Generation of contrast-enhanced synthetic radiological images

Abstract

The present invention relates to the generation of synthetic radiological images. Such a synthetic radiological image is generated with the aid of a machine learning model based on a CT image of an examination area. The CT image depicts the examination area after the application of an MRI contrast agent. The machine learning model is configured and trained to generate a synthetic radiological image that corresponds to an MRI image of the examination area after application of the MRI contrast agent. The present invention relates to: a method for training the machine learning model; a computer-implemented method for generating the synthetic radiological image by means of the trained model; and a computer system and a computer programme product for executing the method for generating the synthetic radiological image.

Description

[0001] Generating contrast-enhanced synthetic radiological images

[0002] The present invention relates to the generation of synthetic radiological images. Such a synthetic radiological image is generated using a machine learning model based on a CT scan of an examination region of an examination subject. The CT scan shows the examination region after the application of an MRI contrast agent. The machine learning model is configured and trained to generate a synthetic radiological image that corresponds to an MRI scan of the examination region after the application of the MRI contrast agent. The present invention relates to a method for training the machine learning model, a computer-implemented method for generating the synthetic radiological image using the trained model, and a computer system and a computer program product for executing the method for generating the synthetic radiological image.

[0003] A person's liver can be affected by multiple benign tumors, which may appear as cystic or solid focal lesions in the liver parenchyma. The liver is further susceptible to malignant tumors, such as metastases from extrahepatic cancers or from primary cancers originating in the liver itself. Worldwide, the two most common types of malignant liver tumors are metastases—particularly colorectal cancer metastases—and hepatocellular carcinoma (HCC). Approximately 20% of colorectal cancer patients have liver metastases at the time of diagnosis, and more than 50% of colorectal cancer patients develop liver metastases during the course of their disease. Hepatocellular carcinoma (HCC) is the most common primary liver cancer. It is also the sixth most common cancer worldwide and the fourth leading cause of cancer-related deaths.

[0004] The accurate and reliable detection and characterization of focal liver lesions in early stages of disease is of high clinical relevance, especially in patients at risk for liver metastases or primary liver cancer, as they are fundamental for appropriate treatment planning and determine the suitability for potentially curative treatment options.

[0005] Magnetic resonance imaging (MRI) is particularly important for the radiological examination of liver lesions. It is characterized by excellent soft tissue contrast and high spatial resolution without exposing the patient to ionizing radiation or iodinated contrast agents.

[0006] The most commonly used contrast agents in MRI are gadolinium-based paramagnetic contrast agents. These agents are administered via an intravenous (iv) bolus injection. Their contrast-enhancing effect is due to the central gadolinium ion (Gd(III), Gd 3+ ) in the chelate complex. When Tl-weighted scanning sequences are used in MRI, the gadolinium ion-induced shortening of the spin-lattice relaxation time (TI) of excited nuclei leads to an increase in signal intensity and thus to an increase in image contrast of the examined tissue.

[0007] According to their distribution pattern in the tissue, gadolinium-based contrast agents can be roughly divided into extracellular and intracellular contrast agents.

[0008] Extracellular contrast agents are low-molecular-weight, water-soluble compounds that, after intravenous administration, are distributed throughout the blood vessels and interstitial space. They are excreted via the kidneys after a relatively short period of circulation. Examples of extracellular MRI contrast agents include the gadolinium chelates gadobutrol (Gadovist®), gadoteridol (Prohance®), gadoteric acid (Dotarem®), gadopentetic acid (Magnevist®), and gadodiamide (Omnican®).

[0009] Intracellular contrast agents are partially absorbed into tissue cells and then excreted again. Intracellular MRI contrast agents based on gadoxetic acid, for example, are characterized by their specific uptake by liver cells, the hepatocytes, their accumulation in functional tissue (parenchyma), and their contrast enhancement in healthy liver tissue before being excreted via bile into the feces. Examples of such contrast agents based on gadoxetic acid are described in US Pat. No. 6,039,931A; they are commercially available, for example, under the brand names Primovist® and Eovist®. Another MRI contrast agent with lower uptake into hepatocytes is gadobenate dimeglumine (Multihance®).

[0010] Gadoxetate disodium (GD, Primovist®) belongs to the group of intracellular contrast agents. It is approved for use in liver MRI to detect and characterize lesions in patients with known or suspected focal liver disease. With its lipophilic ethoxybenzyl moiety, GD exhibits a biphasic distribution: initially, distribution in the intravascular and interstitial space after bolus injection, followed by selective uptake by hepatocytes. GD is excreted unchanged from the body in approximately equal amounts via the kidneys and the hepatobiliary route (50:50 dual excretion mechanism). Due to its selective accumulation in healthy liver tissue, GD is also referred to as a hepatobiliary contrast agent.

[0011] GD is approved at a dose of 0.1 ml / kg body weight (BW) (0.025 mmol / kg BW Gd). The recommended administration of GD involves an undiluted intravenous bolus injection at a flow rate of approximately 2 ml / second, followed by flushing of the IV cannula with normal saline. A standard protocol for liver imaging using GD consists of several planning and pre-contrast sequences. After an IV bolus injection of the contrast agent, dynamic images are usually acquired during the arterial (approximately 30 seconds post-injection, pi), portal venous (approximately 60 seconds pi), and transition phases (approximately 2-5 minutes pi). The transition phase typically already shows some increase in liver signal intensity due to the onset of uptake of the agent into hepatocytes.Additional T2-weighted and diffusion-weighted (DWI) images can be obtained after the dynamic phase and before the late hepatobiliary phase.

[0012] Contrast-enhanced dynamic images from the arterial, portal venous, and transition phases provide crucial information about the time-varying patterns of lesion enhancement (vascularization), which contribute to the characterization of the specific liver lesion. Hepatocellular carcinoma, with its typical arterial phase hyperenhancement (APHE) and contrast washout in the venous phase, can be diagnosed solely based on its unique vascularization pattern observed during dynamic phase imaging, thus protecting patients from an invasive and potentially risky liver biopsy.

