Systems, methods, and computer-readable media for bone volume assessment

Abstract

The present invention relates to a system (100) for bone volume assessment based on MRI patient scan data and / or at least one set of composite CT data, the system comprising at least one processor module (110) and at least one storage module (120) for storing patient data, the processor module (110) being configured to process and / or analyze MRI patient scan data, at least one set of composite CT data and / or reduced patient data, preferably in combination with each other, so as to identify at least one bone volume parameter of a first patient, in particular bone mineral density, fat percentage, calcium concentration, total density, intertrabecular space, etc. Furthermore, the present invention relates to a corresponding method 200 and a computer-readable medium.

Description

[Technical Field]

[0001] The present invention relates to a system, method, and computer-readable medium for evaluating bone volume based on MRI patient scan data and / or at least one set of composite CT data. [Background technology]

[0002] In recent years, attempts have been made to compensate for the inherent limitations of both magnetic resonance imaging (MRI) and computed tomography (CT) imaging by using both technologies in parallel and in combination, according to conventional techniques.

[0003] For example, bone marrow can be affected by various conditions such as post-traumatic bone contusions, diffuse tumor infiltration, or osteoporosis. To date, the main modalities for imaging these conditions have been MRI and CT, for example, in the context of radiotherapy treatment planning, PET attenuation correction, and bone quality assessment, with either one or both being applied in parallel. Therefore, techniques using CT imaging for bone mineral assessment, supported by enhanced image manipulation, have also been analyzed in several studies. - Quantitative bone mineral analysis using dual-energy computed tomography. Genant HK, Boyd D. Invest Radiol. 1977 Nov-Dec;12(6):545-51. PMID:591258 - Precise measurement of vertebral mineral content using computed tomography. Cann CE, Genant HK.J Comput Assist Tomogr. 1980 Aug;4(4):493-500. PMID:7391292 -Quantitative computed tomography for evaluation of spinal cord mineral content: Current status. Genant HK, Cann CE, Ettinger B, Gordan GS, Kolb FO, Reiser U, Arnaud CD. J Comput Assist Tomogr. 1985 May-Jun;9(3):602-4. PMID:11536558

[0004] On the other hand, the advantages of dual-energy computed tomography (DECT) have made CT imaging even more interesting as an additional aid to the conventional MRI diagnoses mentioned above. In particular, the use of DECT enables segmentation and visualization (color coding) of bone marrow based on material breakdown into bone marrow, calcium, water, adipose tissue, calcium hydroxyapatite, and collagen matrix. Therefore, the evaluation of bone mineral density using the DECT method has been investigated in several studies. -Bone mineral assessment: A new dual-energy CT technique. Nickoloff EL, Feldman F, Atherton JV. Radiology. 1988 Jul;168(1):223-8. PMID:3380964 - Evaluation of trabecular bone of the spine based on dual-energy CT. Wesarg S, Kirschner M, Becker M, Erdt M, Kafchitsas K, Khan MF. Methods Inf Med. 2012;51(5):398-405. doi:10.3414 / ME11-02-0034. Epub 2012 Sep 28. PMID:23038636 - Phantom-free in vivo 3D bone mineral density assessment based on dual-energy CT of the lumbar spine. Wichmann JL, Booz C, Wesarg S, Kafchitsas K, Bauer RW, Kerl JM, Lehnert T, Vogl TJ, Khan MF. Radiology. 2014 Jun;271(3):778-84. doi:10.1148 / radiol.13131952. Epub 2014 Jan 16. PMID:24475863

[0005] In a further step, the use of DECT for bone mineral assessment as described above was further studied in relation to phantom-less techniques. - Evaluation of lumbar spine bone mineral density using a novel phantom-less dual-energy CT post-processing algorithm in comparison with dual-energy X-ray absorptiometry. Booz C, Hofmann PC, Sedlmair M, Flohr TG, Schmidt B, D'Angelo T, Martin SS, Lenga L, Leithner D, Vogl TJ, Wichmann JL. Eur Radiol Exp. 2017;1(1):11.doi:10.1186 / s41747-017-0017-2.Epub 2017 Sep 20.PMID:29708178 -Eur Radiol.2015 Jun;25(6):1714-20.doi:10.1007 / s00330-014-3529-7.Epub 2014 Dec 7. Quantitative dual-energy CT for phantom-free evaluation of spongiform bone mineral density of the pedicle: correlation with pedicle screw pull-out strength. Wichmann JL1, Booz C, Wesarg S, Bauer RW, Kerl JM, Fischer S, Lehnert T, Vogl TJ, Khan MF, Kafchitsas K. -American College of Radiology(2013)ACR-SPR-SSR Practice Guideline for the Performance of Quantitative Computed Tomography(QCT)Bone Densitometry(Resolution 32).http: / / www.acr.org / ~ / media / ACR / Documents / PGTS / guidelines / QCT.pdf.Accessed July 2017

