WO2024083466A1

WO2024083466A1 - Automated analysis of radiological images

Info

Publication number: WO2024083466A1
Application number: PCT/EP2023/076726
Authority: WO
Inventors: Felix Karl KREIS; Gesine KNOBLOCH; Matthias LENGA; Ivo Matteo BALTRUSCHAT; Gregor Jost; Gunnar SCHÜTZ; Hubertus Pietsch
Original assignee: Bayer Aktiengesellschaft
Priority date: 2022-10-17
Filing date: 2023-09-27
Publication date: 2024-04-25

Abstract

The present invention relates to the analysis of radiological images. The invention relates inter alia to a computer-implemented method, a computer system and a computer program for segmenting radiological images and/or for identifying anomalies in radiological images.

Description

BHC221036 FC – SV/JW 2023-08-15 - 1 - Automatic analysis of radiological images The present disclosure relates to the analysis of radiological images. Subject matters of the present invention include a computer-implemented method, a computer system and a computer program for segmenting radiological images and/or for identifying abnormalities in radiological images. Medical imaging is the technique and process of imaging the interior of the body for clinical analysis and medical interventions, as well as visually representing the function of specific organs or tissues. The purpose of medical imaging is to make internal structures hidden beneath the skin and bones visible and to diagnose and/or treat diseases. Advances in imaging and machine learning have led to a rapid increase in the potential use of artificial intelligence in various medical imaging tasks, such as risk assessment, detection, diagnosis, prognosis and therapy. Machine learning models are used, among other things, to segment radiological images and/or to identify lesions and/or other abnormalities. WO2021/038203A1, for example, discloses a method for automatically segmenting structural features of blood vessels in radiological images using a trained machine learning model. In a first step, the machine learning model is trained using training data. The training data includes a large number of non-segmented radiological images (as input data) and manually segmented radiological images (as target data). The non-segmented images are presented to the model and the model generates segmented images (output data) based on the non-segmented images and model parameters. The segmented images (output data) are compared with the target data. The model parameters are modified in an optimization process to minimize deviations between the segmented images generated by the model and the target data. Once the model has been trained, it can be used to segment new radiological images, i.e. images not used in training. WO2021/183765A1 discloses a method for identifying lesions using a machine learning model. The model disclosed in WO2021/183765A1 is also trained using training data, the training data comprising a large number of manually labeled radiological images. The labels indicate whether and where lesions can be seen in radiological images. Training such machine learning models requires a great deal of effort. On the one hand, a large number of radiological images must be provided, and on the other hand, they must be manually annotated (segmented and/or labeled). It is conceivable that for some abnormalities there is not enough data available to be used for training. This is particularly true for rare diseases. It would be desirable to be able to segment radiological images and/or identify abnormalities in radiological images without having to provide and annotate a large number of radiological images for training beforehand. In particular, it would be desirable to be able to identify abnormalities in radiological images without having to provide radiological images with abnormalities beforehand. This object is achieved by the subject matter of the present independent patent claims. Preferred embodiments can be found in the dependent patent claims as well as in the present description and in the drawings. A first subject matter of the present disclosure is a computer-implemented method comprising: - Receiving radiological images, the radiological images representing an examination area of at least one examination object at a number N of consecutive points in time, at least after an application of a contrast agent, each radiological image comprising a plurality of image elements, each image element representing a sub-area of the examination area, - for each image element: extracting one or more features from the radiological images, - generating a time series for each sub-area of the examination area, each time series of a sub-area comprising the one or more extracted features of the image elements representing the sub-area for the number N of consecutive points in time, - selecting at least some of the time series, - grouping the selected time series into a number M of groups, - assigning each non-selected time series to one of the M groups, - assigning each image element to the group to which the time series of the sub-area that the image element represents is assigned, - assigning a label from M labels to each image element, each label of the M labels representing a group of the M groups, - outputting and/or storing at least one of the radiological images comprising the characteristics of the image elements of the radiological image and/or transmitting the at least one radiological image comprising the characteristics of the image elements of the radiological image to a separate computer system. A further subject of the present disclosure is a computer system comprising an input unit, a control and computing unit and an output unit, wherein the control and computing unit is configured to - cause the input unit to receive radiological images, wherein the radiological images represent an examination area of at least one examination object at a number N of consecutive points in time at least after an application of a contrast agent, wherein each radiological image comprises a plurality of image elements, wherein each image element represents a sub-area of the examination area, - extract one or more features from the radiological images for each image element, - generate a time series for each sub-area of the examination area, wherein each time series of a sub-area comprises the one or more extracted features of the image elements representing the sub-area for the number N of consecutive points in time, - select at least some of the time series, - group the selected time series into a number M of groups, - assign each non-selected time series to one of the M groups, - assign each image element to the group to which the time series of that sub-area is assigned. is that the image element represents, - to assign one of M identifiers to each image element, each identifier of the M identifiers representing a group of the M groups, - to cause the output unit to output and/or store and/or transmit to a separate computer system at least one of the radiological images comprising the identifiers of the image elements of the radiological image. A further subject of the present invention is a computer program product comprising a data storage device in which a computer program is stored that can be loaded into a working memory of a computer system and causes the computer system to carry out the following steps: - receiving radiological images, the radiological images representing an examination area of at least one examination object at a number N of consecutive points in time, at least after an application of a contrast agent, each radiological image comprising a plurality of image elements, each image element representing a sub-area of the examination area, - for each image element: extracting one or more features from the radiological images, - generating a time series for each sub-area of the examination area, each time series of a sub-area comprising the one or more extracted features of the image elements representing the sub-area for the number N of consecutive points in time, - selecting at least some of the time series, - grouping the selected time series into a number M of groups, - assigning each non-selected time series to one of the M groups, - assigning each image element to the group to which the time series of that is assigned to a sub-area that the image element represents, - assigning a label of M labels to each image element, each label of the M labels representing a group of the M groups, - outputting and/or storing at least one of the radiological images comprising the labels of the image elements of the radiological image and/or transmitting the at least one radiological image comprising the labels of the image elements of the radiological image to a separate computer system. A further subject of the present disclosure is a use of a contrast agent in a radiological examination method, the radiological examination method comprising: - receiving radiological images, the radiological images representing an examination area of at least one examination object at a number N of consecutive times, at least after an application of the contrast agent, each radiological image comprising a plurality of image elements, each image element representing a sub-area of the examination area, - for each image element: extracting one or more features from the radiological images, - generating a time series for each sub-area of the examination area, each time series of a sub-area comprising the one or more extracted features of the image elements representing the sub-area for the number N of consecutive points in time, - selecting at least a portion of the time series, - grouping the selected time series into a number M of groups, - assigning each non-selected time series to one of the M groups, - assigning each image element to the group to which the time series of the sub-area that the image element represents is assigned, - assigning an identifier of M identifiers to each image element, each identifier of the M identifiers representing a group of the M groups, - outputting and/or storing at least one of the radiological images comprising the identifiers of the image elements of the radiological image and/or transmitting the at least one radiological image comprising the identifiers of the image elements of the radiological image to a separate computer system. A further subject of the present disclosure is a contrast agent for use in a radiological examination method, wherein the radiological examination method comprises: - receiving radiological images, wherein the radiological images represent an examination area of at least one examination object at a number N of consecutive points in time, at least after an application of the contrast agent, wherein each radiological image comprises a plurality of image elements, wherein each image element represents a sub-area of the examination area, - for each image element: extracting one or more features from the radiological images, - generating a time series for each sub-area of the examination area, wherein each time series of a sub-area comprises the one or more extracted features of the image elements representing the sub-area for the number N of consecutive points in time, - selecting at least some of the time series, - grouping the selected time series into a number M of groups, - assigning each non-selected time series to one of the M groups, - assigning each image element to the group to which the time series of the sub-area that the image element represents is assigned, - assigning a label of M identifiers for each image element, each identifier of the M identifiers representing a group of the M groups, - outputting and/or storing at least one of the radiological images comprising the identifiers of the image elements of the radiological image and/or transmitting the at least one radiological image comprising the identifiers of the image elements of the radiological image to a separate computer system. A further subject of the present disclosure is a kit comprising - a contrast agent and - a computer program product comprising a data storage device in which a computer program is stored, which can be loaded into a working memory of a computer system. and causes the computer system to carry out the following steps: • Receiving radiological images, the radiological images representing an examination area of at least one examination object at a number N of consecutive points in time, at least after an application of the contrast agent, each radiological image comprising a plurality of image elements, each image element representing a sub-area of the examination area, • for each image element: extracting one or more features from the radiological images, • generating a time series for each sub-area of the examination area, each time series of a sub-area comprising the one or more extracted features of the image elements representing the sub-area for the number N of consecutive points in time, • selecting at least some of the time series, • grouping the selected time series into a number M of groups, • assigning each non-selected time series to one of the M groups, • assigning each image element to the group to which the time series of the sub-area that the image element represents is assigned, • assigning one of M identifiers to each image element, each identifier of the M Characteristics represents a group of the M groups, • outputting and/or storing at least one of the radiological images comprising the characteristics of the image elements of the radiological image and/or transmitting the at least