[0013] Other lesions can also be characterized using dynamic contrast-enhanced MRI.

[0014] In the diagnosis of liver lesions, a hepatobiliary contrast agent has the advantage over an extracellular contrast agent in that it has a higher sensitivity and can therefore better detect smaller carcinomas in particular (see e.g.: RF Hanna et al.: Comparative 13-year meta-analysis of the sensitivity and positive predictive value of ultrasound, CT, and MRI for detecting hepatocellular carcinoma, Abdom Radiol 2016, 41, 71-90; YJ Lee et al.: Hepatocellular carcinoma: diagnostic performance of multidetector CT and MR imaging-a systematic review and metaanalysis, Radiology 2015, 275, 97-109; DK Owens et al.: High-value, cost-conscious health care: concepts for clinicians to evaluate the benefits, harms, and costs of medical interventions, Ann Intern Med 2011, 154, 174-180).

[0015] Computed tomography (CT) has the advantage over magnetic resonance imaging in that surgical interventions can be performed while CT images are being generated. While a surgeon is performing an intervention in the area under examination, they can visualize the area under examination using the CT.

[0016] For example, if a surgeon wishes to perform an operation on a patient's liver, e.g., to perform a biopsy of a liver lesion or to remove a tumor, the contrast between a liver lesion or tumor and healthy liver tissue in a CT scan of the liver will not be as pronounced as in an MRI scan after the application of a hepatobiliary contrast agent. Currently, no hepatobiliary CT-specific contrast agents are known and / or approved for CT. It would therefore be desirable to be able to produce a radiological image using a CT scanner in which liver lesions could be differentiated from healthy liver tissue as well as in an MRI scan after the application of a hepatobiliary MRI contrast agent.

[0017] This object and further objects are achieved by the subject matter of the present invention.

[0018] A first subject of the present invention is a computer-implemented method for training a machine learning model, comprising

[0019] Receiving and / or providing training data, wherein the training data for each examination object of a plurality of examination objects comprises i) an MRI image of an examination region of the examination object after the application of an MRI contrast agent and ii) a CT image of the examination region of the examination object after the application of the MRI contrast agent,

[0020] Providing a machine learning model, wherein the machine learning model is configured to generate a synthetic radiological image based on a CT image representing an examination area of ​​an examination subject after the application of an MRI contrast agent and based on model parameters,

[0021] Training the machine learning model, wherein the training comprises for each examination object of the plurality of examination objects: o Inputting the CT image into the machine learning model o Receiving a synthetic radiological image from the machine learning model o Determining a deviation between the synthetic radiological image and the MRI image o Modifying the model parameters with a view to reducing the deviation

[0022] Storing and / or outputting the trained machine learning model and / or transmitting the trained machine learning model to a separate computer system and / or using the trained machine learning model to generate an artificial MRI image based on a newly measured CT image. A further subject of the present invention is a computer-implemented method for generating a synthetic radiological image, comprising

[0023] Receiving a CT image of an examination area of ​​an examination object after application of an MRI contrast agent,

[0024] Inputting the CT images into a trained machine learning model, wherein the machine learning model is trained on the basis of training data comprising, for each examination object of a plurality of examination objects, i) a CT image of the examination area after the application of the MRI contrast agent and ii) an MRI image after application of the MRI contrast agent, to generate a synthetic radiological image based on the CT image,

[0025] Receiving a synthetic radiological image from the trained machine learning model,

[0026] Outputting and / or storing the synthetic radiological image and / or transmitting the synthetic radiological image to a separate computer system.

[0027] Another object of the present invention is a computer system comprising: a processor; and a memory storing an application program configured to perform an operation when executed by the processor, the operation comprising:

[0028] Receiving a CT image of an examination area of ​​an examination object after application of an MRI contrast agent,

[0029] Inputting the CT images into a trained machine learning model, wherein the machine learning model is trained on the basis of training data comprising, for each examination object of a plurality of examination objects, i) a CT image of the examination area after the application of the MRI contrast agent and ii) an MRI image after the application of the MRI contrast agent, to generate a synthetic radiological image based on the CT image,

[0030] Receiving a synthetic radiological image from the trained machine learning model,

[0031] Outputting and / or storing the synthetic radiological image and / or transmitting the synthetic radiological image to a separate computer system.

[0032] Another object of the present invention is a computer program product that can be loaded into a working memory of a computer system and causes the computer system to carry out the following steps:

[0033] Receiving a CT image of an examination area of ​​an examination object after application of an MRI contrast agent,

[0034] Inputting the CT images into a trained machine learning model, wherein the machine learning model is trained on the basis of training data comprising, for each examination object of a plurality of examination objects, i) a CT image of the examination area after the application of the MRI contrast agent and ii) an MRI image after the application of the MRI contrast agent, to generate a synthetic radiological image based on the CT image,

[0035] Receiving a synthetic radiographic image from the trained machine learning model, outputting and / or storing the synthetic radiographic image, and / or transmitting the synthetic radiographic image to a separate computer system.

[0036] Another object of the present invention is an MRI contrast agent for use in a CT examination method comprising:

[0037] Receiving a CT image of an examination area of ​​an examination object after application of an MRI contrast agent,

[0038] Inputting the CT images into a trained machine learning model, wherein the machine learning model is trained on the basis of training data comprising, for each examination object of a plurality of examination objects, i) a CT image of the examination area after the application of the MRI contrast agent and ii) an MRI image after the application of the MRI contrast agent, to generate a synthetic radiological image based on the CT image,

[0039] Receiving a synthetic radiological image from the trained machine learning model,

[0040] Outputting and / or storing the synthetic radiological image and / or transmitting the synthetic radiological image to a separate computer system.