[0006] Furthermore, over the past few decades, the use of MRI has also increased significantly, for example, in supporting radiotherapy (RT) treatment planning, PET attenuation correction, and bone quality assessment. This increasing trend may be partly due to the superior soft tissue contrast of MRI compared to CT imaging.

[0007] However, considering the respective benefits and combinations of CT and MRI technologies for achieving better medical imaging results, using the two separate technologies in parallel would result in increased workload and additional costs for providing and operating each system.

[0008] Against this backdrop, for example, synthetic CT technology that enables a radiotherapy workflow using only MRI has been developed in recent years.

[0009] Synthetic CT provides images generated from MR scans that simulate CT images in terms of X-ray attenuation and electron density information. The images provide different Hounsfield units (HUs) for different materials, and these HUs can be converted to electron density in a treatment planning system (TPS) using a calibration curve. The purpose of synthetic CT images is to provide information on X-ray attenuation and electron density to obtain dose calculations with the same accuracy as those of CT images, enabling, for example, MR-only radiotherapy planning without the need for additional CT images.

[0010] However, so far, synthetic CT has not been able to provide the same level of information about examined tissue as DECT. Therefore, DECT analysis of tissue is superior to conventional synthetic CT analysis, particularly regarding tissue composition, such as the composition of bone structures. [Overview of the project]

[0011] The object of the present invention is to provide a system that can provide bone volume assessment for each individual patient based on each MRI data and / or synthetic CT data, thereby providing an improved and more efficient process in terms of cost and workload, by identifying additional information from MRI data and / or synthetic CT data resulting from specific data transformation and processing, in order to enable the evaluation of each individual patient using only a single medical imaging process for each individual patient. Furthermore, the object of the present invention is to provide a corresponding method and a computer-readable medium.

[0012] These problems are solved by the system according to claim 1, the method according to claim 13, and the computer-readable medium according to claim 15. Preferred embodiments are defined in the dependent claims respectively.

[0013] According to the present invention, a system for bone volume evaluation based on MRI patient scan data and / or at least one set of synthetic CT data is provided. The system includes - at least one processor module for handling and processing MRI patient scan data, and - at least one storage module for storing patient data, particularly MRI patient scan data, synthetic CT data, and / or related data of at least one first patient, and at least one data connection, particularly a bidirectional data connection, is provided between the processor module and the at least one storage module so that data can be transferred for processing and / or storage. The processor module further - receives at least one set of synthetic CT data of the first patient and / or receives and / or requests the MRI patient scan data of the first patient from the storage module or the MRI scan device, and is configured to apply a first transfer element to the MRI patient scan data of the first patient so that at least one set of synthetic CT data of the first patient is provided. - applies a second transfer element to at least one set of synthetic CT data and / or MRI patient scan data of the first patient so that reduced patient data of the first patient, particularly tissue volume information related to moisture and / or fat volume, etc., is generated, and / or is configured to receive tissue volume information, particularly moisture and / or fat volume related information, etc., from the MRI scan device. - is configured to apply a third transfer element to the reduced patient data and / or MRI patient scan data of the first patient so that total bone volume information is obtained. The processor module is configured to process and / or analyze / evaluate the MRI patient scan data, at least one set of synthetic CT data and / or reduced patient data, preferably in combination with each other, such that at least one bone volume parameter of a first patient, in particular bone mineral density, fat fraction, calcium concentration, total density, trabecular spacing, etc. is determined.

[0014] The present invention is based on the concept that additional information can be collected from each data set by the conversion and further evaluation of MRI patient scan data in order to enable a detailed analysis of bone volume parameters of bone composition, in particular bone mineral density, calcium concentration or further parameters related to bone composition.