one radiological image comprising the characteristics of the image elements of the radiological image to a separate computer system. Further objects are disclosed in the description below. The invention is explained in more detail below, without distinguishing between the objects of the invention (method, computer system, computer program product, use, contrast agent for use, kit). Rather, the following explanations should apply analogously to all objects of the invention, regardless of the context in which they occur (method, computer system, computer program product, use, contrast agent for use, kit). If steps are mentioned in a sequence in the present description or in the patent claims, this does not necessarily mean that the invention is limited to the sequence mentioned. Rather, it is conceivable that the steps are also carried out in a different sequence or in parallel to one another; unless a step builds on another step, which makes it absolutely necessary that the step that builds on it is carried out subsequently (which will become clear in individual cases). The above-mentioned sequences therefore represent preferred embodiments of the invention. The invention is explained in more detail in some places with reference to drawings. The drawings show specific embodiments with specific features and combinations of features, which primarily serve for illustration purposes; the invention should not be understood as being limited to the features and combinations of features shown in the drawings. Furthermore, statements made in the description of the drawings with reference to Features and combinations of features are generally applicable, i.e. can also be transferred to other embodiments and are not limited to the embodiments shown. The present disclosure describes means for segmenting radiological images and/or for identifying abnormalities in radiological images. A “radiological image” is a representation of an examination area of an examination object. The “examination object” is usually a living being, preferably a mammal, very particularly preferably a human. The “examination area” is a part of the examination object, for example an organ or part of an organ such as the liver, brain, heart, kidney, lung, stomach, intestine or part of the organs mentioned or several organs or another part of the body. In one embodiment, the examination area comprises a liver or part of a liver or the examination area is a liver or part of a liver of a mammal, preferably a human. In another embodiment, the examination area comprises a brain or part of a brain or the examination area is a brain or part of a brain of a mammal, preferably a human. In a further embodiment, the examination area comprises a heart or part of a heart or the examination area is a heart or part of a heart of a mammal, preferably a human. In a further embodiment, the examination area comprises a thorax or part of a thorax or the examination area is a thorax or part of a thorax of a mammal, preferably a human. In a further embodiment, the examination area comprises a stomach or part of a stomach or the examination area is a stomach or part of a stomach of a mammal, preferably a human. In a further embodiment, the examination area comprises a pancreas or part of a pancreas or the examination area is a pancreas or part of a pancreas of a mammal, preferably a human. In a further embodiment, the examination area comprises a kidney or part of a kidney or the examination area is a kidney or part of a kidney of a mammal, preferably a human. In a further embodiment, the examination area comprises one or both lungs or part of a lung of a mammal, preferably a human. In a further embodiment, the examination area comprises a breast or part of a breast, or the examination area is a breast or part of a breast of a female mammal, preferably a female human. In a further embodiment, the examination area comprises a prostate or part of a prostate, or the examination area is a prostate or part of a prostate of a male mammal, preferably a male human. The examination area, also called field of view (FOV), represents in particular a volume that is depicted in radiological images. The examination area is typically determined by a radiologist, for example on an overview image (localizer). Of course, the examination area can alternatively or additionally, automatically, for example on the basis of a selected protocol. A radiological image is preferably the result of a radiological examination. “Radiology” is the branch of medicine that deals with the use of predominantly electromagnetic radiation and (including ultrasound diagnostics, for example) mechanical waves for diagnostic, therapeutic and/or scientific purposes. In addition to X-rays, other ionizing radiation such as gamma radiation or electrons are also used. Since a key purpose is imaging, other imaging methods such as sonography and magnetic resonance imaging (MRI) are also considered radiology, although no ionizing radiation is used in these methods. The term “radiology” in the sense of the present disclosure therefore includes in particular the following examination methods: computer tomography, magnetic resonance imaging, sonography. In a preferred embodiment of the present invention, the radiological examination is a magnetic resonance imaging examination. In one embodiment, the radiological examination is an MRI examination in which an MRI contrast agent is used. In another embodiment, the radiological examination is a CT examination in which a CT contrast agent is used. In another embodiment, the radiological examination is a CT examination in which an MRI contrast agent is used. Magnetic resonance imaging, abbreviated to MRI or MRI, is an imaging method that is used primarily in medical diagnostics to depict the structure and function of tissues and organs in the human or animal body. In MR imaging, the magnetic moments of protons in an object under examination are aligned in a basic magnetic field so that a macroscopic magnetization is established along a longitudinal direction. This is then deflected from the rest position by the radiation of high-frequency (HF) pulses (excitation). The return of the excited states to the rest position (relaxation) or the magnetization dynamics is then detected as relaxation signals using one or more RF receiving coils. For spatial coding, rapidly switched magnetic gradient fields are superimposed on the basic magnetic field. The recorded relaxation signals or the detected MR data are initially available as raw data in the frequency space and can be transformed into the spatial space (image space) by subsequent inverse Fourier transformation. A radiological image in the sense of the present disclosure can be an MRI image, a computer tomogram, an ultrasound image or the like. A radiological image in the sense of the present disclosure is preferably a representation of an examination region of an examination object in the spatial space (image space). In a representation in spatial space, the examination area is usually represented by a large number of image elements (e.g. pixels, voxels, doxels, or n-xels, where n indicates the respective dimension), which can be arranged in a grid, for example, with each image element representing a sub-area of the examination area. Each image element is usually assigned a color value or gray value. A format widely used in radiology for storing and Processing representations in spatial space is the DICOM format. DICOM (Digital Imaging and Communications in Medicine) is an open standard for storing and exchanging information in medical image data management. The color value or gray value that can be assigned to each image element usually represents an intensity of a signal that is measured in the respective radiological imaging procedure. Different types of tissue in the examination area usually produce different signal intensities that are made visible in the radiological image. Contrast agents can be used to enhance contrasts between different types of tissue in radiological images. "Contrast agents" are substances or mixtures of substances that improve the representation of structures and functions of the body in radiological imaging procedures. Examples of contrast agents can be found in the literature (see e.g. A. S. L. Jascinth et al.: Contrast Agents in computed tomography: A Review, Journal of Applied Dental and Medical Sciences, 2016, Vol.2, Issue 2, 143 – 149; H. Lusic et al.: X-ray-Computed Tomography Contrast Agents, Chem. Rev. 2013, 113, 3, 1641-1666; https://www.radiology.wisc.edu/wp-content/uploads/2017/10/contrast- agents-tutorial.pdf, M. R. Nough et al.: Radiographic and magnetic resonances contrast agents: Essentials and tips for safe practices, World J Radiol.2017 Sep 28; 9(9): 339–349; L. C. Abonyi et al.: Intravascular Contrast Media in Radiography: Historical Development & Review of Risk Factors for Adverse Reactions, South American Journal of Clinical Research, 2016, Vol. 3, Issue 1, 1-10; ACR Manual on Contrast Media, 2020, ISBN: 978-1-55903-012-0; A. Ignee et al.: Ultrasound contrast agents, Endosc Ultrasound.2016 Nov-Dec; 5(6): 355–362). In computed tomography, iodine-containing solutions are usually used as contrast agents. In magnetic resonance imaging (MRI), superparamagnetic substances (e.g. iron oxide nanoparticles, superparamagnetic iron-platinum particles (SIPPs)) or paramagnetic substances (e.g. gadolinium chelates, manganese chelates, hafnium chelates) are usually used as contrast agents. In the case of sonography, fluids containing gas-filled microbubbles are usually administered intravenously. Examples of contrast agents can be found in the literature (see e.g. A. S. L. Jascinth et al.: Contrast Agents in computed tomography: A Review, Journal of Applied Dental and Medical Sciences, 2016, Vol. 2, Issue 2, 143 – 149; H. Lusic et al.: X-ray- Computed Tomography Contrast Agents, Chem. Rev. 2013, 113, 3, 1641-1666; https://www.radiology.wisc.edu/wp-content/uploads/2017/10/contrast-agents-tutorial.pdf, M. R. Nough et al.: Radiographic and magnetic resonances contrast agents: Essentials and tips for safe practices, World J Radiol. 2017 Sep 28; 9(9): 339–349; L. C. Abonyi et al.: Intravascular Contrast Media in Radiography: Historical Development & Review of Risk Factors for Adverse Reactions, South American Journal of Clinical Research, 2016, Vol.3, Issue 1, 1-10; ACR Manual on Contrast Media, 2020, ISBN: 978-1-55903-012-0; A. Ignee et al.: Ultrasound contrast agents, Endosc Ultrasound.2016 Nov-Dec; 5(6): 355–362). MRI contrast agents exert their effect in an MRI examination by changing the relaxation times of the structures that absorb contrast agents. Two groups of substances can be distinguished: para- and superparamagnetic substances. Both groups of substances have unpaired electrons that induce a magnetic field around the individual atoms or molecules. Superparamagnetic ones lead to a predominant T2 shortening, while paramagnetic contrast agents essentially lead to a T1 shortening. The effect of these contrast agents is indirect, since the contrast agent itself does not emit a signal, but only influences the signal intensity in its surroundings. An example of a superparamagnetic contrast agent is iron oxide nanoparticles (SPIO). Examples of paramagnetic contrast agents are gadolinium chelates such as gadopentetate dimeglumine (trade name: Magnevist^®etc.), gadoteric acid (Dotarem^®, Dotagi^®, Cyclolux^®), gadodiamide (Omniscan^®), gadoteridol (ProHance^®), Gadobutrol (Gadovist^®), gadopiclenol (Elucirem, Vueway) and gadoxetic acid (Primovist^®/Eovist^®). In one embodiment of the present disclosure, the contrast agent is an agent comprising gadolinium(III) 2-[4,7,10-tris(carboxymethyl)-1,4,7,10-tetrazacyclododec-1-yl]acetic acid (also referred to as gadolinium-DOTA or gadoteric acid). In another embodiment, the contrast agent is an agent comprising gadolinium(III) ethoxybenzyl-diethylenetriaminepentaacetic acid (Gd-EOB-DTPA); preferably, the contrast agent comprises the disodium salt of gadolinium(III)-ethoxybenzyl-diethylenetriaminepentaacetic acid (also referred to as gadoxetic acid). In one embodiment of the present disclosure, the contrast agent is an agent comprising gadolinium(III) 2-[3,9-bis[1-carboxylato-4-(2,3-dihydroxypropylamino)-4-oxobutyl]-3,6,9,15-tetrazabicyclo[9.3.1]pentadeca-1(15),11,13-trien-6-yl]-5-(2,3-dihydroxypropylamino)-5-oxopentanoate (also referred to as gadopiclenol) (see, e.g., WO2007/042504 and WO2020/030618 and/or WO2022/013454). In one embodiment of the present disclosure, the contrast agent is an agent comprising dihydrogen[(±)-4-carboxy-5,8,11-tris(carboxymethyl)-1-phenyl-2-oxa-5,8,11-triazatridecane-13-oato(5-)]gadolinate(2-) (also referred to as gadobenic acid). In one embodiment of the present disclosure, the contrast agent is an agent comprising tetragadolinium [4,10-bis(carboxylatomethyl)-7-{3,6,12,15-tetraoxo-16-[4,7,10-tris-(carboxylatomethyl)-1,4,7,10-tetraazacyclododecan-1-yl]-9,9-bis({[({2-[4,7,10-tris-(carboxylatomethyl)-1,4,7,10-tetraazacyclododecan-1-yl]propanoyl}amino)acetyl]-amino}methyl)- 4,7,11,14-tetraazahepta-decan-2-yl}-1,4,7,10-tetraazacyclododecan-1-yl]acetate (also referred to as gadoquatrane) (see, e.g., J. Lohrke et al.: Preclinical Profile of Gadoquatrane: A Novel Tetrameric, Macrocyclic High Relaxivity Gadolinium-Based Contrast Agent. Invest Radiol., 2022, 1, 57(10): 629-638; WO2016193190). In one embodiment of the present disclosure, the contrast agent is an agent that contains a Gd³⁺-Complex of a compound of formula (I)