[0041] A further object of the present invention is the use of an MRI contrast agent in a CT examination method comprising:

[0042] Receiving a CT image of an examination area of ​​an examination object after application of an MRI contrast agent,

[0043] Inputting the CT images into a trained machine learning model, wherein the machine learning model is trained on the basis of training data comprising, for each examination object of a plurality of examination objects, i) a CT image of the examination area after the application of the MRI contrast agent and ii) an MRI image after the application of the MRI contrast agent, to generate a synthetic radiological image based on the CT image,

[0044] Receiving a synthetic radiological image from the trained machine learning model,

[0045] Outputting and / or storing the synthetic radiological image and / or transmitting the synthetic radiological image to a separate computer system.

[0046] A further subject of the present invention is a kit comprising an MRI contrast agent and a computer program product, wherein the computer program product comprises a computer program that can be loaded into a working memory of a computer system and causes the computer system to carry out the following steps:

[0047] Receiving a CT image of an examination area of ​​an examination object after application of an MRI contrast agent,

[0048] Inputting the CT images into a trained machine learning model, wherein the machine learning model is trained on the basis of training data comprising, for each examination object of a plurality of examination objects, i) a CT image of the examination area after the application of the MRI contrast agent and ii) an MRI image after the application of the MRI contrast agent, to generate a synthetic radiological image based on the CT image,

[0049] Receiving a synthetic radiographic image from the trained machine learning model, outputting and / or storing the synthetic radiographic image, and / or transmitting the synthetic radiographic image to a separate computer system.

[0050] The invention is explained in more detail below, without distinguishing between the subject matters of the invention (training method, prediction method, computer system, computer program product, use, contrast agent for use, kit). Rather, the following explanations are intended to apply analogously to all subject matters of the invention, regardless of the context in which they occur (training method, prediction method, computer system, computer program product, use, contrast agent for use, kit).

[0051] If steps are mentioned in a particular order in this description or in the claims, this does not necessarily mean that the invention is limited to that order. Rather, it is conceivable that the steps may be performed in a different order or even in parallel; unless a step builds on another step, which absolutely requires that the subsequent step be performed (which will become clear in individual cases). The specified sequences thus represent preferred embodiments of the invention.

[0052] The invention is explained in more detail at some points with reference to drawings. The drawings depict specific embodiments with specific features and combinations of features, which primarily serve for illustrative 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 regarding features and combinations of features are intended to apply generally, meaning they are also transferable to other embodiments and are not limited to the embodiments shown.

[0053] The invention provides means by which a synthetic radiological image of an examination region of an examination object can be generated.

[0054] The "subject under investigation" is usually a living being, preferably a mammal, and most preferably a human. The subject under investigation is also referred to as the patient in this description.

[0055] The “area of ​​investigation” is a part of the object of investigation, for example an organ or part of an organ such as the liver, brain, heart, kidney, lung, stomach, intestine or part of the aforementioned organs or several organs or another part of the body.

[0056] In a preferred embodiment of the present invention, the examination area is the liver or a part of the liver of a human.

[0057] The examination area, also called the field of view (FOV), represents a volume depicted in radiological images. The examination area is typically defined by a radiologist, for example, on a localizer image. Alternatively or additionally, the examination area can also be defined automatically, for example, based on a selected protocol.

[0058] The synthetic radiological image is based on at least one CT scan. The synthetic radiological image represents an examination area of ​​a subject after the application of an MRI contrast agent.

[0059] The synthetic radiographic image is generated using a trained machine learning model. A "machine learning model" can be understood as a computer-implemented data processing architecture. The model can receive input data and produce output data based on this input data and model parameters. Through training, the model can learn a relationship between the input data and the output data. During training, model parameters can be adjusted to produce a desired output for a given input.

[0060] When training such a model, the model is presented with training data from which it can learn. The trained machine learning model is the result of the training process. The training data includes input data and the correct output data (target data) that the model is supposed to generate based on the input data. During training, patterns are recognized that map the input data to the target data.

[0061] During the training process, the input data from the training data is fed into the model, and the model generates output data. The output data is compared with the target data. Model parameters are adjusted to reduce the deviations between the output data and the target data to a (defined) minimum.

[0062] The deviations can be quantified using a loss function. Such an error function can be used to calculate an error (loss value) for a given pair of output and target data. The goal of the training process can be to modify (adjust) the parameters of the machine learning model so that the error is reduced to a (defined) minimum for all pairs in the training dataset.

[0063] For example, if the output and target data are numbers, the error function can be the absolute difference between these numbers. In this case, a high absolute error may indicate that one or more model parameters need to be significantly changed.

[0064] For example, for output data in the form of vectors, difference metrics between vectors such as the mean square error, a cosine distance, a norm of the difference vector such as a Euclidean distance, a Chebyshev distance, an Lp norm of a difference vector, a weighted norm, or any other type of difference metric between two vectors can be chosen as the error function.

[0065] For higher-dimensional outputs, such as two-dimensional, three-dimensional, or higher-dimensional outputs, an element-wise difference metric can be used. Alternatively or additionally, the output data can be transformed, e.g., into a one-dimensional vector, before calculating an error value.

[0066] In this case, the machine learning model is trained using training data to generate a synthetic radiological image based on at least one CT scan of the examination area of ​​the subject after the application of an MRI contrast agent. This image corresponds, in particular in terms of its quality, contrast distribution, and resolution, to an MRI scan of the examination area of ​​the subject after the application of an MRI contrast agent. In other words, the machine learning model is trained to generate (predict) an artificial MRI image based on at least one CT scan of the examination area of ​​the subject after the application of an MRI contrast agent, which represents the examination area of ​​the subject after the application of the MRI contrast agent.In other words, the artificial radiological image is generated on the basis of at least one CT image, but has a contrast distribution that corresponds to an MRI image.