[0015] Furthermore, since bone mineral density correlates with bone strength and thus, for example, the pull-out strength of pedicle screws, a further concept underlying the present invention is to optimize healthcare treatment, i.e. for the optimized placement of screws in the context of, for example, unstable vertebral fractures, existing spinal instability, degenerative scoliosis, spinal fixation in anterior strut grafting (see Eur Radiol. 2015 Jun;25(6):1714-20.doi:10.1007 / s00330-014-3529-7.Epub 2014 Dec 7. 「Quantitative dual-energy CT for phantomless evaluation of cancellous bone mineral density of the vertebral pedicle: correlation with pedicle screw pull-out strength」, Wichmann JL1, Booz C, Wesarg S, Bauer RW, Kerl JM, Fischer S, Lehnert T, Vogl TJ, Khan MF, Kafchitsas K).

[0016] However, since the original MRI imaging data does not provide such information in an obvious and immediate manner, the data collected during MRI imaging or by generating synthetic CT data must be further processed and / or analyzed, in particular to highlight various aspects specific to the corresponding bone composition, so that such further processed data, additional information about the bone structure of the first patient and its combination of compositions, can be obtained.

[0017] In particular, to achieve appropriate data processing starting from MRI patient scan data and / or synthetic CT data in order to gather sufficient information for the evaluation of bone structure, i.e., to evaluate bone volume parameters such as mineral density of a first patient and further relevant parameters of bone structure, multiple transfer elements may be provided by the present invention.

[0018] MRI patient scan data specifically refers to the data of the first patient acquired during a particular MRI scan. Therefore, MRI patient scan data can also refer to the specific body parts of the first patient that are focused on during the MRI scan process.

[0019] Alternatively, it may be possible that synthetic CT data for specific parts of the first patient's body is already provided.

[0020] In the context of the present invention, the processor module may be considered, for example, as a single processor unit, computer, etc., suitable for (appropriate) data processing and / or handling.

[0021] Furthermore, the storage module can store and read / transfer data, and therefore could be any type of storage device such as a solid-state disk, hard drive, or database for reading and writing data.

[0022] Furthermore, the processor module is configured to apply multiple, specifically three, different transfer elements for the purpose of appropriate data processing, particularly for the processing of appropriate MRI patient scan data, in order to provide information on bone volume parameters such as bone mineral density.

[0023] In particular, different transfer elements can be provided as data processing elements and / or algorithms to appropriately handle, analyze, and / or evaluate the data of each individual first patient, starting from the MRI patient scan dataset of each individual first patient. Thus, the synthetic CT data may be generated and reduced patient data that gives specific information about tissue volume, such as, for example, preferably, water and / or fat volume-related information and total bone volume information of the first patient, separated (or separable) from the fat volume information of the reduced patient data.

[0024] For example, in the context of the present invention, a single MRI patient dataset can form the basis of the entire process. These MRI patient datasets can preferably be obtained by a single MRI sequence. From such an MRI sequence, at least one 3D image dataset can be obtained that can later be used to generate at least one composite CT dataset. Furthermore, multiple different composite CT datasets can also be generated based on the MRI imaging sequences of individual patients.

[0025] In particular, the application of multiple transfer elements can be considered as step-by-step data so that the system's processor modules are configured appropriately.

[0026] In the case of MRI patient scan data, three transfer elements are applied in particular to generate appropriate information according to the present invention.

[0027] If synthetic CT data of the first patient is provided, in particular, two transfer elements, namely the second and third transfer elements, can be applied to generate appropriate information regarding the bone volume parameters of the first patient.

[0028] In particular, this stepwise data handling / processing makes it possible to extract bone volume information from the initially provided MRI patient scan data.

[0029] As a result, the present invention provides a more reliable and efficient process for identifying bone volume information, i.e., for bone volume assessment, based exclusively on MRI patient scan data.

[0030] According to a preferred embodiment, the first transfer element is learned based on CT patient scan data and MRI patient scan data of multiple patients to generate synthetic CT data. Furthermore, in other embodiments, the learning data for learning and / or adapting the first transfer element, in particular the CT patient scan data and / or MRI patient scan data of multiple patients, is stored in the system's storage module and / or received by the system from remote data storage.

[0031] In particular, the first transfer element can generate synthetic CT patient data based on MRI patient scan data. The first transfer element can be trained and optimized based on a dataset of multiple patients in order to improve the results provided by the first transfer element.

