Computer-implemented method for classifying a body type
A computer-implemented method is for classifying a body type of at least one person. In an embodiment, the method includes receiving at least one image data record of the respective person, which maps at least one subarea of the person; and ascertaining a body type for the person by an optimization method. A respective person model is used for each of the possible body types, which as a function of at least one person parameter determines an expected person geometry of the person described by the person model. The body type is selected by the optimization method by a similarity measure for the similarity of the image data record or a person geometry ascertained from the image data record being optimized with the expected person geometry by selecting the body type.
1. A computer-implemented method for classifying a body type of at least one person, the computer-implemented method comprising:
receiving at least one respective image data record of at least one respective person, the at least one respective image data record mapping at least one subarea of the at least one respective person; and
ascertaining a body type for the at least one respective person by an optimization method, the body type being selected from a group of body types including a rectangular body shape, a triangular body shape, an inverse triangular body shape, a trapeziform body shape, an oval body shape, and an hourglass body shape, wherein
a respective person model is defined for each possible body type of the group of body types, each respective person model providing an expected person geometry as a function of at least one person parameter,
the body type is ascertained by the optimization method by
inputting the at least one person parameter of the at least one respective person into each respective person model to obtain a set of expected person geometries, the set of expected person geometries including the expected person geometry of each possible body type of the group of body types,
determining a similarity measure between the at least one respective image data record and the expected person geometry for each respective person model, and
selecting the body type for the at least one respective person as the expected person geometry with a highest similarity measure to the at least one respective image data record,
a respective body type is determined for all people in a group of people,
each of a plurality of feature vectors, including the respective body type and the at least one person parameter as entries, is determined for each respective person and a cluster analysis of the plurality of feature vectors is carried out, by which the plurality of feature vectors are assigned to a fixed number of clusters or a number of clusters determined within a scope of the cluster analysis,
a respective characteristic feature vector is determined for each respective cluster, and
each respective cluster corresponds to a body type of the group of body types;
determining at least one operating parameter of a medical therapy device or a medical diagnostics device based upon the body type of the at least one respective person;
adjusting an operating protocol for a therapy application for the at least one respective person by transferring the at least one operating parameter to a control unit of the medical therapy device or the medical diagnostics device; and
operating the medical therapy device or the medical diagnostics device to perform the therapy application based on the adjusted operating protocol.
2. The computer-implemented method of claim 1 , wherein the at least one person parameter is at least one of
determined in a scope of the optimization method, or
received together with the at least one respective image data record.
3. The computer-implemented method of claim 2 , wherein at least one of:
at least one of size, weight or gender of the person is used as the at least one person parameter, or
at least one variable, dependent upon the at least one of the size or the weight is used as the at least one person parameter.
4. The computer-implemented method of claim 2 , wherein the similarity measure depends on at least one detection parameter, and wherein the at least one detection parameter at least one of
relates to detection of the at least one respective image data record and respectively determined scope of the optimization method, or
is received together with the at least one respective image data record.
5. The computer-implemented method of claim 4 , wherein the at least one detection parameter relates to at least one of a position or orientation of the at least one respective person with respect to a detection device used to detect the at least one respective image data record.
6. The computer-implemented method of claim 1 , wherein at least one of:
at least one of size, weight or gender of the person is used as the at least one person parameter, or
at least one variable, dependent upon the at least one of the size or the weight is used as the at least one person parameter.
7. The computer-implemented method of claim 1 , wherein the similarity measure depends on at least one detection parameter, and wherein the at least one detection parameter at least one of
relates to detection of the at least one respective image data record and respectively determined scope of the optimization method, or
is received together with the at least one respective image data record.
8. The computer-implemented method of claim 7 , wherein the at least one detection parameter relates to at least one of a position or orientation of the at least one respective person with respect to a detection device used to detect the at least one respective image data record.
9. The computer-implemented method of claim 1 , wherein a further image data record is detected for at least one further person, and wherein at least one control parameter, upon which detection of the further image data record depends, is determined as a function of the respective characteristic feature vector for each respective cluster.
10. The computer-implemented method of claim 9 , wherein the at least one control parameter is at least one of an x-ray dose or a contrast agent quantity.
11. The computer-implemented method of claim 1 , wherein as the function of the at least one person parameter, the respective person model determines an expected three-dimensional body surface of the at least one respective person described by the respective person model as the expected person geometry.
12. The computer-implemented method of claim 1 , wherein the at least one respective image data record describes a two-dimensional x-ray recording, and wherein the similarity measure depends on an expansion of the expected person geometry at right angles to an image plane of the two-dimensional x-ray recording and on an absorption intensity, described by x-ray detection, of x-ray radiation through the at least one respective person.
13. The computer-implemented method of claim 1 , wherein the expected person geometry of the ascertained body type describes a two-dimensional outline or a three-dimensional surface of the at least one respective person or the at least one mapped subarea of the at least one respective person.
14. The computer-implemented method of claim 1 , wherein at least one of an image data record of a two-dimensional x-ray image, an image data record of at least one of a three-dimensional computed tomography examination, a magnetic resonance tomography examination, ultrasound measuring data, or image data of a 3D camera is used as the at least one respective image data record.
15. A non-transitory computer program product for a processor, storing program instructions, to carry out the computer-implemented method of claim 1 when carried out on the processor.
16. A non-transitory machine-readable data carrier, storing a computer program including program instructions, to carry out the computer-implemented method of claim 1 when carried out on a processor.
17. The computer-implemented method of claim 1 , wherein the medical therapy device or the medical diagnostics device is a radiation device or an imaging device.
18. The computer-implemented method of claim 1 , wherein each of the respective characteristic feature vector for each respective cluster determines a control parameter used in a first parameterization of new image data.
19. A processing device of a medical therapy device or a medical diagnostics device, the processing device comprising:
at least one processor, the at least one processor being configured to at least cause the processing device to
receive at least one respective image data record of at least one respective person, the at least one respective image data record mapping at least one subarea of the at least one respective person; and
ascertain a body type for the at least one respective person by an optimization method, the body type being selected from a group of body types including a rectangular body shape, a triangular body shape, an inverse triangular body shape, a trapeziform body shape, an oval body shape, and an hourglass body shape, wherein
a respective person model is defined for each possible body type of the group of body types, each respective person model providing an expected person geometry as a function of at least one person parameter,
the body type is ascertained by the optimization method by
inputting the at least one person parameter of the at least one respective person into each respective person model to obtain a set of expected person geometries, the set of expected person geometries including the expected person geometry of each possible body type of the group of body types,
determining a similarity measure between the at least one respective image data record and the expected person geometry for each respective person model, and
selecting the body type for the at least one respective person as the expected person geometry with a highest similarity measure to the at least one respective image data record,
a respective body type is determined for all people in a group of people,
each of a plurality of feature vectors, including the respective body type and the at least one person parameter as entries, is determined for each respective person and a cluster analysis of the plurality of feature vectors is carried out, by which the plurality of feature vectors are assigned to a fixed number of clusters or a number of clusters determined within a scope of the cluster analysis,
a respective characteristic feature vector is determined for each respective cluster, and
each respective cluster corresponds to a body type of the group of body types,
determine at least one operating parameter of the medical therapy device or the medical diagnostics device based upon the body type of the at least one respective person;
adjust an operating protocol for a therapy application for the at least one respective person based on the at least one operating parameter; and
operating the medical therapy device or the medical diagnostics device to perform the therapy application based on the adjusted operating protocol.