[0067] The training data includes, for each of a plurality of examination objects, at least one CT scan and at least one MRI scan, wherein the at least one CT scan represents the examination region of the respective examination object after the application of an MRI contrast agent, and the at least one MRI scan represents the examination region of the respective examination object after the application of the MRI contrast agent. The examination region is usually the same for all examination objects. Different examination regions can also be selected for different examination objects.

[0068] Typically, the contrast of structures in a CT scan differs from that in an MRI scan. The machine vision model is trained to transfer the contrast distribution in an MRI scan to a CT scan. This allows structures in a CT scan that are contrast-enhanced in a comparable MRI scan to be highlighted more clearly in the CT scan. For example, a high contrast between healthy overlying tissue and lesions in an MRI scan can be transferred to a CT scan after the application of a hepatobiliary MRI contrast agent. In other words, this corresponds to the conversion of a hepatobiliary MRI contrast agent into a liver-specific CT contrast agent.

[0069] Preferably, the at least one CT scan and the at least one MRI scan are in the same format. They preferably have the same image size (e.g., number of pixels or voxels in the image dimensions). Preferably, they are each representations of the examination area in a spatial representation. Preferably, the at least one CT scan and the at least one MRI scan are in the form of digital image data.

[0070] The term "digital" means that the representations can be processed by a machine, usually a computer system. "Processing" refers to the well-known methods of electronic data processing (EDP). An example of a common format for digital image data is the DICOM format (DICOM: Digital Imaging and Communications in Medicine) – an open standard for storing and exchanging information in medical image data management.

[0071] Both the at least one CT scan and the at least one MRI scan represent the examination area of ​​the respective examination subject after the application of an MRI contrast agent. Typically, the contrast agent is the same in both cases. Preferably, but not necessarily, the same amount of MRI contrast agent is applied.

[0072] The at least one CT scan and the at least one MRI scan can represent the examination area at the same time after the application of the MRI contrast agent or at different times. For example, it is possible for the at least one MRI scan to represent the liver or part of the liver of an examination subject in the hepatobiliary phase after the application of a hepatobiliary MRI contrast agent (e.g., 10 to 20 minutes after the application of the hepatobiliary MRI contrast agent), while the at least one CT scan represents the liver or part of the liver at an earlier time (e.g., within the first 10 minutes after the application of the contrast agent) or at a later time (e.g., more than 20 minutes after the application of the contrast agent) after the application of the hepatobiliary contrast agent.

[0073] The MRI contrast agent is preferably an intracellular contrast agent. Most preferably, the MRI contrast agent is a hepatobiliary contrast agent. A hepatobiliary contrast agent is a contrast agent that is specifically absorbed by healthy liver cells, the hepatocytes. Examples of hepatobiliary contrast agents are contrast agents based on gadoxetic acid. These are described, for example, in US Pat. No. 6,039,931A. They are commercially available, for example, under the brand names Primovist® or Eovist®.

[0074] The contrast-enhancing effect of Primovist® / Eovist® is mediated by the stable gadolinium complex Gd-EOB-DTPA (gadolinium-ethoxybenzyl-diethylenetriamine-pentaacetic acid). DTPA forms a complex with the paramagnetic gadolinium ion that exhibits extremely high thermodynamic stability. The ethoxybenzyl moiety (EOB) mediates hepatobiliary uptake of the contrast agent.

[0075] In a particularly preferred embodiment, the contrast agent used is a substance or a mixture of substances containing gadoxetic acid or a salt of gadoxetic acid as the contrast-enhancing agent. Most preferably, this is the disodium salt of gadoxetic acid (Gd-EOB-DTPA disodium).

[0076] In one embodiment of the present disclosure, the contrast agent is an agent comprising a Gd 3+ -Complex of a compound of formula (I)

[0077] (I) , wherein

[0078] Are a group selected from where # represents the connection to X,

[0079] X represents a group consisting of

[0080] 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,

[0081] R1 , R 2 and R 3 independently represent a hydrogen atom or a group selected from C1-C3 alkyl, -CH2OH, -(CH2)2OH and -CH2OCH,

[0082] R 4 a 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-,

[0083] R 5 represents a hydrogen atom, and

[0084] R 6 represents a hydrogen atom, or a stereoisomer, tautomer, hydrate, solvate or salt thereof, or a mixture thereof.

[0085] In one embodiment of the present disclosure, the contrast agent is an agent comprising a Gd 3+ -Complex of a compound of formula (II)

[0086] Are a group selected from where # represents the connection to X,

[0087] X represents a group consisting of CH2, (CH2)2, (CFF , (CH2)4 and *-(CH2)2-O-CH2- # is selected, where * represents the connection to Ar and # represents the link to the acetic acid residue,

[0088] R 7 represents a hydrogen atom or a group selected from C1-C3 alkyl, -CH2OH, -(CH2)2OH and -CH2OCH3;

[0089] R 8 a group selected from

[0090] C2-C4-alkoxy, (H3C-CH2O)-(CH2)2-O-, (H3C-CH2O)-(CH2)2-O-(CH2)2-O- and

[0091] (H3C-CH2O)-(CH2)2-O-(CH2)2-O-(CH2)2-O-;

[0092] R 9 and R 10 independently represent a hydrogen atom; or a stereoisomer, tautomer, hydrate, solvate or salt thereof, or a mixture thereof.

[0093] The term "Ci-C3 alkyl" means a linear or branched, saturated, monovalent hydrocarbon group containing 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 containing 2, 3, or 4 carbon atoms.

[0094] 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.

[0095] In one embodiment of the present disclosure, the contrast agent is an agent comprising gadolinium 2,2',2"-(10-{l-carboxy-2-[2-(4-ethoxyphenyl)ethoxy]ethyl}-l,4,7,10-tetraazacyclododecane-l,4,7-triyl)triacetate (see, for example, WO2022 / 194777, Example 1). In one embodiment of the present disclosure, the contrast agent is an agent comprising gadolinium 2,2',2"-{10-[l-carboxy-2-{4-[2-(2-ethoxyethoxy)ethoxy]phenyl}ethyl]-

[0096] 1.4.7.10-tetraazacyclododecane-1,4,7-triyl}triacetate (see e.g. WO2022 / 194777, Example 2).