[0032] Therefore, MRI patient scans and corresponding CT patient scans can be used to provide configuration and / or adaptability to the first transfer element so that the first transfer element can generate synthetic CT patient data based exclusively on the MRI patient scan data of the first patient.

[0033] Such CT and / or MRI patient scan data from multiple patients can be stored in a storage module so that the processor module can request the learning information to configure and / or adapt the first transfer element.

[0034] Alternatively, the above data information for multiple patients, namely MRI patient scan data and corresponding CT patient scan data as training data, can also be stored and retrieved from remote data storage such as a central server.

[0035] According to other preferred embodiments, the first transfer element is provided in the form of a machine learning element / model, in particular a deep learning element / model, a neural network element, preferably a transformer, encoder-decoder, spread, U-net, V-net, recurrent neural network, recurrent interface network (RIN) or perseverer I / O or any other type of neural network of machine learning element / model.

[0036] Therefore, the first transfer element is learnable to be enabled and configured based on a given dataset, in particular based on MRI and CT patient scan data, to provide synthetic CT data of a first patient starting from corresponding MRI patient scan data of a first patient.

[0037] Furthermore, the first transfer element can similarly be provided in any other form of machine learning element.

[0038] In particular, quality and effectiveness can improve over time through the quality of the training data provided and the increasing quantity of this training data.

[0039] Therefore, the first transfer element may be further improved and configured to become more effective over time and provide improved quality results.

[0040] In a further embodiment of the present invention, CT patient scan data of multiple patients is provided in the form of two or more sets of CT patient scan data each referencing different energy levels, particularly dual-energy CT patient scan data or unenhanced multi-energy scan data.

[0041] In particular, the quantity and quality of information provided by CT patient scan data can be improved by providing multiple sets of CT patient scan data, i.e., sets acquired from the same patient at different energies. This information can also be used to generate at least one composite CT scan data set that references a specific tube voltage energy level. Furthermore, the above reference to a specific tube voltage energy level can also be achieved by a specific calibration process of the generated composite CT scan dataset, such that a specific composite CT scan dataset referencing a specific tube voltage energy is obtained as a result.

[0042] Therefore, the basis for the configuration of the first transfer element, i.e., the basis for each training data, may be provided with the above-mentioned additional multi-energy information to improve the configuration / adaptability of the first transfer element.

[0043] According to one embodiment, the second transfer element is provided in the form of an analytical element of type Dixon, IDEAL, MAGO, or MAGORINO, such as machine learning-based water-lipolysis, water-lipolysis based on in-phase and / or out-of-phase acquisition, etc., to identify fat and / or water volume-related information from MRI patient scan data and / or synthetic CT data.

[0044] In particular, the second transfer element preferably provides separation of fat and water content based on MRI patient scan data, especially from the first patient. Thus, the estimation of total bone volume can be improved by applying the second transfer element to the identification of fat volume (information).

[0045] Preferably, Dixon analysis may be applied as a second transfer element, namely, for example, echo asymmetry and the iterative breakdown of water and fat by least squares estimation (IDEAL).

[0046] Furthermore, the second transfer element may also be provided in the form of alternative techniques in specific backgrounds of water and / or fat volume-related information from the first patient's MRI patient scan data or synthetic CT data, such as acquisition in phase and out of phase, and water-lipolysis based on machine learning (e.g., U.S. Patent Application Publication No. 2021 / 0270919; Triay Bagur et al., "Magnitude-intrinsic water-fat ambiguity can be resolved with multipeak fat modeling and a multipoint search method", Magnetic Resonance in Medicine, Volume 82, Issue 1 p.460-475; Bray et al., MAGORINO: "Magnitude-only fat fraction and R*2 estimation with Rician noise modeling", Magnetic Resonance in Medicine, Volume 89, Issue 3 p.1173-1192; Liu et al., "Robust water-fat separation based on deep learning model exploring multi-echo nature of mGRE", Magnetic Resonance in Medicine, Volume 89, Issue 3 p.1173-1192; Liu et al., "Robust water-fat separation based on deep learning model exploring multi-echo nature of mGRE", Magnetic Resonance in Medicine, Volume 89, Issue 3 p.1173-1192; (See Resonance in Medicine, Volume 85, Issue 5, pp. 2828-2841).