(I) , wherein Ar is a group selected from

where _#represents the connection to X, X represents a group consisting of CH₂, (CH₂)₂, (CH₂)₃, (CH₂)₄and *-(CH₂)₂-O-CH₂-^#is selected, where * represents the connection to Ar and^#represents the link to the acetic acid residue, R¹, R²and R³independently a hydrogen atom or a group selected from C₁-C₃- Alkyl, -CH₂OH, -(CH₂)₂OH and -CH₂OCH₃represent, R⁴a group selected from C₂-C₄-Alkoxy, (H₃C-CH₂)-O-(CH₂)₂-O-, (H₃C-CH₂)-O-(CH₂)₂-O- (CH₂)₂-O- and (H₃C-CH₂)-O-(CH₂)₂-O-(CH₂)₂-O-(CH₂)₂-O- represents R⁵represents a hydrogen atom, and R⁶represents a hydrogen atom, or a stereoisomer, tautomer, hydrate, solvate or salt thereof, or a mixture thereof. In one embodiment of the present disclosure, the contrast agent is an agent comprising a Gd³⁺-Complex of a compound of formula (II)

(II) , wherein Ar is a group selected from R

where _#represents the link to X, X represents a group consisting of CH₂, (CH₂)₂, (CH₂)₃, (CH₂)₄and *-(CH₂)₂-O-CH₂-^#is selected, where * represents the connection to Ar and^#represents the link to the acetic acid residue, R⁷a hydrogen atom or a group selected from C₁-C₃-alkyl, -CH₂OH, -(CH2)2OH and -CH2OCH3; R^8tha group selected from C₂-C₄-Alkoxy, (H₃C-CH₂O)-(CH₂)₂-O-, (H₃C-CH₂O)-(CH₂)₂-O-(CH₂)₂-O- and (H₃C-CH₂O)-(CH₂)₂-O-(CH₂)₂-O-(CH₂)₂-O- represents; R⁹and R¹⁰independently represent a hydrogen atom; or a stereoisomer, tautomer, hydrate, solvate or salt thereof, or a mixture thereof. The term "C₁-C₃-Alkyl" means a linear or branched, saturated, monovalent hydrocarbon group having 1, 2 or 3 carbon atoms, e.g. methyl, ethyl, n-propyl and isopropyl. The term "C₂-C₄-Alkyl" means a linear or branched, saturated, monovalent hydrocarbon group having 2, 3 or 4 carbon atoms. The term "C₂-C₄-Alkoxy" means a linear or branched, saturated, monovalent group of the formula (C₂-C₄-alkyl)-O-, in which the term "C₂-C₄-Alkyl" is as defined above, e.g. a methoxy, ethoxy, n-propoxy or isopropoxy group. In one embodiment of the present disclosure, the contrast agent is an agent comprising gadolinium 2,2',2''-(10-{1-carboxy-2-[2-(4-ethoxyphenyl)ethoxy]ethyl}-1,4,7,10-tetraazacyclododecane-1,4,7-triyl)triacetate (see e.g. WO2022/194777, Example 1). In one embodiment of the present disclosure, the contrast agent is an agent comprising gadolinium 2,2',2''-{10-[1-carboxy-2-{4-[2-(2-ethoxyethoxy)ethoxy]phenyl}ethyl]- 1,4,7,10-tetraazacyclododecane-1,4,7-triyl}triacetate (see e.g. WO2022/194777, Example 2). In one embodiment of the present disclosure, the contrast agent is an agent comprising gadolinium 2,2',2''-{10-[(1R)-1-carboxy-2-{4-[2-(2-ethoxyethoxy)ethoxy]phenyl}ethyl]-1,4,7,10-tetraazacyclododecane-1,4,7-triyl}triacetate (see e.g. WO2022/194777, Example 4). In one embodiment of the present disclosure, the contrast agent is an agent comprising gadolinium (2S,2'S,2''S)-2,2',2''-{10-[(1S)-1-carboxy-4-{4-[2-(2-ethoxyethoxy)ethoxy]phenyl}butyl]-1,4,7,10-tetraazacyclododecane-1,4,7-triyl}tris(3-hydroxypropanoate) (see e.g. WO2022/194777, Example 15). In one embodiment of the present disclosure, the contrast agent is an agent comprising gadolinium 2,2',2''-{10-[(1S)-4-(4-butoxyphenyl)-1-carboxybutyl]-1,4,7,10-tetraazacyclododecane-1,4,7-triyl}triacetate (see e.g. WO2022/194777, Example 31). In one embodiment of the present disclosure, the contrast agent is an agent comprising gadolinium 2,2',2''-{(2S)-10-(carboxymethyl)-2-[4-(2-ethoxyethoxy)benzyl]-1,4,7,10-tetraazacyclododecane-1,4,7-triyl}triacetate. In one embodiment of the present disclosure, the contrast agent is an agent comprising gadolinium 2,2',2''-[10-(carboxymethyl)-2-(4-ethoxybenzyl)-1,4,7,10-tetraazacyclododecane-1,4,7-triyl]triacetate. In one embodiment of the present disclosure, the contrast agent is an agent comprising gadolinium(III) 5,8-bis(carboxylatomethyl)-2-[2-(methylamino)-2-oxoethyl]-10-oxo-2,5,8,11-tetraazadodecane-1-carboxylate hydrate (also referred to as gadodiamide). In one embodiment of the present disclosure, the contrast agent is an agent comprising gadolinium(III) 2-[4-(2-hydroxypropyl)-7,10-bis(2-oxido-2-oxoethyl)-1,4,7,10-tetrazacyclododec-1-yl]acetate (also referred to as gadoteridol). In one embodiment of the present disclosure, the contrast agent is an agent comprising gadolinium(III) 2,2',2''-(10-((2R,3S)-1,3,4-trihydroxybutan-2-yl)-1,4,7,10-tetraazacyclododecane-1,4,7-triyl)triacetate (also referred to as gadobutrol or Gd-DO3A-butrol). The techniques described in this disclosure can be used for automated segmentation of a radiological image (segmentation method) and/or for automated identification of abnormalities in a radiological image. The term "automated" means without human intervention. The term "segmentation" refers to the process of dividing a radiological image into several segments, which are also called image segments, image regions or image objects. Segmentation is usually performed to localize objects and boundaries (lines, curves, planes, surfaces, etc.) in radiological images. From a segmented radiological image, the localized objects can be separated from the background, visually highlighted (e.g.: colored), measured, counted or otherwise quantified. During segmentation, each image element (pixel/voxel/doxel/n-xel) of a radiological image is assigned an identifier so that image elements with the same identifier have certain features in common. In a first step of the segmentation process, radiological images are received. The term “receiving” includes both the retrieval of radiological images and the receipt of radiological images that are transmitted, for example, to the computer system of the present disclosure. The radiological images can be received from a computer tomograph, a magnetic resonance tomograph or an ultrasound scanner. The radiological images can be read from one or more data storage devices and/or transmitted from a separate computer system. The radiological images represent an examination area of one or more examination objects. The examination area is usually the same for all examination objects (e.g. the liver). The radiological images represent the examination area at different times, at least after the application of a contrast agent. In one embodiment of the present disclosure, the radiological images represent the examination area at different times before and after the application of a contrast agent. In other words, the radiological images form a sequence in which the dynamic behavior of the contrast agent in the examination area is shown. The times form a series of consecutive times: t₁, t₂, …, t_NThe term “consecutive” means that there is a first point in time t₁and a second time t₂where the second point in time is later than the first point in time t₁There may be further points in time, for example a third point in time t₃. The third point in time is then later than the first point in time t₁and the second time t₂. There may be further points in time. Between successive points in time there may be a period of time in which no point in time occurs. In other words: the term "successive" does not mean that the points in time immediately follow one another, i.e. it does not mean that there may be no period of time between the points in time. Preferably, at least one point in time occurs before an application of a contrast agent, while preferably at least two points in time occur after the application of the contrast agent. In other words: the radiological images preferably comprise at least one radiological image that represents an examination area of an examination object before the application of a contrast agent (i.e. without contrast agent) and at least two radiological images that represent the examination area of the examination object after the application of the contrast agent. It is possible for a contrast agent to be applied multiple times, i.e. for there to be more than one application of a contrast agent. It is possible for several contrast agents to be applied. The several contrast agents can be applied simultaneously and/or one after the other. The number N of points in time t₁ are_Nis preferably at least 3. The number N of points in time is particularly preferably in the range of 5 to 20, but is not limited to 20 points in time. The points in time can have a constant distance or a varying distance from one another. If radiological images of the examination area are received from several (different) examination objects, the points in time at which they were generated are preferably the same. If there is therefore a first sequence of radiological images of a Examination area of a first examination object, which defines the examination area of the first examination object at the times t₁= -20 seconds, t₂= 0, t₃= 60 seconds and t₄= 120 seconds, the radiological images of a second sequence also present represent the examination area of a second examination object preferably also at the times t₁= -20 seconds, t₂= 0, t₃= 60 seconds and t₄= 120 seconds. However, it is not necessary for the times to be the same. Each radiological image comprises a large number of image elements (pixels/voxels/doxels, n-xels). Depending on the resolution, a two-dimensional radiological image can, for example, have a number of 512 pixels x 512 pixels = 262144 pixels. However, the number of image elements can also be larger or smaller. Each image element represents a sub-area of the examination area. It is possible to co-register the various radiological images. The “co-registration” (also called “image registration” in the state of the art) serves to bring two or more radiological images of the same examination area into optimal agreement with one another. One of the radiological images is defined as the reference image, and the other radiological images are called object images. In order to optimally adapt the object images to the reference image, a compensating transformation is calculated. It is possible to use an optimization process to identify the transformation that best adapts an object image to the reference image, so that image elements of the object image represent the same sub-areas of the examination area as the corresponding ones (i.e. image elements with the same coordinates) of the reference image. Once the transformation has been found for an object image, the transformation can be transferred to the other object images. However, it is also conceivable that each object image is individually matched to the reference image. Methods for co-registration are described in the prior art (see, for example: B. Zitová et al: Image Registration Methods: A Survey, Image and Vision Computing, 2003, 21, 977-1000). It is possible that further/other pre-processing steps are carried out on the radiological images, such as filtering, changing the resolution, color space transformation and/or others. For each image element, one feature or several features are extracted from the radiological images. An extracted feature can be a radiomics feature, such as a signal intensity and/or a gradient (e.g. after executing a Sobel and/or Kirsch gradient operator). The term “radiomics” refers to the quantitative extraction of, for example, histogram and/or texture features from radiological images (see, for example, M. E. Mayerhoefer et al: Introduction to Radiomics, The Journal of Nuclear Medicine, 2020, Vol.61, No.4, pages 488-495; C. Haarburger et al.: Radiomics Feature Reproducibility Under Inter-Rater Variability in Segmentations of CT Images, Sci Rep 10, 12688 (2020), https://doi.org/10.1038/s41598-020-69534-6; V. Parekh et al.: Radiomics: a new application from established techniques, Expert Rev Precis Med Drug Dev. 2016, 1(2): 207–226). An extracted feature can, for example, be an intensity value of a signal (signal intensity). In the case of an X-ray image, the intensity value can, for example, be the intensity of X-rays after they have passed through the examination area. In the case of an MRI image, the intensity value can, for example, be the intensity of a T1-weighted signal or a T2-weighted signal or another intensity value of an MRI measurement sequence. The intensity value can be, for example, the gray value (in the case of a gray value representation) or the color value (in the case of a color representation) of the respective image element of the respective radiological image. For each sub-area of the examination area, a time series is generated on the basis of the received radiological images. Each time series comprises the characteristics of those image elements that represent the examination area for the number N of consecutive points in time. In other words: each time series is formed for a sub-area on the basis of the image elements that represent the sub-area in the radiological images; each image element represents the sub-area at a different point in time; the characteristics of the image elements as a function of the consecutive points in time form a time series. Usually, each time series has exactly one value of a feature (e.g. intensity value of a signal) for each point in time of the consecutive points in time. A time series can, for example, have the following form: X = (x₁, x₂, …, x_N), where X represents the time series, x₁the value of a feature of an image element at time t₁, x₂the value of the feature of the image element at time t2 and xN the value of the feature of the image element at time t_N. Preferably, a temporal standardization is carried out. For example, the time of application of a contrast agent can be defined as the zero point (t = 0). All other times can be expressed in relation to this zero point. A time "30 seconds before the application of a contrast agent" can, for example, be expressed as "t = -30 seconds" and a time "3 minutes and 10 seconds after the application of the contrast agent" as "t = 190 seconds". The time of application of a contrast agent is preferably the time at which the application is complete, i.e. the intended amount of contrast agent, for example in the form of a bolus, has been completely introduced into a part of the body of the object being examined (e.g. into the arm vein in a person). Analogously, the time at which a radiological image is created is the time at which the creation is complete. If radiological images of different examination objects were not created at the same relative times or if time points are missing in a time series, the missing values in the time series can be supplemented, for example, by interpolation and/or extrapolation. It is also conceivable that the temporal progression of the signal intensities in a dynamic contrast-enhanced radiological examination for different examination objects varies due to different anatomy and/or physiology. It is therefore possible that the time series of different examination objects are standardized with regard to the intensity values and the time points. An example of standardization is described further down in the description. It is also conceivable to take further steps to create the values for the time series, e.g. calculating the relative or absolute signal intensity increase from the contrast-enhanced radiological images and the native image, and/or taking further steps that are common in the processing of medical image data, such as co-registering the images and/or changing the image resolution, calculating an image gradient and/or calculating local statistical values (radiomics features). The time series values created in this way can then be standardized, as is usual in time series analysis, so that they all have the same variance and the same mean, for example. At least part of the time series is selected in order to group the selected time series. The term "clustering" refers to methods for identifying similarity structures in data and grouping similar objects into groups, which are also called clusters. Such methods are also known as cluster analysis. The aim of grouping is to identify new groups in the data (in contrast to classification, in which data is assigned to existing classes). The selected part can be, for example, 1%, 5%, 10%, 15%, 20%, 30%, 50%, 80%, 90% or another percentage. The selected part can also be 100%, i.e. all time series can be selected and grouped. However, since grouping often takes place in an iterative process (see the example described below) and is therefore comparatively computationally intensive, a part that is smaller than 100% is preferably selected. Following the cluster analysis, the remaining, non-selected time series can be assigned to the groups formed in a classification process that is less computationally intensive. The selected part should therefore be representative of the set of available time series, i.e. all sub-areas that show a characteristic temporal behavior after application of a contrast agent should be represented. The selection can be made randomly. However, it is also possible to specifically select time series from defined sub-areas of the examination area. For example, it is possible for a radiologist to specify sub-areas in one or more radiological images (for example using an input device such as a computer mouse) whose time series should be included in the grouping, for example because they show different types of tissue in one or more radiological images and/or the radiologist knows from experience that the sub-areas show a characteristic signal curve in the dynamic contrast-enhanced radiological examination. The selected time series are grouped into a number M of groups. During grouping, each time series is assigned to exactly one of the M groups. There are a variety of methods for time-series clustering (see e.g.: E. Özkoç, Esma (2020), Clustering of Time-Series Data, 10.5772/intechopen.84490; A. Sardá-Espinosa: Comparing Time-Series Clustering Algorithms in R Using the dtwclust Package, The R Journal Vol. 11/01, June 2019 ISSN 2073-4859; S. Aghabozorgi et al.: Time-series clustering - A decade review, Information Systems, 2015, 53, 16-38). Time series clustering is usually based on determining a similarity measure or a distance measure. A similarity measure usually indicates how similar two time series are to each other. The larger the similarity measure, the greater the similarity is usually. A distance measure usually indicates how different two time series are from each other. The larger the distance measure, the more the time series usually differ from each other. Distance measures can often be converted into similarity measures and vice versa. If you look at two time series, a first time series X = (x₁, x₂, …, x_N) and a second time series Y = (y₁, y₂, …, y_N), then a similarity measure s(X,Y), which quantifies the similarity between the two time series, can be calculated from a distance measure d(X,Y), which quantifies the difference between the two time series, for example according to the following equation: ^{( ) 1} ^{^^^^ X, Y =} 1_{+ ^^^^(X, Y)}If the two time series X and Y are identical, the distance measure is 0 and the similarity measure is calculated according to the above equation as s(X,Y) = 1. An example of a distance measure between two time series X = (x1, x2, ..., xN) and Y = (y1, y2, ..., yN) is the Manhattan distance:

Another example of a distance measure between two time series X = (x₁, x₂, …, x_N) and Y = (y₁, y₂, …, yN) is the Euclidean distance:

Another example of a distance measure between two time series X = (x₁, x₂, …, x_N) and Y = (y₁, y₂, …, y_N) is the Minkowksi distance of order p:

where p is an integer. For p = 1, the Minkowksi distance corresponds to the Manhattan distance; for p = 2, the Minkowksi distance corresponds to the Euclidean distance. Another example of a distance measure between two time series X = (x1, x2, …, xN) and Y = (y1, y2, …, y_N) is dynamic time warping (described in H. Sakoe, S. Chiba, “Dynamic programming algorithm optimization for spoken word recognition,” IEEE Transactions on Acoustics, Speech and Signal Processing, vol. 26(1), pp. 43–49, 1978. and https://tslearn.readthedocs.io/en/stable/user_guide/dtw.html). It is defined as an optimization problem: ^^^^( ^^^^, ^^^^) = m ^^i^^n� ^

where π is a path that aligns the time series such that the Euclidean distance between aligned (i.e. resampled) time series is minimal. Other examples of distance measures are Short Time-Series Distance (STS) and DISSIM distance. Instead of calculating the difference or similarity of time series based on intensity values, features (for example in the form of feature vectors) can be extracted from the time series and the similarity/distance between them calculated. Examples of similarity measures/distance measures based on extracted features are Pearson’s correlation coefficient and periodogram-based distance. It is also possible to subject the time series (and/or radiological images) to one or more transformations before determining similarity measures or distance measures for the transformed time series. Examples of such transformations are Fourier transform and wavelet transform. The distance measures and similarity measures described here, as well as others that can be used, are described in numerous publications on the topic of time-series clustering (see e.g.: E. Özkoç, Esma (2020), Clustering of Time-Series Data, 10.5772/intechopen.84490; A. Sardá-Espinosa: Comparing Time-Series Clustering Algorithms in R Using the dtwclust Package, The R Journal Vol.11/01, June 2019 ISSN 2073-4859; S. Aghabozorgi et al.: Time-series clustering - A decade review, Information Systems, 2015, 53, 16-38). If a similarity measure or a distance measure is defined for the time series in question, the time series are grouped based on the similarity measure or distance measure. Numerous methods also exist for this. An example is described below for illustration purposes. The time series X and Y described above can be understood as vectors in an N-dimensional space. Each time series defines a point in the N-dimensional space through its vector (starting from the origin of the N-dimensional space). Usually there are not just two but a large number of such vectors/points in the N-dimensional space (as many as time series have been selected). A center of gravity can be calculated for a group (a cluster) of points. The center of gravity can, for example, be the point in the N-dimensional space that is defined by a vector whose elements are formed by the arithmetically averaged mean values of the elements of all vectors belonging to the cluster. In other words: all time series j that belong to a cluster are each represented by a vector X_j= (x_j1, x_j2, …, x_jn) represents the center of gravity of the center of gravity of the center of gravity of the center of gravity._s= (x_s1, x_s2, …, x_sn) can be calculated; the calculation can be done by element-wise arithmetic averaging, i.e., the first element x_s1can be calculated by the arithmetic mean of the first elements of the vectors X_jformed, the second element x_s2by the arithmetic mean of the second elements of the vectors X_jand so on. The point specified by the centroid vector starting from the zero point of the N-dimensional space is the centroid of the group (cluster). A centroid of a group can be considered as the center of the group (group center/cluster center). An example of an iterative algorithm for grouping the time series is: 1. A number M of cluster centers are randomly created in the N-dimensional space. 2. Each time series is assigned to the cluster center from which it has a minimum distance d. 3. For each cluster, the cluster center is recalculated as the centroid of the time series forming the cluster. 4. Steps 2 and 3 are repeated as often as the assignment of the time series to the clusters changes. The term "minimal distance" means that there is no cluster center for a time series that is closer to the time series than the cluster center that is the closest to it. The number M of groups into which the time series are grouped can be determined by a radiologist. Preferably, the number M of groups corresponds to the number of sub-areas of the examination area that show different/characteristic dynamic behavior after application of a contrast agent. This is explained in more detail using the example of a dynamic contrast-enhanced magnetic resonance imaging examination of a human liver using a hepatobiliary contrast agent. A hepatobiliary contrast agent is characterized by the fact that it is specifically absorbed by liver cells, the hepatocytes, accumulates in the functional tissue (parenchyma) and increases the contrast in healthy liver tissue. An example of a hepatobiliary contrast agent is the disodium salt of gadoxetic acid (Gd-EOB-DTPA disodium), which is described in US Patent No.6,039,931A under the brand name Primovist^®and Eovist^®is commercially available. Other hepatobiliary contrast agents are described in WO2022/194777, among others. Another MRI contrast agent with lower uptake into the hepatocytes is gadobenate dimeglumine (Multihance®). After the intravenous administration of a hepatobiliary contrast agent in the form of a bolus into a vein in a person's arm, the contrast agent initially reaches the liver via the arteries. These are shown with contrast enhancement in the corresponding MRI images. The phase in which the liver arteries are shown with contrast enhancement in MRI images is called the "arterial phase". The contrast agent then reaches the liver via the liver veins. While the contrast in the liver arteries is already decreasing, the contrast in the liver veins reaches a maximum. The phase in which the liver veins are shown with contrast enhancement in MRI images is called the "portal venous phase". The portal venous phase is followed by the "transitional phase," in which the contrast in the liver arteries continues to decrease, and the contrast in the liver veins also decreases. When using a hepatobiliary contrast agent, the contrast in the healthy liver cells gradually increases in the transitional phase. 10-20 minutes after its injection, a hepatobiliary contrast agent leads to a significant signal enhancement in the healthy liver parenchyma. This phase is referred to as the "hepatobiliary phase." The contrast agent is only slowly excreted from the liver cells; accordingly, the hepatobiliary phase can last two hours or more. The phases mentioned are described in more detail in the following publications, for example: J. Magn. Reson. Imaging, 2012, 35(3): 492–511, doi:10.1002/jmri.22833; Clujul Medical, 2015, Vol.88 no.4: 438-448, DOI: 10.15386/cjmed-414; Journal of Hepatology, 2019, Vol. 71: 534–542, http://dx.doi.org/10.1016/j.jhep.2019.05.005). The phases mentioned are shown schematically in Fig. 1. Fig. 1 shows the course of the signal intensities in a magnetic resonance imaging examination for three different sub-areas of an examination area that includes the liver of a person as a function of time t. The examination subject was administered a hepatobiliary contrast agent at time t = 0. The T1-weighted signal intensities are plotted on the ordinate as intensity values I. The graph marked with A shows the temporal course of the signal intensity in a hepatic artery. The graph marked with The graph marked V shows the temporal progression of the signal intensity in a liver vein. The graph marked L shows the temporal progression of the signal intensity in healthy liver cells. All three signal curves are characteristic and can be easily distinguished from one another. The signal curves are similar in all healthy people. In order to group signal curves (time series) from different examination objects (people), the signal curves of the different examination objects can be normalized. The time of application of the contrast agent can be set to time 0 for all examination objects. The characteristic signal curve in the liver artery of an examination object can be used as a reference. The signal curves in the liver artery of the other examination objects can be adapted to the reference signal curve by means of a transformation (e.g. stretching, compression) so that they are aligned as best as possible. The transformation can then be applied to all other signal curves (time series) of all examination objects. If the radiological images include the liver and the gallbladder, an accumulation of contrast medium in the gallbladder can often be observed. Depending on which areas of the human body are captured by the radiological images, a number of 3 to 5 groups can be determined. If only the liver is captured, the arteries, veins and healthy liver cells differ in their dynamic appearance during the contrast-enhanced MRI examination (see Fig. 1) and a grouping into three groups can be made. If the gallbladder is captured in addition to the liver, this differs from the sub-areas already mentioned; a grouping into four groups can be made. If, in addition to the liver and gallbladder, sub-areas are captured to which the contrast medium does not reach or only reaches a comparatively small amount, a grouping into five groups can be made. Alternatively, mathematical heuristics can also be used to determine the number M of groups, for example the elbow method or the silhouette coefficient method. If the selected time series are grouped, M groups have been formed and each selected time series is assigned to one of the M groups. In a further step, the remaining, non-selected time series are assigned to the groups already formed. Each non-selected time series is assigned to exactly one group. This process is also known as classification. Classification can be carried out, for example, by calculating the distance (the distance measure) to the group center for each non-selected time series. Each non-selected time series is assigned to the group from whose group center it has a minimum distance. If all time series are assigned to a group of the M groups, the group membership can be transferred to the corresponding image elements that belong to the time series. Each time series has a number N of image elements. These N image elements can be assigned to the group to which the time series is assigned. If all image elements are assigned, the image elements can be labeled according to their group membership. Each image element is assigned an identifier, whereby the identifier represents the group to which the image element is assigned. This means that there are M different identifiers and each image element is assigned one of the M identifiers. All image elements of a radiological image with the same identifier can form a segment in the radiological image. In the case of the example shown in Fig.1, for example, all image elements representing hepatic arteries can be assigned to a first group, all image elements representing hepatic veins, be assigned to a second group and all image elements that represent healthy liver cells are assigned to a third group. The identifier that is/will be assigned to each group can be a number, for example. The number can define a color tone in which the respective image elements (the segment) should be displayed. For example, for a number of M = 3 groups, three identifiers can be defined, e.g. the identifiers 0, 1 and 2. Each image element is assigned one of the identifiers 0, 1 or 2. The identifier "0" can mean, for example, that the image element in a radiological image should be displayed in a red color tone. The identifier "1" can mean, for example, that the image element in a radiological image should be displayed in a green color tone. The identifier "2" can mean, for example, that the image element in a radiological image should be displayed in a blue color tone. If the radiological images are in a format where hues are defined according to the RGB color space with 256 hues (RGB stands for the primary colors red, green and blue), the labels can also indicate the corresponding hues: (255, 0, 0), (0, 255, 0), (0, 0, 255). However, these examples are for illustration purposes only. There are many other ways to define labels. It is also conceivable that the color saturation, hue or transparency of the label changes as a function of the distance measure, so that the label becomes more noticeable to the human eye the greater the deviation of the underlying time series from the cluster centroid. In a further step, the radiological images comprising the characteristics for the image elements can be output (e.g. displayed on a screen or printed out on a printer), stored (e.g. on a data storage of the computer system of the present disclosure or a data storage on a cloud server) or transmitted to a separate computer system (e.g. via a wired and/or wireless network). In one embodiment, the user (e.g. a radiologist) can select the number M of groups. For example, he can specify how many groups the time series should be divided into by entering data into the computer system of the present disclosure, or, as explained above, use a mathematical heuristic to help determine the number M. The user can, for example, try out different numbers of groups and have the segmented radiological images output (e.g. displayed on a screen) to check which number M produces a better division of the radiological images into segments. Once the groups have been formed, sequences of radiological images that were not included in the grouping can also be segmented. For example, a sequence of radiological images of a different (new) examination object can be segmented. A sequence of an examination object from which radiological images have been included in a grouping (i.e. have been used for the grouping) can also be segmented, wherein the sequence was recorded in a different time period (for example earlier or later) than the radiological images that were included in the grouping. The process for segmenting new radiological images can, for example, comprise: • Receiving a new sequence of radiological images, wherein the radiological images of the new sequence represent an examination region of an examination object at a number N of consecutive points in time at least after an application of a contrast agent, wherein each radiological image comprises a plurality of image elements, wherein each image element represents a sub-area of the examination region, • for each image element: extracting one or more features from the radiological images of the new sequence of radiological images, - generating a time series for each sub-area of the examination area based on the radiological images of the new sequence, each time series of a sub-area comprising the one or more extracted features of the image elements representing the sub-area for the number N of consecutive points in time, • assigning each time series to one of the M groups, • assigning the image elements of each time series to the group to which the time series is assigned, • assigning one of the M characteristics to each image element, each characteristic of the M characteristics representing a group of the M groups, • outputting and/or storing the new sequence of radiological images comprising the characteristics and/or transmitting the new sequence of radiological images comprising the characteristics to a separate computer system. The solutions described in this description can, as already described, also be used to identify abnormalities. On the one hand, groups can be formed on the basis of healthy examination objects and a group assignment can be checked on the basis of diseased examination objects. If one or more sub-areas cannot be clearly assigned to an existing group during group assignment (classification), this is an abnormality that may indicate an illness. On the other hand, changes in radiological images of an examination object can be detected. In a first step, groups are formed on the basis of radiological images of an examination object, whereby the radiological images represent an examination area of the examination object in a first time period. Radiological images of the examination area of the same examination object obtained in a later second time period can be assigned to the groups formed in a second step. If one or more sub-areas cannot be clearly assigned to an existing group during group assignment (classification), this is an abnormality that may indicate a change in the health status of the examination object (e.g. a deterioration or an improvement). It is also possible to include abnormalities in radiological images in the grouping in order to check for one or more other examination objects whether such abnormalities also occur in the one or more other examination objects. In a first step, a grouping is carried out based on radiological images that show one or more abnormalities. One or more groups represent sub-areas that show one or more abnormalities. In a second step, a group assignment is carried out based on radiological images of another object under examination. If, in this group assignment, sub-areas are assigned to a group that represents an abnormality, the abnormality is also present in the other object under examination. The identification of abnormalities according to the three examples mentioned above is explained in more detail below. In a (first) embodiment, radiological images are received from at least one healthy object under examination. "Healthy" means