[0097] In one embodiment of the present disclosure, the contrast agent is an agent comprising gadolinium 2,2',2"-{10-[(lR)-l-carboxy-2-{4-[2-(2-ethoxyethoxy)ethoxy]phenyl}ethyl]-

[0098] 1.4.7.10-tetraazacyclododecane-1,4,7-triyl}triacetate (see e.g. WO2022 / 194777, Example 4).

[0099] In one embodiment of the present disclosure, the contrast agent is an agent comprising gadolinium (2S,2'S,2"S)-2,2',2"-{10-[(lS)-l-carboxy-4-{4-[2-(2-ethoxyethoxy)ethoxy]phenyl}butyl]-l,4,7,10-tetraazacyclododecane-l,4,7-triyl}tris(3-hydroxypropanoate) (see, for example, WO2022 / 194777, Example 15).

[0100] In one embodiment of the present disclosure, the contrast agent is an agent comprising gadolinium 2,2',2"-{10-[(lS)-4-(4-butoxyphenyl)-l-carboxybutyl]-l,4,7,10-tetraazacyclododecane-1,4,7-triyl}triacetate (see, for example, WO2022 / 194777, Example 31).

[0101] In one embodiment of the present disclosure, the contrast agent is an agent comprising gadolinium -2,2',2"-{(2.S)-1 O-(carboxymethyl)-2-|4-(2-ethoxyethoxy)benzyl]-1,4,7,10-tetraazacyclododecane-1,4,7-triyl}triacetate.

[0102] 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.

[0103] 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-

[0104] 2.5.8.11-tetraazadodecane-l-carboxylate hydrate (also known as gadodiamide).

[0105] 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-I-yl]acetate (also referred to as gadoteridol).

[0106] 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).

[0107] The training data may include additional data as input and / or target data for each of the plurality of objects under investigation.

[0108] For example, in addition to the at least one CT image representing the examination area of ​​the examination subject after the application of the MRI contrast agent, the training data may include as input data at least one further CT image representing the examination area of ​​the examination subject before the application of the contrast agent, ie without contrast agent.

[0109] The training data can also include several CT images as input data that were generated at different times before and / or after the application of the MRI contrast agent.

[0110] The training data can also include multiple CT scans as input data, which were generated under different measurement conditions and / or using different measurement parameters. If multiple CT scans are fed to the machine learning model as input data, a registration process is usually performed beforehand.

[0111] Such registration (also called "co-registration" or image registration) is a process in digital image processing and serves to optimally match two or more images (e.g., pictures) of the same scene, or at least of similar scenes. One of the images is designated as the reference image, and the others are called object images. To optimally adjust these object images to the reference image, a compensating transformation is calculated. The images to be registered differ from one another because they were taken from different positions, at different times, or with different sensors.

[0112] In the case of the present invention, different CT images may have been acquired at different times and / or may contain different amounts of a contrast agent.

[0113] The goal of image registration is therefore to find the transformation that best matches a given object image with the reference image. The goal is that, if possible, each pixel / voxel of an image represents the same region of interest in an object as the pixel / voxel of another (co-registered) image with the same coordinates.

[0114] Methods for image registration are described in the state of the art (see e.g.: EH Seeley et al.: Co-registration of multi-modality imaging allows for comprehensive analysis of tumor-induced hone disease, Bone 2014, 61, 208-216; C. Bhushan et al.: Co-registration and distortion correction of diffusion and anatomical images based on inverse contrast normalization, Neuroimage 2015, 15, 115: 269-80; US20200214619; US20090135191; EP3639272A).

[0115] The training data may also include other data as input data, such as information about the examination subject (e.g. age, gender, height, weight, body mass index, resting heart rate, heart rate variability, body temperature, information about the lifestyle of the examination subject, such as alcohol consumption, smoking and / or physical activity and / or the diet of the examination subject, medical intervention parameters such as regular medication, occasional medication or other previous or current medical interventions and / or other information about previous and current treatments of the examination subject and reported health conditions and / or combinations thereof), information about the examination area (e.g. size and / or shape of the examination area and / or the location of the examination area within the examination subject) and / or information about the at least one CT scan (e.g.under which conditions it was generated and / or which measurement parameters and / or measurement sequences were used to generate it, when it was generated) and / or information on the MRI contrast agent used (e.g. which contrast agent was used, in what quantity, how it was administered, where it was administered).

[0116] When training the machine learning model, the input data is fed into the machine learning model. The machine learning model is configured to generate at least one synthetic radiological image based on the input data and model parameters. The at least one synthetic radiological image can be compared with the at least one MRI image of the training data (with the target data).

[0117] Using an error function, the deviations between the at least one synthetic radiological image and the at least one MRI image can be quantified. Suitable error functions for quantifying deviations between two radiological images include, for example, the LI error function (LI loss), the L2 error function (L2 loss), the Lp error function (Lp loss), the structural similarity index measure (SSIM), the VGG error function (VGG loss), the perceptual loss, or a combination of the above-mentioned functions or other error functions. Further details on error functions can be found in the scientific literature (see e.g.: R. Mechrez et al. : The Contextual Loss for Image Transformation with Non-Aligned Data, 2018, arXiv: 1803.02077v4; H. Zhao et al:. Loss Functions for Image Restoration with Neural Networks , 2018, arXiv: 1511.08861v3).

[0118] Using a gradient method or another optimization method, the model parameters can be modified so that the error for all pairs comprising the synthetic radiological image (output data) and the at least one MRI image (target data) is reduced and preferably falls below a predefined threshold and / or is reduced to a defined minimum.

