which is configured to acquire images (I1, . . . , IN) within the adapted angular range (AR). Still further, the system comprises a segmentator (30), which is configured to generate a segmentation model based on the acquired images (I1, . . . , IN) using the determined segmentation parameter (SP).
摘要:
The present invention relates to a system (1) for adaptive segmentation. The system (1) comprises a configurator (10), which is configured to determine an adapted angular range (AR) with respect to an operation mode of the system (1) and which is configured to determine a segmentation parameter (SP) based on the adapted angular range (AR). Further, the system comprises an imaging sensor (20), which is configured to acquire images (I1, . . . , IN) within the adapted angular range (AR). Still further, the system comprises a segmentator (30), which is configured to generate a segmentation model based on the acquired images (I1, . . . , IN) using the determined segmentation parameter (SP).
摘要:
The present invention relates to an imaging system (10) for a vertebral level, an identification method for a vertebral level, a computer program element for controlling such system and a computer readable medium having stored such computer program element. The imaging system (10) comprises a determination unit (11), a definition unit (12), an imaging unit (13), and a processing unit (14). The determination unit (11) determines a target vertebral level. The definition unit (12) defines an anatomical landmark of a spine. The imaging unit (13) provides a series of X-ray images along the spine based on the landmark. The processing unit (14) identifies the target vertebral level in at least one of the X-ray images. The processing unit (14) further stitches the X-ray images to a continuous panoramic image of the spine and identifies the target vertebral level in the panoramic image.