that no signs of illness can be seen in the radiological images. The radiological images represent an examination area of the at least one healthy examination object at a number N of consecutive points in time at least after an application of a contrast agent. Each radiological image comprises a large number of image elements. Each image element represents a sub-area of the examination area. For each image element of at least some of the sub-areas, one or more features are extracted from the radiological images. For at least some of the sub-areas of the examination area, time series are generated on the basis of the radiological images. Each time series comprises values of the one or more features for the number N of consecutive points in time. The time series are grouped in a cluster analysis into a number M of groups, as described in this description. This means that M groups are available that can be used for classification. In a further step, radiological images are received from another examination object. For the other examination object, it is to be examined whether the radiological images contain any abnormalities that indicate, for example, an illness. The radiological images preferably represent the examination area of the further examination object at the number N of consecutive points in time at least after the application of a contrast agent. The contrast agent is preferably the same as in the case of the at least one healthy examination object. Each radiological image comprises a plurality of image elements. Each image element represents a sub-area of the examination area. For each image element, one or more features are extracted from the radiological images. For each sub-area of the examination area, time series are generated on the basis of the radiological images. Each time series comprises values of the one or more features for the number N of consecutive points in time. If radiological images are available that represent the examination area (at least partially) at other points in time, time series that show values of the one or more features for the number N of consecutive points in time can be generated by interpolation or extrapolation. The time series are assigned to the M groups present in a classification process. If, during the assignment, the minimum distance of a time series to a group center is greater than a predefined threshold, this is an indication that the corresponding sub-area shows a different dynamic behavior. It is an indication that the time series does not fit into the group. Since the time series has an even greater distance to the other groups, this is an indication that the time series does not fit into any group. Time series that cannot be assigned to a group, e.g. because the minimum distance to a group center is greater than a predefined threshold and/or because the distances to all group centers are greater than a predefined threshold, can be provided with a label indicating that the time series cannot be assigned. The threshold can be set, for example, by a radiologist. It is also conceivable that the threshold is set automatically by the computer system of the present disclosure. is set. It is conceivable, for example, that the threshold value is set based on statistical calculations. It is conceivable, for example, that for a group from whose group center a time series has a minimum distance, a standard deviation of the distances of all time series from the group center is calculated and the threshold value is a multiple (e.g. double or triple or another multiple) of the standard deviation. In radiological images there can always be so-called outliers, i.e. image elements that have an intensity value that differs significantly from the intensity values of neighboring image elements. Such outliers can, for example, be the result of disturbances in the generation of the radiological images. If an individual time series cannot be clearly assigned to a group, it is possible that the time series includes one or more outliers. Therefore, in one embodiment, time series are only marked as unassignable if there is a minimum number of neighboring image elements that cannot be assigned. This minimum number can be, for example, 10 or 20 or 100 or more or less. Individual outliers can, for example, be assigned to the group to which the majority of the time series from neighboring sub-areas are assigned. If time series are assigned to the M groups, the image elements of the time series can also be assigned to the M groups and marked with one of the M identifiers that represent the groups. In the case of time series that are marked as unassignable, the image elements of the time series can also be marked as unassignable. The radiological images comprising the identifiers can be output and/or stored and/or transmitted to a separate computer system. If the radiological images comprising the identifiers are output (e.g. displayed on a screen or output on a printer), the image elements marked as unassignable can be displayed with a separate color tone, preferably with a color tone that is easily recognizable to humans in the context of the other colors of the radiological images. A user (e.g. radiologist) can then immediately see which image elements could not be assigned and which sub-areas therefore show a different dynamic behavior during the contrast-enhanced radiological examination. The user can inspect the sub-areas and look at the dynamic behavior of the contrast agent in these sub-areas by comparing the radiological images that represent the examination area at successive points in time in order to draw conclusions about the causes of the changed behavior. In this way, for example, lesions in the liver can be identified. In this way, for example, malignant tumors can be distinguished from benign lesions in the liver. Fig. 2 shows an example and schematic of the signal curve of various abnormalities in radiological images. In addition to the signal curves of the various abnormalities, Fig. 1 shows the signal curve A in the hepatic artery from Fig. 1 as a reference. As in Fig. 1, the T1w signal intensities are plotted as intensity values on the ordinate. The times are plotted on the abscissa. The signal curve marked with (a) shows the signal curve of a focal nodular hyperplasia (FNH). The signal curve marked with (b) shows the signal curve of a hepatocellular adenoma. The signal curve marked with (c) shows the signal curve of a liver hemangioma. The signal curve marked with (d) shows the signal curve of a hepatocellular carcinoma or of hypervascular metastases. The signal curve marked with (e) shows the signal curve of hypovascular metastases. It can be seen that each abnormality (a) to (e) has a characteristic signal curve. The signal curves of the abnormalities also differ from the signal curves of liver arteries, liver veins and healthy liver cells. The abnormalities can therefore be identified in a method as described here. In a (second) embodiment, radiological images are received from an examination object. The radiological images represent an examination area of the examination object at a number N of consecutive points in time at least after an application of a contrast agent. Each radiological image comprises a large number of image elements. Each image element represents a sub-area of the examination area. For each image element of at least some of the sub-areas, one or more features are extracted from the radiological images. Time series are generated for at least some of the sub-areas of the examination area based on the radiological images. Each time series comprises values of the one or more features for the number N of consecutive points in time. At least some of the time series are selected and the selected time series are grouped into a number M of groups as described in this description. This results in M groups that can be used for classification. At a later point in time (e.g. a few days or weeks or months or years later), the examination subject is again subjected to a contrast-enhanced radiological examination. Further radiological images of the examination area of the examination subject are generated. The radiological images represent the examination area at the number N of consecutive points in time at least after the application of a contrast agent. The contrast agent is preferably the same as in the previous examination. The generated radiological images are received by the computer system of the present disclosure. Each radiological image comprises a plurality of image elements. Each image element represents a sub-area of the examination area. For each image element, one or more features are extracted from the radiological images. For each sub-area of the examination area, a time series is generated based on the radiological images. Each time series contains values of one or more characteristics for the number N of consecutive points in time. The time series are assigned to the M groups in a classification process. If the minimum distance of a time series to a group focus is greater than a predefined threshold value during the assignment, this is an indication that the corresponding sub-area shows a different dynamic behavior. It is an indication that the time series does not fit into the group. Since the time series is even further away from the other groups, this is an indication that the time series does not fit into any group. Time series that cannot be assigned to a group, e.g. because the minimum distance to a group center is greater than a predefined threshold and/or because the distances to all group centers are greater than a predefined threshold, are provided with a label indicating that the time series are not assignable. The threshold can be set, for example, by a radiologist. It is also conceivable that the threshold is set automatically by the computer system of the present disclosure. If an individual time series cannot be clearly assigned to a group, the time series may include one or more outliers. Therefore, in one embodiment, time series are only marked as not assignable if there is a minimum number of neighboring image elements that are not assignable. If time series are assigned to the M groups, the image elements of the time series can also be assigned to the M groups and marked with one of the M labels representing the groups. In the case of time series that are marked as not assignable, the image elements of the time series can also be marked as not assignable. The radiological images comprising the characteristics can be output and/or stored and/or transmitted to a separate computer system. If the radiological images comprising the characteristics are output (e.g. displayed on a screen or output on a printer), the image elements marked as unassignable can be displayed with a separate color tone, preferably with a color tone that is easily recognizable to humans in the context of the other colors of the radiological images. A user (e.g. radiologist) can then immediately recognize which image elements were unassignable and which sub-areas therefore show a different dynamic behavior in the new contrast-enhanced radiological examination than in the previous radiological examination. The user can inspect the sub-areas and look at the dynamic behavior of the contrast agent in these sub-areas by comparing the radiological images that represent the examination area at successive points in time in order to draw conclusions about the causes of the changed behavior. In this way, for example, lesions in the liver can be identified. In this way, for example, malignant tumors can be distinguished from benign lesions in the liver. In a (third) embodiment, radiological images of at least one examination object are received that show at least one abnormality. Such an abnormality can be a sign of a disease, e.g. one or more lesions and/or one or more tumors. Such an abnormality affects one or more sub-areas of the examination area. Such an abnormality is characterized by the fact that the one or more sub-areas show a characteristic behavior after the application of a contrast agent. In other words: time series of this one or more sub-areas differ from time series of other sub-areas of the examination area. The radiological images represent the examination area at a number N of consecutive points in time at least after an application of a contrast agent. Each radiological image comprises a plurality of image elements. Each image element represents a sub-area of the examination area. For each image element of at least some of the sub-areas, one or more features are extracted from the radiological images. Time series are generated for at least some of the sub-areas of the examination area based on the radiological images. Each time series includes values of one or more features for the number N of consecutive points in time. The time series are grouped in a cluster analysis into a number M of groups, as described in this description. The number of groups is chosen so that time series from sub-areas that have one or more abnormalities fall into one group. After grouping, there are M groups that can be used for classification. In a further step, radiological images are received from another examination object. For the other examination object, it is to be examined whether the radiological images also contain one or more abnormalities that, for example, indicate a disease. The radiological images represent the examination area of the other examination object at the number N of consecutive points in time at least after the application of a contrast agent. Each radiological image includes a large number of image elements. Each image element represents a sub-area of the examination area of the other examination object. For each image element, one or more features are extracted from the radiological images. For each sub-area of the examination area, a time series is generated based on the radiological images. Each time series includes values of one or more features for the number N of consecutive points in time. The time series are assigned to the M groups in a classification process. If, during this classification, one or more time series are assigned to the group that represent time series of sub-areas with one or more abnormalities, then such abnormalities are also present in the radiological images of the other object under examination. If the time series are assigned to the M groups, the image elements of the time series can also be assigned to the M groups and marked with one of the M characteristics that represent the groups. The radiological images comprising the characteristics can be output and/or saved and/or transmitted to a separate computer system. If the radiological images are output including the characteristics (e.g. displayed on a screen or output on a printer), the image elements that fall into the group of abnormalities can be displayed with their own color tone, preferably with a color tone that is easily recognizable to humans in the context of the other colors of the radiological images. A user (e.g. radiologist) can then immediately recognize which image elements also show one or more abnormalities. The abnormalities that are included in the grouping can be, for example, focal nodular hyperplasias, hepatocellular adenomas, liver hemangiomas, hepatocellular carcinomas, hypervascular metastases and/or hypovascular metastases, because, as Fig. 2 shows, these show a characteristic signal curve that can generate their own group when grouped. Fig. 3 shows an example and schematically an embodiment of the computer-implemented method of the present disclosure in the form of a flow chart. The method (100) comprises the following steps: (110) receiving radiological images, the radiological images representing an examination area of at least one examination object at a number N of consecutive points in time at least after an application of a contrast agent, each radiological image comprising a plurality of image elements, each image element representing a sub-area of the examination area, (120) for each image element: extracting one or more features from the radiological images, (130) generating a time series for each sub-area of the examination area, each time series of a sub-area comprising the one or more extracted features of the image elements representing the sub-area for the number N of consecutive points in time, (140) selecting at least some of the time series, (150) grouping the selected time series into a number M of groups, (160) assigning each non-selected time series to one of the M groups, (170) assigning the image elements of each time series to the group to which the time series is assigned, (180) Assigning a label from M labels to each image element, each label of the M labels representing a group of the M groups, (190) Outputting and/or storing the radiological images comprising the labels and/or transmitting the radiological images comprising the labels to a separate computer system. Fig. 4 shows by way of example and schematically the generation of time series and the grouping of time series. In a first step, three radiological images are received, a first radiological image I₁, a second radiological image I₂and a third radiological image I₃. In the radiological images shown in Fig. 4 I₁, I₂and I₃are two-dimensional raster graphics. Each of the radiological images I1, I2 and I3 represents the examination area of an object under examination. The radiological images I₁, I₂and I₃represent the examination area of the object under examination at different times. The first radiological image I₁represents the investigation area of the object under investigation at a first time t₁. The second radiological image I₂represents the investigation area of the object under investigation at a second time t₂The third radiological image I3 represents the examination area of the object under examination at a third time t₃. The three points in time follow one another: time t₂follows the time t₁and time t3 follows time t2: t1