[0119] Once the model has been trained on the basis of a large number of pairs of input data and target data and the error (L) has reached a defined minimum for all pairs of output data and target data, the trained machine learning model and / or the modified model parameters can be stored in a data storage and / or transmitted to a separate computer system and / or used for prediction.

[0120] Fig. 1 shows an exemplary and schematic embodiment of the method for training the machine learning model.

[0121] The machine learning model is trained using training data (TD). Typically, the training data includes input data and target data for each of a plurality of examination objects. Fig. 1 shows only one dataset of an examination object, which consists of the input data (I) and the target data (T). In this example, the input data (I) consists of at least one CT scan (CT 1), which represents a portion of a liver of an examination subject at a time after the application of a hepatobiliary MRI contrast agent, and at least one unenhanced CT image (CT°), which represents the liver of the examination subject without a contrast agent. As described, the use of a unenhanced CT image is optional. As described, the input data can include further data. In this example, the target data (T) consists of an MRI image, which represents the portion of the liver at a time in the hepatobiliary phase after the application of the MRI contrast agent.

[0122] The input data (I) is fed into the machine learning model (MLM). The machine learning model (MLM) is configured to generate output data (O) based on the input data (I) and model parameters (MP). The output data (O) represents a predicted MRI image (MRI*) that represents the part of the liver of the subject at a time point in the hepatobiliary phase after the application of the hepatobiliary contrast agent. In particular, with regard to their quality, contrast distribution, and resolution, the output data (O) correspond to an MRI image of the subject's examination area after the application of the hepatobiliary MRI contrast agent.

[0123] Using an error function (LF), deviations between the predicted MRI image (MRI*) and the MRI image (MRI) of the training data are quantified. An error (L) can be calculated for each pair of output data and target data. The error (L) can be used to modify model parameters (MP) so that the error (L) is reduced to a defined minimum. Once the model has been trained based on numerous pairs of input data and target data and the error (L) has reached a defined minimum for all pairs, the trained machine learning model can be used for prediction. This is shown schematically in Fig. 2.

[0124] Fig. 2 shows, by way of example and schematically, an embodiment of the method for generating a synthetic radiological image, in this case the generation of an artificial MRI image of a liver or part of the liver of a patient, wherein the MRI image represents the liver or part of the liver in a hepatobiliary phase after the application of a hepatobiliary contrast agent.

[0125] The prediction is made using a trained machine learning model (MLM). The machine learning model can, for example, have been trained as described in relation to Fig. 1.

[0126] The prediction is based on patient data (PD). The patient data (PD) includes at least one CT scan (CT 1) representing a liver or part of a liver of an examination subject after the application of the hepatobiliary contrast agent. In the present example, the patient data (PD) further includes at least one native CT scan (CT°). The at least one native MRI scan (CT°) is optional and shows the liver or part of the liver without a contrast agent. The patient data may include further data, particularly if the machine learning model was also trained based on further input data.

[0127] The patient data (PD) is fed into the trained machine learning model (MLM). The machine learning model (MLM) is configured and trained to generate a synthetic radiological image based on the patient data (PD) and the model parameters (MP) modified (learned) during training. The synthetic radiological image corresponds, particularly in terms of its quality, contrast distribution, and resolution, to an MRI image depicting the liver or part of the liver of the subject in the hepatobiliary phase after the application of the hepatobiliary contrast agent.

[0128] The output of the predicted MRI image (MRI*) can be displayed on a monitor, printed on a printer, stored on a data storage device, and / or transmitted to a separate computer system.

[0129] The machine learning model can, for example, be or include an artificial neural network.

[0130] An artificial neural network comprises at least three layers of processing elements: a first layer with input neurons (nodes), an Nth layer with at least one output neuron (node), and N-2 inner layers, where N is a natural number and greater than 2.

[0131] The input neurons serve to receive the input data, in particular the at least one CT image of a liver or part of the liver of an examination subject and optionally at least one native CT image. Typically, there is one input neuron for each pixel or voxel of a CT image. Additional input neurons can be present for additional input values ​​(e.g., information on the examination region, the examination subject, conditions prevailing when the CT images were generated, information on the state represented by the CT image, and / or information on the time or time period at / in which the CT image was generated). The output neurons can serve to output a predicted MRI image representing the examination region in the hepatobiliary phase.

[0132] The processing elements of the layers between the input neurons and the output neurons are connected in a predetermined pattern with predetermined connection weights.

[0133] Preferably, the artificial neural network is a so-called convolutional neural network (CNN for short) or it includes one.

[0134] A CNN usually consists essentially of filters (convolutional layer) and aggregation layers (pooling layer), which repeat alternately, and at the end of one or more layers of “normal” fully connected neurons (dense / fully connected layer).

[0135] The neural network can be trained, for example, using a backpropagation method. The goal is to achieve the most reliable prediction possible for the second MRI scan. The quality of the prediction is described by an error function. The goal is to minimize the error function. In the backpropagation method, an artificial neural network is trained by changing the connection weights.

[0136] In the trained state, the connection weights between the processing elements contain information regarding the relationship between the at least one CT scan and optionally the at least one native CT scan and the MRI scan, which can be used for prediction purposes.

[0137] A cross-validation method can be used to split the data into training and validation sets. The training set is used for backpropagation training of the network weights. The validation set is used to test the predictive accuracy of the trained network when applied to unknown data.

[0138] The artificial neural network may have an autoencoder architecture; for example, the artificial neural network may have an architecture such as the U-Net (see, for example, O. Ronneberger et al.: U-net: Convolutional networks for biomedical image segmentation, International Conference on Medical image computing and computer-assisted intervention, pages 234-241, Springer, 2015, https: / / doi.org / 10.1007 / 978-3-319-24574-4_28).

[0139] The artificial neural network can be a Generative Adversarial Network (GAN) (see, e.g., M.-Y. Liu et al. '. Generative Adversarial Networks for Image and Video Synthesis: Algorithms and Applications, arXiv:2008.02793; J. Henry et al. '. Pix2Pix GAN for Image-to-Image Translation, DOI: 10. 13140 / RG.2.2.32286.66887).