[0047] In summary, bone volume assessment can preferably be performed based on at least one set of synthetic CT patient data generated using a transfer element learned from one type of CT scan data from multiple patients. Furthermore, fat percentage images can be provided obtained by using MRI-based water-fat separation transfer elements / algorithms such as DIXON, IDEAL, MAGO, MAGORINO, machine learning-based water-lipolysis, or acquisition-based water-lipolysis (based on in-phase and / or out-of-phase / opposite-phase echo times).

[0048] In a further embodiment of the present invention, a third transfer element is provided in the form of a Nickoloff model element for determining total bone volume, particularly based on MRI patient scan data and / or reduced patient data of a first patient.

[0049] In particular, the third transfer element can provide additional information regarding total bone volume in order to obtain appropriate results regarding bone volume parameters such as bone mineral density, fat percentage, calcium concentration, and total density intertrabecular space.

[0050] For this purpose, preferably, the Nickoloff model can be applied as a third transfer element in the context of the present invention.

[0051] The Nickoloff model, which is suitably applied in the background of the present invention, is also known from the publication "Bone mineral assessment: new dual-energy CT approach" Nickoloff EL, Feldman F, Atherton JV. Radiology. 1988 Jul;168(1):223-8. PMID:3380964.

[0052] Alternatively, the third transfer element can be provided in the form of other models or algorithms that enable appropriate data processing to allow for further identification of bone volume parameters based on MRI patient scan data and / or reduced patient data.

[0053] In particular, the Nickoloff model employs both water content and red bone marrow as a single fraction. On the other hand, for example, by combining it with a Dixon element as a second transport element, the red bone marrow fraction (especially water content) and the yellow bone marrow fraction (especially fat content) can be separated from each other, particularly in terms of comparing the data information provided by the second and third transport elements.

[0054] According to one preferred embodiment, MRI patient scan data is provided, in particular, by the application of (dual echo / multi echo) gradient echo sequences (ME-GRE), (turbo) spin echo sequences (TSE), dual echo steady state (DESS), multi echo steady state (MESS) sequences, equilibrium sequences (e.g., bSSFP), (multi echo) ultrashort echo time (UTE) sequences, zero TE (zTE) sequences, synthetic MRI sequences, MR fingerprinting-based sequences, MR-STAT-based sequences, etc., so that enhanced identification of the fat and / or bone volume of the first patient is provided by the transfer elements.

[0055] Therefore, in the context of the present invention, it is also possible to generate MRI patient scan data comprising sequences / sets that focus on different information due to the respective sequences to which they are applied.

[0056] The resulting MRI patient scan data, as a whole, can provide further information due to the application of the aforementioned specific sequencing techniques, such as multi-echo, or dual-echo GRE.

[0057] In other embodiments, MRI patient scan data and / or CT patient scan data are provided and / or stored in DICOM format, HDF5 format, NIfTI format, etc.

[0058] Therefore, each dataset is available in a standard format for medical imaging, ensuring maximum quality and availability.

[0059] Furthermore, according to a further embodiment of the present invention, the system comprises at least one MRI scanning device, and a data connection, in particular a one-way data connection, is provided between the MRI scanning device, at least one processor module and / or at least one storage module.

[0060] Therefore, the system according to the present invention may include an MRI scanning device itself that includes direct data connections to a processor module and / or a storage module.

[0061] Alternatively, the MRI scanning device that collects MRI patient scan data may be located remotely so that the corresponding MRI patient scan data can be provided to the system according to the present invention, for example, via a central server, remote data connection, etc.

[0062] As a result, the system according to the present invention does not necessarily need to have an MRI scanning device specifically assigned to it. Rather, the system according to the present invention can be provided in terms of an independent system, thereby providing MRI patient scan data from different MRI scanning devices via remote data connection.

[0063] Therefore, the system according to the present invention can be considered as one that does not include an MRI scanning device (its own / separate / individual), or as one in which an MRI scanning device is provided within the system according to the present invention.

[0064] In one embodiment, the system further comprises at least one calibration phantom element, which is applied once to the determination of a general calibration coefficient for each of the MRI patient scan data of multiple patients or for each of the MRI patient scan data of multiple patients, in order to identify individual calibration coefficients for multiple patients. Furthermore, according to one preferred embodiment, at least one calibration phantom element is provided in the form of multiple tubes and / or balls filled with a homogeneous or variable moisture-fat ratio, preferably in the range of 0 to 100 percent, particularly in the range of 20 to 80 percent.