Each of the radiological images I1, I2 and I3 comprises a large number of image elements. In the example shown in Fig. 4, each of the radiological images comprises 10x12=120 image elements arranged in a grid. Usually, the number of image elements is greater than 120. In the example shown in Fig. 4, each of the radiological images I₁, I₂and I₃three image elements are highlighted and provided with reference symbols. The first radiological image I1 comprises the three image elements IE₁₁, IE₁₂and IE₁₃(and other image elements). The second radiological image I₂includes the three image elements IE₂₁, IE₂₂and IE₂₃(and other image elements). The third radiological image I₃includes the three image elements IE₃₁, IE₃₂and IE₃₃(and other image elements). Each image element of each radiological image represents a sub-area of the examination area of the object under examination. The image elements IE₁₁, IE₂₁and IE₃₁correspond to each other, i.e. they represent the same part of the examination area of the object under investigation. The image elements IE12, IE22 and IE32 also correspond to each other; they also represent the same part of the examination area of the object under investigation; however, this part of the area differs from the part of the image elements IE₁₁, IE₂₁and IE₃₁represent. The image elements IE also correspond₁₃, IE₂₃and IE₃₃with each other; they also represent the same part of the examination area of the object under investigation; this part, however, differs from the part of the image elements IE₁₁, IE₂₁and IE₃₁and from the part of the image elements IE12, IE22 and IE32 represent. The image elements IE₁₁, IE₂₁and IE₃₁represent the subarea T₁. The image elements IE₁₂, IE₂₂and IE₃₂represent the subarea T₂. The image elements IE₁₃, IE₂₃and IE₃₃represent the subarea T₃. The image element IE₁₁represents the subarea T₁at time t₁. The image element IE₂₁represents the sub-area T1 at time t2. The image element IE31 represents the sub-area T1 at time t₃. The image element IE₁₂represents the subarea T₂at time t₁. The image element IE₂₂represents the subarea T₂at time t₂. The image element IE₃₂represents the subarea T₂at time t₃. The image element IE₁₃represents the subarea T₂at time t₁. The image element IE₂₃represents the subarea T₂at time t₂. The image element IE33 represents the partial area T2 at time t3. In a further step, one or more features are extracted from the radiological images for each image element. In the example shown in Fig.4, the extracted feature is a hatching with which the respective image element is represented. The image element IE₁₁is shown with a hatching, which in the example shown in Fig.4 is referred to as feature M2. For the image element IE11, the feature M2 from the radiological image I₁extracted. The image element IE₂₁is shown with a hatching, which in the example shown in Fig.4 is referred to as feature M₃For the image element IE₂₁The feature M₃from radiological image I₂extracted. The image element IE₃₁is shown with a hatching, which in the example shown in Fig.4 is referred to as feature M₁For the image element IE₃₁The feature M₁extracted from the radiological image I3. The image element IE₁₂is shown with a hatching, which in the example shown in Fig.4 is referred to as feature M₂For the image element IE₁₂The feature M₂from radiological image I₁extracted. The image element IE₂₂is shown with a hatching, which in the example shown in Fig.4 is referred to as feature M3. For the image element IE22, the feature M3 from the radiological image I₂extracted. The image element IE₃₂is shown with a hatching, which in the example shown in Fig.4 is referred to as feature M₁For the image element IE₃₂The feature M₁from radiological image I₃extracted. The image element IE13 is shown with a hatching, which in the example shown in Fig.4 is referred to as feature M₁For the image element IE₁₃The feature M₁from radiological image I₁extracted. The image element IE₂₃is shown with a hatching, which in the example shown in Fig.4 is referred to as feature M₁For the image element IE₂₃The feature M₁from radiological image I₂extracted. The image element IE33 is shown with a hatching, which in the example shown in Fig.4 is represented as feature M₁For the image element IE₃₃The feature M₁from radiological image I₃extracted. In a further step, a time series is generated for each sub-area of the examination area. Each time series contains the extracted feature for the number of consecutive points in time. The time series are shown graphically in Fig.4. The time series XT1 for the sub-area T1 contains the feature M2 of the image element IE11 at time t₁, the feature M₃of the image element IE₂₁at time t₂and the feature M₁of the image element IE₃₁at time t₃. The time series X_T2for sub-area T₂includes the feature M₂of the image element IE₁₂at time t₁, the feature M₃of the image element IE₂₂at time t₂and the feature M₁of the image element IE₃₂at time t₃. The time series X_T3for sub-area T₃includes the feature M₁of the image element IE₁₃at time t1, the feature M1 of the image element IE23 at time t2 and the feature M1 of the image element IE33 at time t₃. For sub-area T₁the time series can also be formulated as follows: X_T1= (M₂, M₃, M₁). For sub-area T₂the time series can also be formulated as follows: X_T2= (M₂, M₃, M₁). For the sub-area T3, the time series can also be formulated as follows: XT3 = (M1, M1, M1). The time series are grouped in a further step. In the example shown in Fig.4, the time series for the sub-area T₁and sub-area T₂identical; they are assigned to the same group C₁In the example shown in Fig.4, the time series for the sub-area T₃from the time series of sub-area T₁and sub-area T₂; it will be assigned to Group C₂In a further step, all image elements of all radiological images are assigned to the group to which the time series of the sub-area that the image elements represent is assigned. This is not shown in Fig.4. The image elements IE₁₁, IE₂₁and IE₃₁represent the subarea T₁. The time series X_T1of sub-area T₁was assigned to Group C₁assigned; this also includes the image elements IE₁₁, IE₂₁and IE₃₁Group C₁The image elements IE₁₂, IE₂₂and IE₃₂represent the subarea T₂. The time series X_T2of the sub-area T2 was assigned to group C1; this also includes the image elements IE12, IE22 and IE₃₂Group C₁The image elements IE₁₃, IE₂₃and IE₃₃represent the subarea T₃. The time series X_T3of sub-area T₃was assigned to Group C₂assigned; this also includes the image elements IE₁₃, IE₂₃and IE₃₃Group C₂assigned. In a next step, each image element is assigned a label, with each label representing the group to which the respective image element was assigned. In other words: image elements that were assigned to the same group receive the same label; image elements that were assigned to different groups usually (but not necessarily) receive different labels. This is not shown in Fig.4. The image elements IE₁₁, IE₂₁, IE₃₁, IE₁₂, IE₂₂and IE₃₂receive the same identifier because they have been assigned to the same group C1. The sub-areas T represented by the image elements IE11, IE21, IE31, IE12, IE22 and IE32₁and T₂show a comparable (in the case of the example shown in Fig.4 identical) dynamic (temporal) behavior. They can be identified as belonging together in a radiological image, for example. The image elements IE₁₃, IE₂₃and IE₃₃also receive the same identifier because they were assigned to the same group C2. The sub-area T3 represented by the image elements IE13, IE23 and IE33 shows a different dynamic (temporal) behavior than the sub-areas T₁and T₂. The sub-area T₃can be shown differently in a radiological image than the subareas T₁and T₂. In a further step, the radiological images can be output including the characteristics (e.g. displayed on a monitor or printed out with a printer), stored on a data storage device and/or transmitted to a separate computer system (e.g. via a network). For example, all image elements to which the same characteristic has been assigned can be displayed in the same way (e.g. in the same color and/or brightness) on a monitor and/or printed out on a printer so that a radiologist can recognize which sub-areas show comparable (e.g. identical) dynamic behavior and which sub-areas differ in their dynamic behavior. Further embodiments of the present disclosure are: 1. A computer-implemented method for segmenting a plurality of radiological images, comprising: • receiving the plurality of radiological images, the radiological images representing an examination area of an examination subject at a number N of consecutive points in time at least after an application of a contrast agent, each radiological image comprising a plurality of image elements, each image element representing a partial area of the examination area, • for each image element: o for each point in time of the N consecutive points in time: extracting one or more features from the radiological images, o generating a time series based on the one or more extracted features, • assigning each time series to one of M groups, the assigning comprising: o selecting at least a portion of the time series, o grouping the selected time series into a number of M groups, o assigning each non-selected time series to one of the M groups, • assigning each image element to that group of the M groups to which the time series of the image element is assigned, • assigning a label of M identifiers for each image element, each identifier of the M identifiers representing a group of the M groups, • Outputting and/or storing at least one of the radiological images comprising the identifiers of the image elements of the radiological image and/or transmitting the at least one radiological image comprising the identifiers of the image elements of the radiological image to a separate computer system. 2. The method according to embodiment 1, wherein a portion of the radiological images represents the examination area of the examination object before application of the contrast agent and a portion of the radiological images represents the examination area of the examination object after application of the contrast agent. 3. The method according to embodiment 1 or 2, wherein the examination object is a mammal, preferably a human, and the examination area is a part of the examination object, preferably the liver, the brain, the heart, the kidney, the lung, the stomach, the intestine or a part of the organs mentioned or several organs or another part of the body of the examination object. 4. The method according to one of embodiments 1 to 3, wherein the examination object is a human and the examination area is or includes the liver of the human. 5. The method according to one of embodiments 1 to 4, wherein the radiological images are MRI images. 6. The method according to one of embodiments 1 to 5, wherein the contrast agent is a hepatobiliary contrast agent. 7. The method according to one of embodiments 1 to 6, wherein the number N of consecutive points in time is in the range from 5 to 20. 8. The method according to any one of embodiments 1 to 7, wherein the one or more features is a signal intensity, the signal intensity preferably being represented by a shade of gray or color. 9. The method according to any one of embodiments 1 to 8, wherein the step of grouping the selected time series into a number M of groups comprises: (1) randomly forming a number M of group centers, (2) assigning each time series to the group center to which it has a minimum distance, (3) recalculating the group center for each group, (4) repeating steps (2) and (3) as often as the assignment of the time series to the groups changes. 10. The method according to any one of embodiments 1 to 9, wherein the step of assigning each non-selected time series to one of the M groups comprises: - calculating a distance of the time series to the centers of the groups, - assigning the time series to the group to whose center it has the minimum distance. 11. The method according to one of embodiments 1 to 10, further comprising: - receiving the number M of groups from a user. 12. A computer-implemented method for identifying abnormalities in radiological images, comprising: • receiving and/or generating radiological images, the radiological images representing an examination area of at least one healthy examination subject at a number N of consecutive points in time at least after an application of the contrast agent, each radiological image comprising a plurality of image elements, each image element representing a partial area of the examination area, • for each image element: o for each point in time of the N consecutive points in time: extracting one or more features from the radiological images, o generating a time series based on the one or more extracted features, • Assigning each time series to one of M groups, the assignment comprising: o selecting at least a portion of the time series, o grouping the selected time series into a number of M groups, • receiving and/or generating further radiological images, the further radiological images representing the examination area of a further examination object at the number N of consecutive points in time at least after the application of the contrast agent, each further radiological image comprising a plurality of image elements, each image element representing a partial area of the examination area, • for each image element of the further radiological images: o for each point in time of the N consecutive points in time: extracting one or more features from the radiological images, o generating a time series based on the one or more extracted features, • assigning each time series to one of the M groups in a classification process, time series of a minimum number of neighboring image elements that have a distance to a group center of gravity that is greater than a predefined threshold value are marked as not assignable, • outputting the non-assignable assignable image elements and/or a message about the non-assignable image elements. 13. The method according to embodiment 12, wherein the radiological images and the further radiological images represent the examination area before and after the application of the contrast agent. 14. The method according to embodiment 12 or 13, wherein the examination object is a mammal, preferably a human, and the examination area is a part of the examination object, preferably the liver, the brain, the heart, the kidney, the lung, the stomach, the intestine or a part of the organs mentioned or several organs or another part of the body of the examination object. 15. The method according to one of embodiments 12 to 14, wherein the examination object is a human and the examination area is or comprises the human's liver. 16. The method according to one of embodiments 12 to 15, wherein the radiological images and the further radiological images are MRI images. 17. The method according to any one of embodiments 12 to 16, wherein the contrast agent is a hepatobiliary contrast agent. 18. The method according to any one of embodiments 12 to 17, wherein the number N of consecutive time points is in the range of 5 to 20. 19. The method according to any one of embodiments 12 to 18, wherein the one or more features is a signal intensity, wherein the signal intensity is preferably represented by a shade of gray or color. 20. The method according to any one of embodiments 12 to 19, wherein the step of grouping the selected time series into a number M of groups comprises: (1) randomly forming a number M of group centers, (2) assigning each time series to the group center to which it has a minimum distance, (3) Recalculate the group center for each group, (4) Repeat steps (2) and (3) as often as the assignment of the time series to the groups changes. 21. A computer-implemented method for identifying abnormalities in radiological images, comprising: • receiving and/or generating radiological images, the radiological images representing an examination area of a healthy examination subject at a number N of consecutive points in time at least after an application of the contrast agent, each radiological image comprising a plurality of image elements, each image element representing a partial area of the examination area, • for each image element: o for each point in time of the N consecutive points in time: extracting one or more features from the radiological images, o generating a time series based on the one or more extracted features, • assigning each time series to one of M groups, the assigning comprising: o selecting at least a portion of the time series, o grouping the selected time series into a number of M groups, • receiving and/or generating further radiological images, the further radiological images representing the examination area of the examination subject at a number N of later consecutive points in time at least after the application of the contrast agent, each further radiological image comprises a plurality of image elements, wherein each image element represents a partial region of the examination region, • for each image element of the further radiological images: o for each point in time of the N consecutive points in time: extracting one or more features from the radiological images, o generating a time series based on the one or more extracted features, • assigning each time series to one of the M groups in a classification process, wherein time series of a minimum number of adjacent image elements that have a distance to a group center of gravity that is greater than a predefined threshold are marked as unassignable, • outputting the unassignable image elements and/or a message about the unassignable image elements. 