[0140] The artificial neural network can be a recurrent neural network or comprise one. Recurrent or feedback neural networks are neural networks that, in contrast to feedforward networks, are characterized by connections between neurons in one layer and neurons in the same or a previous layer. The artificial neural network can, for example, comprise a long short-term memory (LSTM) (see, for example, Y. Gao et al.: Fully convolutional structured LSTM networks for joint 4D medical image segmentation, DOI: 10.1109 / ISBI.2018.8363764).

[0141] The artificial neural network can be a transformer network (see, for example, D. Karimi et al. '. Convolution-Free Medical Image Segmentation using Transformers, arXiv:2102.13645 [eess.IV]).

[0142] The invention may be carried out in whole or in part by means of a computer system.

[0143] Fig. 3 shows an exemplary and schematic embodiment of the computer system according to the invention. The computer system (1) comprises a processing unit (20) connected to a memory (50).

[0144] The processing unit (20) may comprise one or more processors alone or in combination with one or more memories. The processing unit (20) may be conventional computer hardware capable of processing information such as digital images, computer programs, and / or other digital information. The processing unit (20) typically consists of an arrangement of electronic circuits, some of which may be embodied as an integrated circuit or as multiple interconnected integrated circuits (an integrated circuit is sometimes referred to as a "chip"). The processing unit (20) may be configured to execute computer programs that may be stored in a main memory of the processing unit (20) or in the memory (50) of the same or another computer system.

[0145] The memory (50) 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 (50) may comprise volatile and / or non-volatile memory and may be permanently installed or removable. Examples of suitable memories include RAM (Random Access Memory), ROM (Read-Only Memory), a hard disk, flash memory, a removable computer diskette, an optical disc, magnetic tape, or a combination of the above. Optical discs may include read-only compact discs (CD-ROM), read / write compact discs (CD-R / W), DVDs, Blu-ray discs, and the like.

[0146] In addition to the memory (50), the processing unit (20) can also be connected to one or more interfaces (11, 12, 30, 41, 42) for displaying, transmitting, and / or receiving information. The interfaces can include one or more communication interfaces (11, 12) and / or one or more user interfaces (30, 41, 42). The one or more communication interfaces can 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 devices, or the like. The one or more communication interfaces can be configured to transmit and / or receive information via physical (wired) and / or wireless communication connections.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.

[0147] The user interfaces may comprise a display (30). A display (30) 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 (PDP), or the like. The user input interface(s) (41, 42) 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 include 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 (60) may be stored in memory (50) and executed by the processing unit (20), which is thereby programmed to perform the functions described in this specification. The retrieval, loading, and execution of instructions of the computer program (60) may occur sequentially, such that one instruction is retrieved, loaded, and executed at a time. However, the retrieval, loading, and / or execution may also occur in parallel.

[0148] The machine learning model according to the invention can also be stored in the memory (50).

[0149] 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.

[0150] The present invention can be used for various purposes. Examples of applications are described below, without intending to limit the invention to these.

[0151] One possible application example concerns the use of MRI contrast agents in computed tomography examinations.

[0152] MRI contrast agents typically have a lower contrast-enhancing effect in a CT scan than CT contrast agents. Nevertheless, it can be advantageous to use an MRI contrast agent in a CT scan. One example is a minimally invasive intervention in a patient's liver, where a surgeon monitors the procedure using a CT scanner. Computed tomography (CT) has the advantage over magnetic resonance imaging that it allows for greater surgical interventions in the area under examination while CT images are being generated. However, there are only a few interventional instruments and surgical devices that are MRI-compatible. In addition, access to the patient is limited due to the magnets used in MRI.While a surgeon is performing an operation in the area being examined, he or she can use the CT to create an image of the area being examined and follow the procedure on a monitor.

[0153] For example, if a surgeon wishes to perform an intervention on a patient's liver, e.g., to perform a biopsy of a liver lesion or to remove a tumor, the contrast between a liver lesion or tumor and healthy liver tissue in a CT scan of the liver will not be as pronounced as in an MRI scan following the application of a hepatobiliary contrast agent. Currently, no hepatobiliary CT-specific contrast agents are known and / or approved for CT. The use of an MRI contrast agent, particularly a hepatobiliary MRI contrast agent in computed tomography, therefore combines the ability to differentiate between healthy and diseased liver tissue with the ability to perform an intervention while simultaneously visualizing the liver.

[0154] The comparatively low contrast enhancement achieved by the MRI contrast agent can be increased with the help of the present invention without having to administer a higher dose than the standard dose.

[0155] A first CT image can be generated without MRI contrast agent, and a second CT image can be generated after the application of an MRI contrast agent whose amount corresponds to the standard amount. Based on these generated CT images, a synthetic CT image can be generated, as described in this disclosure, in which the contrast induced by the MRI contrast agent can be varied within wide limits by changing the amplification factor a. Contrasts can be achieved that can otherwise only be achieved by applying an amount of MRI contrast agent that is higher than the standard amount.

Claims

Patent claims 1. A computer-implemented method for training a machine learning model, comprising Receiving and / or providing training data, wherein the training data for each examination object of a plurality of examination objects comprises i) an MRI image of an examination region of the examination object after the application of an MRI contrast agent and ii) a CT image of the examination region of the examination object after the application of the MRI contrast agent, Providing a model of the machine element, wherein the model of the machine element is configured to generate a synthetic radiological image based on a CT image representing an examination area of ​​an examination object after the application of an MRI contrast agent, and to generate a synthetic radiological image based on model parameters, Training the machine learning model, wherein the training comprises for each examination object of the plurality of examination objects: o Inputting the CT image into the machine learning model o Receiving a synthetic radiological image from the machine learning model o Determining a deviation between the synthetic radiological image and the MRI image o Modifying the model parameters with a view to reducing the deviation Storing and / or outputting the trained machine learning model and / or transmitting the trained machine learning model to a separate computer system and / or using the trained machine learning model to generate an artificial MRI image based on a new measured CT image.