[0065] Therefore, by providing standardized reference values ​​for the morphology of calibration phantom elements, particularly calibration phantom elements having multiple tubes / balls with varying / different water-fat ratios, the accuracy of at least one resulting bone volume parameter can be improved.

[0066] Furthermore, in a further embodiment, the present invention relates to a method for evaluating bone volume based on MRI patient scan data and / or at least one set of synthetic CT data by applying the system described in any one of claims 1 to 12, wherein the method is as follows: - A step of receiving and / or requesting at least one set of synthetic CT data for the first patient, or The steps include receiving and / or requesting MRI patient scan data of a first patient from a storage module or MRI scanning device, and applying a first transfer element to the MRI patient scan data of the first patient so that at least one set of composite CT data of the first patient is generated, - The steps of applying a second transfer element to at least one set of synthetic CT data and / or MRI patient scan data of the first patient so as to generate reduced patient data of the first patient that provides tissue volume information of the first patient, particularly water and / or fat volume-related information, and / or The process involves receiving tissue volume information, particularly water and / or fat volume-related information, from an MRI scanning device (130), - The step of applying a third transfer element to the reduced patient data and / or MRI patient scan data of the first patient so that total bone volume information is generated, Equipped with, To identify at least one bone volume parameter of the first patient, particularly bone mineral density, fat percentage, calcium concentration, total density, intertrabecular space, etc., based on the MRI patient scan data of the first patient, the MRI patient scan data, at least one set of synthetic CT data and / or reduced patient data are processed, analyzed and / or evaluated by a processor module, especially in combination with each other.

[0067] In particular, according to the method of the present invention, bone volume assessment can be performed based on at least one synthetic CT / at least one set of synthetic CT data generated using a method learned from CT scans of multiple patients.

[0068] Furthermore, fat percentage images obtainable using MRI-based water-lipolysis algorithm elements such as DIXON or IDEAL must be presented / provided.

[0069] Then, bone volume parameters such as mineral density and / or other related parameters can be obtained using the Nickoloff method or a process based on the Nickoloff method.

[0070] Finally, total bone volume (V TB The minimum parameters must be identified so that the CT value and fat volume (V) can be directly calculated, and the CT value and fat volume (V) must be identified. F Since both of ) are known, preferably only one equation should be used / solved.

[0071] Furthermore, the method according to the present invention is provided as something that can be executed by a system according to the present invention that is sufficiently and appropriately configured.

[0072] According to another embodiment of the present invention, the method is as follows: The steps include receiving MRI patient scan data from multiple patients, The steps include receiving CT patient scan data from multiple patients, The steps include: applying MRI patient scan data and CT patient scan data, preferably in the form of machine learning elements, deep learning elements, neural network elements, etc., to learn a first transfer element, thereby generating at least one set of synthetic CT data based on the MRI patient scan data of the first patient; To further prepare.

[0073] In particular, the method according to the present invention ensures that a first transfer element that generates / generates synthetic CT data for a first patient from MRI patient scan data of a first patient can provide synthetic CT data obtained as a result of appropriate and sufficient quality.

[0074] The above quality can be ensured according to the method of the present invention by appropriately training a first transfer element, which is preferably provided in the form of machine learning elements such as deep learning elements or neural network elements, with MRI patient scan data and corresponding CT patient scan data from multiple patients.

[0075] Therefore, by providing the above data for an increasing amount from multiple patients, the first transfer element can also be improved over time.

[0076] A further aspect of the present invention provides a computer-readable medium that includes instructions for causing at least one computer (processor) or the like to perform the method described in the present invention.

[0077] In particular, computer-readable media can be non-temporary computer-readable media.

[0078] Therefore, computer-readable and executable instructions can be stored on any type of storage medium, such as CDs, DVDs, flash memory devices, USB sticks, hard drives, solid-state disks, etc.