22. The method according to embodiment 21, wherein the radiological images and the further radiological images represent the examination region before and after the application of the contrast agent. 23. The method according to embodiment 21 or 22, wherein the examination object is a mammal, preferably a human, and the examination area is a part of the examination object, preferably the liver, the brain, the heart, the kidney, the lung, the stomach, the intestine or a part of the organs mentioned or several organs or another part of the body of the examination object. 24. The method according to one of embodiments 21 to 23, wherein the examination object is a human and the examination area is or includes the liver of the human. 25. The method according to one of embodiments 21 to 25, wherein the radiological images and the further radiological images are MRI images. 26. The method according to one of embodiments 21 to 25, wherein the contrast agent is a hepatobiliary contrast agent. 27. The method according to one of embodiments 21 to 26, wherein the number N of consecutive points in time is in the range from 5 to 20. 28. The method according to one of embodiments 21 to 27, wherein the one or more features is a signal intensity, wherein the signal intensity is preferably represented by a shade of gray or color. 29. The method according to any one of embodiments 21 to 28, wherein the step of grouping the selected time series into a number M of groups comprises: (1) randomly forming a number M of group centers, (2) assigning each time series to the group center to which it has a minimum distance, (3) recalculating the group center for each group, (4) repeating steps (2) and (3) as often as the assignment of the time series to the groups changes. 30. A computer-implemented method for identifying abnormalities in radiological images, comprising: • receiving and/or generating radiological images, the radiological images representing an examination area of at least one examination object at a number N of consecutive points in time at least after an application of the contrast agent, each radiological image comprising a plurality of image elements, each image element representing a partial area of the examination area, the examination area of the at least one examination object having an abnormality that can be attributed to a disease of the at least one examination object, • for each image element: o for each point in time of the N consecutive points in time: extracting one or more features from the radiological images, o generating a time series based on the one or more extracted features, • assigning each time series to one of M groups, the assigning comprising: o selecting at least a portion of the time series, o grouping the selected time series into a number of M groups, at least one group having time series of such image elements that cover at least a partial area of the examination area with the Represent abnormality, • Receiving and/or generating further radiological images, wherein the further radiological images represent the examination area of a further examination object at the number N of consecutive points in time at least after the application of the contrast agent, wherein each further radiological image comprises a plurality of image elements, wherein each image element represents a partial area of the examination area, • for each image element of the further radiological images: o for each point in time of the N consecutive points in time: extracting one or more features from the radiological images, o generating a time series based on the one or more extracted features, • assigning each time series to one of the M groups in a classification process, • in the event that a minimum number of time series of the at least one group of time series of image elements have been assigned to the abnormality: issuing a message that the abnormality is present in the examination area of the further examination object. 31. The method according to embodiment 30, wherein the radiological images and the further radiological images represent the examination area before and after the application of the contrast agent. 32. The method according to embodiment 30 or 31, wherein the examination object is a mammal, preferably a human, and the examination area is a part of the examination object, preferably the liver, the brain, the heart, the kidney, the lung, the stomach, the intestine or a part of the organs mentioned or several organs or another part of the body of the examination object. 33. The method according to one of embodiments 30 to 32, wherein the examination subject is a human and the examination region is or comprises the liver of the human. 34. The method according to one of embodiments 30 to 33, wherein the radiological images and the further radiological images are MRI images. 35. The method according to one of embodiments 30 to 34, wherein the contrast agent is a hepatobiliary contrast agent. 36. The method according to one of embodiments 30 to 35, wherein the number N of consecutive points in time is in the range from 5 to 20. 37. The method according to one of embodiments 30 to 36, wherein the one or more features is a signal intensity, wherein the signal intensity is preferably represented by a shade of gray or color. 38. The method according to any one of embodiments 30 to 37, wherein the step of grouping the selected time series into a number M of groups comprises: (1) randomly forming a number M of group centers, (2) assigning each time series to the group center to which it has a minimum distance, (3) recalculating the group center for each group, (4) repeating steps (2) and (3) as often as the assignment of the time series to the groups changes. 39. Computer system comprising means for carrying out a method according to any one of embodiments 1 to 38. 40. Computer program product comprising a data store in which a computer program is stored that can be loaded into a working memory of a computer system and causes the computer system to carry out a method according to any one of embodiments 1 to 38. The methods described in this disclosure can be carried out in whole or in part by a computer system. A "computer system" is a system for electronic data processing that processes data using programmable calculation rules. Such a system usually comprises a "computer", the unit that includes a processor for carrying out logical operations, and a peripheral. In computer technology, "peripherals" refer to all devices that are connected to the computer and are used to control the computer and/or as input and output devices. Examples of these are monitors (screens), printers, scanners, mice, keyboards, drives, cameras, microphones, loudspeakers, etc. Internal connections and expansion cards are also considered peripherals in computer technology. Fig. 5 shows an example and schematically a computer system according to the present disclosure. The computer system (1) shown in Fig. 5 comprises an input unit (10), a control and computing unit (20) and an output unit (30). The control and computing unit (20) is used to control the computer system (1), coordinate the data flows between the units of the computer system (1) and carry out calculations. The control and computing unit (20) is configured to • cause the input unit (10) to receive radiological images, the radiological images representing an examination area of at least one examination object at a number N of consecutive points in time at least after an application of a contrast agent, each radiological image comprising a plurality of image elements, each image element representing a sub-area of the examination area, • generate a time series for each sub-area of the examination area based on the radiological images, each time series comprising intensity values for the number N of consecutive points in time, • select at least some of the time series, • group the selected time series into a number M of groups, • assign each non-selected time series to one of the M groups, • assign the image elements of each time series to the group to which the time series is assigned, • assign one of M identifiers to each image element, each identifier of the M identifiers representing a group of the M groups, • output the radiological images comprising the identifiers, store them and/or transmit them to a separate computer system. Fig. 6 shows, by way of example and schematically, another embodiment of the computer system according to the invention. The computer system (1) comprises a processing unit (21) which is connected to a memory (22). The processing unit (21) and the memory (22) form a control and calculation unit as shown in Fig. 5. The processing unit (21) can comprise one or more processors alone or in combination with one or more memories. The processing unit (21) can be conventional computer hardware which is capable of processing information such as digital images, computer programs and/or other digital information. The processing unit (21) usually consists of an arrangement of electronic circuits, some of which are implemented as an integrated circuit or as several interconnected integrated circuits. circuits (an integrated circuit is sometimes also referred to as a "chip"). The processing unit (21) may be configured to execute computer programs that may be stored in a working memory of the processing unit (21) or in the memory (22) of the same or another computer system. The memory (22) may be ordinary computer hardware capable of storing information such as digital images (e.g. representations of the examination area), data, computer programs and/or other digital information either temporarily and/or permanently. The memory (22) may comprise volatile and/or non-volatile memory and may be permanently installed or removable. Examples of suitable memories are RAM (Random Access Memory), ROM (Read-Only Memory), a hard disk, flash memory, a removable computer diskette, an optical disc, a magnetic tape or a combination of the above. The optical discs may include read-only compact discs (CD-ROM), read/write compact discs (CD-R/W), DVDs, Blu-ray discs, and the like. In addition to the memory (22), the processing unit (21) may also be connected to one or more interfaces (11, 12, 31, 32, 33) to display, transmit, and/or receive information. The interfaces may include one or more communication interfaces (32, 33) and/or one or more user interfaces (11, 12, 31). The one or more communication interfaces may be configured to send and/or receive information, e.g., to and/or from an MRI scanner, a CT scanner, an ultrasound camera, other computer systems, networks, data storage, or the like. The one or more communication interfaces may be configured to transmit and/or receive information via physical (wired) and/or wireless communication links. The one or more communication interfaces may include one or more interfaces for connecting to a network, e.g., using technologies such as cellular, Wi-Fi, satellite, cable, DSL, fiber optic, and/or the like. In some examples, the one or more communication interfaces may include one or more short-range communication interfaces configured to connect devices using short-range communication technologies such as NFC, RFID, Bluetooth, Bluetooth LE, ZigBee, infrared (e.g., IrDA), or the like. The user interfaces may include a display (31). A display (31) may be configured to display information to a user. Suitable examples include a liquid crystal display (LCD), a light-emitting diode display (LED), a plasma display panel (PDP), or the like. The user input interface(s) (11, 12) may be wired or wireless and may be configured to receive information from a user into the computer system (1), e.g. for processing, storage and/or display. Suitable examples of user input interfaces are a microphone, an image or video capture device (e.g. a camera), a keyboard or keypad, a joystick, a touch-sensitive surface (separate from or integrated with a touchscreen), or the like. In some examples, the user interfaces may include automatic identification and data capture (AIDC) technology for machine-readable information. This may include barcodes, radio frequency identification (RFID), magnetic stripes, optical character recognition (OCR), integrated circuit cards (ICC), and the like. The user interfaces may further include one or more interfaces for communicating with peripheral devices such as printers and the like. One or more computer programs (40) can be stored in the memory (22) and executed by the processing unit (21), which is thereby programmed to perform the functions described in this description. The retrieval, loading and execution of instructions of the computer program (40) can be carried out sequentially, so that one command is retrieved, loaded and executed at a time. However, the retrieval, loading and/or execution can also be carried out in parallel. The system according to the invention can be designed as a laptop, notebook, netbook and/or tablet PC, it can also be a component of an MRI scanner, a CT scanner or an ultrasound diagnostic device. The present invention also relates to a computer program product. Such a computer program product comprises a non-volatile data carrier such as a CD, a DVD, a USB stick or another medium for storing data. A computer program is stored on the data carrier. The computer program can be loaded into a working memory of a computer system (in particular into a working memory of a computer system of the present disclosure) and cause the computer system to carry out the following steps: • receiving radiological images, the radiological images representing an examination area of at least one examination object at a number N of consecutive points in time at least after an application of a contrast agent, each radiological image comprising a plurality of image elements, each image element representing a sub-area of the examination area, • for each image element: extracting one or more features from the radiological images, • generating a time series for each sub-area of the examination area, each time series of a sub-area comprising the one or more extracted features of the image elements representing the sub-area for the number N of consecutive points in time, • selecting at least some of the time series, • grouping the selected time series into a number M of groups, • assigning each non-selected time series to one of the M groups, • assigning the image elements of each time series to the group to which the time series is assigned, • assigning a label of M identifiers for each image element, each identifier of the M identifiers representing a group of the M groups, • outputting and/or storing the radiological images comprising the identifiers and/or transmitting the radiological images comprising the identifiers to a separate computer system. The computer program product can also be marketed in combination (in a compilation) with the contrast agent. Such a compilation is also referred to as a kit. Such a kit comprises the contrast agent and the computer program product. It is also possible for such a kit to comprise the contrast agent and means that allow a buyer to obtain the computer program, e.g. download it from an internet page. These means can comprise a link, i.e. an address of the internet page from which the computer program can be obtained, e.g. from which the computer program can be downloaded to a computer system connected to the internet. These means can comprise a code (e.g. an alphanumeric character string or a QR code, or a DataMatrix code or a barcode or another optically and/or electronically readable code) with which the buyer gains access to the computer program. Such a link and/or code can, for example, be printed on a packaging of the contrast agent and/or on an information leaflet for the contrast agent. A kit is therefore a combination product comprising a contrast agent and a computer program (e.g. in the form of access to the computer program or in the form of executable program code on a data carrier) that is offered for sale together.