2. Computer-implemented method for generating a synthetic radiological image comprising Receiving a CT image of an examination area of ​​an examination object after application of an MRI contrast agent, Inputting the CT images into a trained machine learning model, wherein the machine learning model is trained on the basis of training data comprising, for each examination object of a plurality of examination objects, i) a CT image of the examination area after the application of the MRI contrast agent and ii) an MRI image after the application of the MRI contrast agent, to generate a synthetic radiological image based on the CT image, Receiving a synthetic radiological image from the trained machine learning model, Outputting and / or storing the synthetic radiological image and / or transmitting the synthetic radiological image to a separate computer system.

3. Method according to one of claims 1 or 2, wherein the object to be examined is a living being, preferably a mammal, most preferably a human.

4. A computer system comprising: a processor; and a memory storing an application program configured to perform an operation when executed by the processor, the operation comprising: Receiving a CT image of an examination area of ​​an examination object after application of an MRI contrast agent, Inputting the CT images into a trained machine learning model, wherein the machine learning model is trained on the basis of training data comprising, for each examination object of a plurality of examination objects, i) a CT image of the examination area after the application of the MRI contrast agent and ii) an MRI image after the application of the MRI contrast agent, to generate a synthetic radiological image based on the CT image, Receiving a synthetic radiological image from the trained machine learning model, Outputting and / or storing the synthetic radiological image and / or transmitting the synthetic radiological image to a separate computer system.

5. A computer program product that can be loaded into the memory of a computer system and causes the computer system to perform the following steps: Receiving a CT image of an examination area of ​​an examination object after application of an MRI contrast agent, Inputting the CT images into a trained machine learning model, wherein the machine learning model is trained on the basis of training data comprising, for each examination object of a plurality of examination objects, i) a CT image of the examination area after the application of the MRI contrast agent and ii) an MRI image after the application of the MRI contrast agent, to generate a synthetic radiological image based on the CT image, Receiving a synthetic radiological image from the trained machine learning model, Outputting and / or storing the synthetic radiological image and / or transmitting the synthetic radiological image to a separate computer system.

6. MRI contrast agent for use in a CT examination procedure comprising: Receiving a CT image of an examination area of ​​an examination object after application of an MRI contrast agent, Inputting the CT images into a trained machine learning model, wherein the machine learning model, based on training data, generates, for each examination object of a plurality of examination objects, i) a CT image of the examination area after the application of the MRI contrast agent and ii) an MRI Image after application of the MRI contrast agent, is trained to generate a synthetic radiological image based on the CT image, Receiving a synthetic radiological image from the trained machine learning model, Outputting and / or storing the synthetic radiological image and / or transmitting the synthetic radiological image to a separate computer system.

7. Use of an MRI contrast agent in a CT examination procedure comprising: Receiving a CT image of an examination area of ​​an examination object after application of an MRI contrast agent, Inputting the CT images into a trained machine learning model, wherein the machine learning model is trained on the basis of training data comprising, for each examination object of a plurality of examination objects, i) a CT image of the examination area after the application of the MRI contrast agent and ii) an MRI image after the application of the MRI contrast agent, to generate a synthetic radiological image based on the CT image, Receiving a synthetic radiological image from the trained machine learning model, Outputting and / or storing the synthetic radiological image and / or transmitting the synthetic radiological image to a separate computer system.

8. Kit comprising an MRI contrast agent and a computer program product, wherein the computer program product comprises a computer program that can be loaded into a working memory of a computer system and causes the computer system to perform the following steps: Receiving a CT image of an examination area of ​​an examination object after application of an MRI contrast agent, Inputting the CT images into a trained machine learning model, wherein the machine learning model is trained on the basis of training data comprising, for each examination object of a plurality of examination objects, i) a CT image of the examination area after the application of the MRI contrast agent and ii) an MRI image after the application of the MRI contrast agent, to generate a synthetic radiological image based on the CT image, Receiving a synthetic radiological image from the trained machine learning model, Outputting and / or storing the synthetic radiological image and / or transmitting the synthetic radiological image to a separate computer system.

9. The method, computer system, computer program product, MRI contrast agent, use, or kit according to any one of claims 1 to 8, wherein training the machine learning model for each examination object of the plurality of examination objects comprises: o inputting the CT image into the machine learning model o receiving a synthetic radiological image from the machine learning model o determining a deviation between the synthetic radiological image and the MRI image o Modify the model parameters to reduce the deviation.

10. MRI contrast agent for use according to claim 6 or use according to claim 7, wherein the MRI contrast agent comprises a Gd 3+ -Complex of a compound of formula (I) (I) , wherein Are 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, R 1 , R 2 and R 3 independently represent a hydrogen atom or a group selected from C1-C3 alkyl, -CH2OH, -(CH2)2OH and -CH2OCH, R 4 a 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-, R 5 represents a hydrogen atom, and R 6 represents a hydrogen atom, or a stereoisomer, tautomer, hydrate, solvate or salt thereof, or a mixture thereof, or wherein the MRI contrast agent comprises a Gd 3+ -Complex of a compound of formula (II) Are a group selected from where # represents the connection to X, X represents a group consisting of CH2, (CH2)2, (CI , (CH2)4 and *-(CH2)2-O-CH2- # is selected, where * represents the connection to Ar and # represents the link to the acetic acid residue, R 7 represents a hydrogen atom or a group selected from C1-C3 alkyl, -CH2OH, -(CH2)2OH and -CH2OCH3; R 8 a 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-; R 9 and R 10 independently represent a hydrogen atom; or a stereoisomer, tautomer, hydrate, solvate or salt thereof, or a mixture thereof.

Images