[0079] Further details and advantageous effects of the present invention are disclosed herein in conjunction with the drawings. [Brief explanation of the drawing]

[0080] [Figure 1] Figure 1 is a schematic diagram of an embodiment of a system for evaluating bone volume. [Figure 2] Figure 2 is a schematic flowchart of the method for evaluating bone volume. [Figure 3]Figure 3 is a schematic diagram of the method relating to Figure 2, including the system calibration step. [Modes for carrying out the invention]

[0081] Figure 1 shows a schematic diagram of an embodiment of System 100 for bone volume assessment, in particular for identifying at least one bone volume parameter such as bone mineral density, fat percentage, calcium concentration, total density, and intertrabecular space.

[0082] According to Figure 1, the system 100 comprises at least one processor module 110 and at least one storage 120.

[0083] Optionally, the system 100 may further include at least one MRI scanning device 130.

[0084] The processor module 110 and the storage module 120 are interconnected by a bidirectional data connection 140.

[0085] Furthermore, the optional MRI scanning device 130 can be connected to the processor module 110 and / or the storage module 120.

[0086] In particular, the data connection 140 with the MRI scanning device 130 may be provided as a one-way data connection 140.

[0087] Alternatively, the system 100 may be provided without at least one MRI scanning device 130.

[0088] In particular, one or more MRI scanning devices 130 may be located remotely, for example, while the MRI scanning devices 130 are data-connected to the system 100, and in particular to at least one processor module 110 and at least one storage 120 of the system 100.

[0089] Furthermore, as a further alternative, at least one processor module 110 and at least one storage module 120 may be implemented within at least one MRI scanning device 130.

[0090] With respect to the MRI scanning device 130, the system 100 can preferably be provided as an embedded system incorporated within the corresponding MRI scanning device 130.

[0091] Furthermore, the remote data storage and / or server 150 can be connected to the system 100, as shown in Figure 1.

[0092] In particular, the remote data storage 150 may include a data connection 140, preferably a bidirectional data connection 140, with respect to the processor module 110 of the system 100.

[0093] Figure 2 shows a flowchart of an embodiment of the corresponding method for evaluating bone volume.

[0094] In the first step, MRI patient scan data 212 of the first patient may be received and / or requested by the processor module 110 from the storage module 120, the MRI scanning device 130, or the remote data storage 150 (210).

[0095] Therefore, depending in particular whether the corresponding system 100 includes an MRI scanning device 130, the MRI patient scan data 212 may be directly transferred to the processor module 110 or requested / received from the storage module 120 or the (central) remote data storage 150.

[0096] In the next step, the first transfer element 220 is applied / used by the processor module 110 to the MRI patient scan data 212 so that the composite CT data 222 is generated, identified and / or configured.

[0097] In particular, machine learning elements such as neural networks may be used as a first transfer element 220 for converting MRI patient scan data 212 into synthetic CT data 222.

[0098] In another step of method 200, a second transfer element 230 is applied to the first patient's MRI patient scan data 212 and / or composite CT data 222 to generate, identify and / or constitute reduced patient data 232.

[0099] For example, the second transfer element 230 may be provided as a Dixon analysis element for identifying fat volume information, which can particularly optimize the subsequent estimation of total bone volume.

[0100] Therefore, as an alternative, it is also possible to apply machine learning-based water-lipolysis or other types of (analytical) sequencing, such as IDEAL, MAGO, MAGORINO, or acquisition-based water-lipolysis (based on in-phase and / or out-of-phase / opposite-phase echo times).

[0101] In a further step, the third transfer element 240 is applicable to the MRI patient scan data 212 and / or reduced patient data 232 of the first patient so that at least one total bone volume information 242 is generated, identified and / or configured.

[0102] The third transfer element 240 may preferably be provided in the form of a Nickoloff model so that the total bone volume information 242 is identifiable.

[0103] Based on this, at least one bone volume parameter can be identified / specified (250).

[0104] Therefore, in summary, the method 200 shown in Figure 2 can analyze and process MRI patient scan data 210;212 in such a manner that at least one bone volume parameter of the first patient, such as bone mineral density, fat percentage, calcium concentration, total density intertrabecular space, etc., can be identified (250).

[0105] Furthermore, the method 200 shown in Figure 2 also provides the basis for learning, i.e., constructing and adapting the first transfer element 220.

[0106] In particular, MRI patient scan data 202 and CT patient scan data 204 of multiple patients can be provided, for example, to the remote data storage / server 150 or storage module 120 of the system 100.

[0107] In the next step, the first transfer element 220 is learned (206) based on MRI and CT patient scan data 202;204 from multiple patients to generate improved synthetic CT data 222 from a single set of MRI patient scan data 210.

