Back to results
US 12,718,367 B2Granted
Classification of dynamically contrast-enhanced medical image data of a liver
AssigneeSiemens Healthineers AG
InventorsRobert Grimm, Cornelius Jacob
Life of this patent
Estimating the term…
Abstract
A method for classifying medical image data using a classification algorithm configured to make an image-based correlation between an image data set and a phase of a plurality of defined phases relative to a time of administration of a contrast agent with additional plausibility checking. The image data may include at least three image data sets in each case imaging the examination region in each case at a time of capture within a period of less than two hours.
Claims16
A method for classifying medical image data, comprising the following method steps:
- acquiring medical image data of an examination region including a liver of an object under examination, the medical image data including at least three image data sets, each of which images the examination region in each case at a time of capture within a period of less than two hours, wherein each of the at least three image data sets comprises metadata associated with at least one capture parameter and the time of capture;
- providing a classification algorithm configured to make an image-based correlation between an image data set and a phase of a plurality of defined phases relative to a time of administration of a contrast agent;
- extracting at least three image data sets from the medical image data based on the metadata;
- generating initially classified image data sets by applying the classification algorithm to the at least three extracted image data sets, wherein at least one phase is assigned to each of the three extracted image data sets;
- performing a plausibility check on the initially classified image data sets to generate verified classified image data, wherein:
- in response to an inadequate plausibility:
- outputting of an indication of the inadequate plausibility and/or performing a correction of the classification of the initially classified image data sets, and
- in response to a confirmed plausibility:
- retaining the classification of the initially classified image data sets;
- and
- providing an electronic signal representing the verified classified image data sets.
8 more elements
A classification system comprising:
- a first interface configured to acquire medical image data of an examination region including a liver of an object under examination, the medical image data including at least three image data sets, each of which images the examination region in each case at a time of capture within a period of less than two hours, wherein each of the at least three image data sets comprises metadata associated with at least one capture parameter and the time of capture;
- a second interface;
- and
- a computer configured to:
- perform, using a classification algorithm, an image-based correlation between an image data set and a phase of a plurality of defined phases relative to a time of administration of a contrast agent;
- extract at least three image data sets from the medical image data based on the metadata;
- generate initially classified image data sets by applying the classification algorithm to the at least three extracted image data sets, wherein at least one phase is assigned to each of the three extracted image data sets;
- perform a plausibility check on the initially classified image data sets to generate verified classified image data, wherein:
- in response to an inadequate plausibility:
- outputting of an indication of the inadequate plausibility and/or performing a correction of the classification of the initially classified image data sets, and
- in response to a confirmed plausibility:
- retaining the classification of the initially classified image data sets;
- and
- provide, via the second interface, an electronic signal representing the verified classified image data.
11 more elements
An apparatus comprising:
- one or more processors;
- and
- memory storing instructions that, when executed by the one or more processors, cause the apparatus to:
- acquire medical image data of an examination region including a liver of an object under examination, the medical image data including at least three image data sets, each of which images the examination region in each case at a time of capture within a period of less than two hours, wherein each of the at least three image data sets comprises metadata associated with at least one capture parameter and the time of capture;
- provide a classification algorithm configured to make an image-based correlation between an image data set and a phase of a plurality of defined phases relative to a time of administration of a contrast agent;
- extract at least three image data sets from the medical image data based on the metadata;
- generate initially classified image data sets by applying the classification algorithm to the at least three extracted image data sets, wherein at least one phase is assigned to each of the three extracted image data sets;
- perform a plausibility check on the initially classified image data sets to generate verified classified image data, wherein:
- in response to an inadequate plausibility:
- outputting of an indication of the inadequate plausibility and/or performing a correction of the classification of the initially classified image data sets, and
- in response to a confirmed plausibility:
- retaining the classification of the initially classified image data sets;
- and
- provide an electronic signal representing the verified classified image data sets.
11 more elements
Dependencies are inferred from the claim text.