Claims

Patent claims 1. Computer-implemented method comprising: - receiving radiological images, the radiological images representing an examination area of at least one examination object at a number N of consecutive points in time, at least after an application of a contrast agent, each radiological image comprising a plurality of image elements, each image element representing a sub-area of the examination area, - for each image element: extracting one or more features from the radiological images, - generating a time series for each sub-area of the examination area, each time series of a sub-area comprising the one or more extracted features of the image elements representing the sub-area for the number N of consecutive points in time, - selecting at least a portion of the time series, - grouping the selected time series into a number M of groups, - assigning each non-selected time series to one of the M groups, - assigning each image element to the group to which the time series of the sub-area that the image element represents is assigned, - assigning a label from M labels to each image element, each label of the M labels representing a group of M groups, - outputting and/or storing at least one of the radiological images comprising the characteristics of the image elements of the radiological image and/or transmitting the at least one radiological image comprising the characteristics of the image elements of the radiological image to a separate computer system. 2. The method according to claim 1, wherein a portion of the radiological images represents the examination region of the at least one examination object before the application of the contrast agent and a portion of the radiological images represents the examination region of the at least one examination object after the application of the contrast agent. 3. The method according to claim 1 or 2, wherein the at least one examination object is a mammal, preferably a human, and the examination region is a part of the examination object, preferably the liver, the brain, the heart, the kidney, the lung, the stomach, the intestine or a portion of the aforementioned organs or several organs or another part of the body of the examination object. 4. The method according to one of claims 1 to 3, wherein the radiological images are computer tomography images, MRI images or ultrasound images. 5. The method according to one of claims 1 to 4, wherein the number N of consecutive points in time is in the range of 5 to 20. 6. The method according to one of claims 1 to 5, wherein the one or more features is a signal intensity, wherein the signal intensity is preferably represented by a shade of gray or color. 7. The method according to one of claims 1 to 6, wherein the step of grouping the selected time series into a number M of groups comprises: (1) randomly forming a number M of group centers, (2) assigning each time series to the group center to which it has a minimum distance, (3) recalculating the group center for each group, (4) repeating steps (2) and (3) as often as the assignment of the time series to the groups changes. 8. The method according to one of claims 1 to 7, wherein the step of assigning each non-selected time series to one of the M groups comprises: - calculating a distance of the time series to the centers of the groups, - assigning the time series to the group to whose center it has the minimum distance. 9. The method according to one of claims 1 to 8, further comprising: - receiving the number M of groups from a user. 10. The method according to one of claims 1 to 9, further comprising: - receiving new radiological images, the new radiological images representing the examination area of the at least one examination object or another examination object at successive points in time at least after an application of the contrast agent, each new radiological image comprising a plurality of image elements, each image element representing a partial area of the examination area, - for each image element of the new radiological images: extracting the one or more features from the new radiological images, - generating a time series for each partial area of the examination area based on the new radiological images, each time series of a partial area comprising the one or more extracted features of the image elements representing the partial area for the number N of successive points in time, - assigning each time series of the new radiological images to one of the M groups, - assigning the image elements of each time series of the new radiological images to the group to which the time series is assigned, - assigning one of the M identifiers to each image element of the new radiological images, each identifier the M identifiers represent a group of the M groups, - outputting and/or storing the new radiological images comprising the identifiers and/or transmitting the new radiological images comprising the identifiers to a separate computer system. 11. The method according to one of claims 1 to 10, further comprising: - in the event that the minimum distance of a time series to a center of a group is greater than a predefined threshold value: assigning an identifier to the time series and the image elements of the time series, the identifier indicating that the time series cannot be assigned. 12. Computer-implemented method for segmenting radiological images, comprising: - receiving radiological images, the radiological images representing an examination region of at least one examination object at a number N of consecutive points in time, at least after an application of a contrast agent, each radiological image comprising a plurality of image elements, each image element representing a partial region of the examination region, - for each image element: extracting one or more features from the radiological images, - generating a time series for each sub-area of the examination area, each time series of a sub-area comprising the one or more extracted features of the image elements representing the sub-area for the number N of consecutive points in time, - selecting at least some of the time series, - grouping the selected time series into a number M of groups, - assigning each non-selected time series to one of the M groups, - assigning each image element to the group to which the time series of the sub-area represented by the image element is assigned, - assigning an identifier from M identifiers to each image element, each identifier of the M identifiers representing a group of the M groups, - outputting and/or storing at least one of the radiological images comprising the identifiers of the image elements of the radiological image and/or transmitting the at least one radiological image comprising the identifiers of the image elements of the radiological image to a separate computer system. 13. Computer-implemented method for identifying abnormalities in radiological images, comprising: - receiving radiological images, the radiological images representing an examination area of at least one examination object at a number N of consecutive points in time, at least after an application of a contrast agent, each radiological image comprising a plurality of image elements, each image element representing a sub-area of the examination area, - for each image element: extracting one or more features from the radiological images, - generating a time series for each sub-area of the examination area, each time series of a sub-area comprising the one or more extracted features of the image elements representing the sub-area for the number N of consecutive points in time, - selecting at least some of the time series, - grouping the selected time series into a number M of groups, - assigning each non-selected time series to one of the M groups, - assigning each image element to the group to which the time series of the sub-area that the image element represents is assigned, - assigning one of M identifiers to each Image element, each identifier of the M identifiers representing a group of the M groups, - outputting and/or storing at least one of the radiological images comprising the identifiers of the image elements of the radiological image and/or transmitting the at least one radiological image comprising the identifiers of the image elements of the radiological image to a separate computer system. 14. Computer system comprising an input unit, a control and computing unit and an output unit, the control and computing unit being configured, - to cause the input unit to receive radiological images, the radiological images representing an examination area of at least one examination object at a number N of consecutive points in time at least after an application of a contrast agent, each radiological image comprising a plurality of image elements, each image element representing a sub-area of the examination area, - to extract one or more features from the radiological images of the examination area for each image element, - to generate a time series for each sub-area of the examination area, each time series of a sub-area comprising the one or more extracted features of the image elements representing the sub-area for the number N of consecutive points in time, - to select at least some of the time series, - to group the selected time series into a number M of groups, - to assign each non-selected time series to one of the M groups, - to assign the image elements of each time series to the group to which the time series is assigned, - to assign one of M identifiers to each image element, each identifier of the M identifiers representing a group of the M groups, - to cause the output unit, to output and/or store at least one of the radiological images comprising the characteristics of the image elements of the radiological image and/or to transmit them to a separate computer system. 15. Computer program product comprising a data storage device in which a computer program is stored that can be loaded into a working memory of a computer system and causes the computer system to carry out the following steps: - receiving radiological images, the radiological images representing an examination area of at least one examination object at a number N of consecutive points in time, at least after an application of a contrast agent, each radiological image comprising a plurality of image elements, each image element representing a sub-area of the examination area, - for each image element: extracting one or more features from the radiological images, - generating a time series for each sub-area of the examination area, each time series of a sub-area comprising the one or more extracted features of the image elements representing the sub-area for the number N of consecutive points in time, - selecting at least some of the time series, - grouping the selected time series into a number M of groups, - assigning each non-selected time series to one of the M groups, - assigning each image element to the group to which the time series of that sub-area is assigned. which the image element represents, - assigning an identifier of M identifiers to each image element, each identifier of the M identifiers representing a group of the M groups, - outputting and/or storing at least one of the radiological images comprising the identifiers of the image elements of the radiological image and/or transmitting the at least one radiological image comprising the characteristics of the image elements of the radiological image to a separate computer system. 16. Use of a contrast agent in a radiological examination method, the radiological examination method comprising: - receiving radiological images, the radiological images representing an examination area of at least one examination object at a number N of consecutive points in time, at least after an application of the contrast agent, each radiological image comprising a plurality of image elements, each image element representing a sub-area of the examination area, - for each image element: extracting one or more features from the radiological images, - generating a time series for each sub-area of the examination area, each time series of a sub-area comprising the one or more extracted features of the image elements representing the sub-area for the number N of consecutive points in time, - selecting at least some of the time series, - grouping the selected time series into a number M of groups, - assigning each non-selected time series to one of the M groups, - assigning each image element to the group to which the time series of the sub-area that the image element represents is assigned, - assigning a label from M labels each image element, wherein each identifier of the M identifiers represents a group of the M groups, - outputting and/or storing at least one of the radiological images comprising the identifiers of the image elements of the radiological image and/or transmitting the at least one radiological image comprising the identifiers of the image elements of the radiological image to a separate computer system. 17. A contrast agent for use in a radiological examination method, the radiological examination method comprising: - receiving radiological images, the radiological images representing an examination area of at least one examination object at a number N of consecutive points in time, at least after an application of the contrast agent, each radiological image comprising a plurality of image elements, each image element representing a sub-area of the examination area, - for each image element: extracting one or more features from the radiological images, - generating a time series for each sub-area of the examination area, each time series of a sub-area comprising the one or more extracted features of the image elements representing the sub-area for the number N of consecutive points in time, - selecting at least some of the time series, - grouping the selected time series into a number M of groups, - assigning each non-selected time series to one of the M groups, - assigning each image element to the group to which the time series of the sub-area that the image element represents is assigned, - Assigning an identifier from M identifiers to each image element, wherein each identifier of the M identifiers represents a group of the M groups, - outputting and/or storing at least one of the radiological images comprising the identifiers of the image elements of the radiological image and/or transmitting the at least one radiological image comprising the identifiers of the image elements of the radiological image to a separate computer system. 18. Kit comprising - a contrast agent and - a computer program product comprising a data storage device in which a computer program is stored that can be loaded into a working memory of a computer system and causes the computer system to carry out the following steps: o receiving radiological images, the radiological images representing an examination area of at least one examination object at a number N of consecutive points in time, at least after an application of the contrast agent, each radiological image comprising a plurality of image elements, each image element representing a sub-area of the examination area, o for each image element: extracting one or more features from the radiological images, o generating a time series for each sub-area of the examination area, each time series of a sub-area comprising the one or more extracted features of the image elements representing the sub-area for the number N of consecutive points in time, o selecting at least some of the time series, o grouping the selected time series into a number M of groups, o assigning each non-selected time series to one of the M groups, o assigning each image element to the group to which the time series of the sub-area represented by the image element is assigned, o Assigning an identifier from M identifiers to each image element, wherein each identifier of the M identifiers represents a group of the M groups, o Outputting and/or storing at least one of the radiological images comprising the identifiers of the image elements of the radiological image and/or transmitting the at least one radiological image comprising the identifiers of the image elements of the radiological image to a separate computer system.