[0108] In particular, the first transfer element 220 can be learned once to achieve at least a temporarily fixed setting based on the currently available set of MRI and CT patient scan data 202;204 (206).

[0109] Alternatively, the first transfer element 220 may be learned temporarily, for example once a month, upon request, for example, when a large amount of additional new MRI / CT patient scan data 202;204 becomes available, or (generally) continuously, for example, whenever a new set of MRI / CT patient scan data becomes available (206).

[0110] Preferably, the first transfer element 220 may be provided in the form of a machine learning element.

[0111] For example, machine learning elements may be provided as deep learning elements, neural network elements, and so on.

[0112] Furthermore, in the context of the present invention, for example, a machine learning element such as the first transfer element 220 may be learned by / based on the output of a Nickoloff model element applied to CT patient scan data of multiple patients.

[0113] The quality of the composite CT data generated by the application of the first transfer element 220 can be improved in an efficient and beneficial manner based on MRI and CT patient scan data 202;204 of multiple patients.

[0114] In particular, according to embodiments of the method of the present invention, bone volume assessment can be performed based on synthetic CT data 222 of one synthetic CT / 1 set of a first patient, generated using a method which is learned from one type of CT scan received from multiple patients 202;204 (206).

[0115] Furthermore, for example, by applying the second transfer element 230, fat percentage images obtainable using MRI-based water-lipolysis algorithm elements such as DIXON, IDEAL, and machine learning elements must be presented / provided.

[0116] Furthermore, other relevant parameters related to bone mineral density and bone volume information can be obtained, for example, using the Nickoloff method in which a third transfer element 240 is applied.

[0117] Finally, total bone volume (V TB ) so that the CT value and fat volume (V F Since both of ) are known, only one equation must be used / solved (250).

[0118] More specifically, the above method according to the present invention can utilize the Nickoloff method based on the following formula with exemplary parameter values given in Koch et al., "Accuracy and precision of volumetric bone mineral density assessment using dual-source dual-energy versus quantitative CT": a phantom study, European Radiology Experimental (2021) 5:43 (https: / / doi.org / 10.1186 / s41747-021-00241-1) (240). [Number]

[0119] These exemplary equations (1) and (2) relate the exemplary intensities X 90 and X 150 (given in HU) in at least one (synthetic) CT series acquired at tube energies 90 and / or 150 kV to the fractional volume occupied by the matrix material (bone mineral + collagen) V TB and the volume of adipose tissue V F By this, the synthetic CT series / set is reconstructed to represent CT acquired at a specific tube energy, although not acquired at that specific tube energy.

[0120] On the other hand, regarding the present invention, the parameter V F is already available from an MRI patient scan (data) sequence for a first patient. Thus, according to the present invention, only one synthetic CT data set is required to solve the above equation correspondingly.

[0121] Furthermore, according to the present invention, in particular, it is possible to use tube energies of 20 to 150 kV, preferably 90 to 150 kV, for at least one, preferably two (synthetic) CT series.

[0122] The values ​​for t and g are 0.92 g / cm³, respectively. 3 and 1.02 g / cm³ 3 It is possible. On the other hand, other variables (μ 90 gamma 90 , β 90 , μ 150 gamma 150 , β 150 δ) is an energy-related constant, for example, parameter δ consists of a constant value of -1000 HU. Further variables can be collected from Table 1 below, consisting of parameter values ​​for different energy levels / photon energies. [Table 1]

[0123] By calculating the average strength of trabecular bone in both CT datasets, V TB and V F The following values ​​can be obtained. Finally, the bone mineral density (BMD) value ρBM (unit: g / cm³) 3 (given by) but the material constant l = 3.06 g / cm³ is given by equation (3). 3 And by applying λ=2.11, V TB It can be calculated from this.

number

[0124] Further parameters can be determined by equations (4) to (6).

number

[0125] Therefore, bone volume assessment, particularly of bone mineral density and / or other related bone volume parameters, can be identified and evaluated in an efficient manner, especially with respect to cost, time, and resources, by the method according to the present invention.

[0126] In particular, the results of the bone mineral density assessment described above may be considered as absolute values. However, if these values ​​are not fully calibrated, they may be considered as relative values, and they can still be very useful within the individual range. These relative values ​​may preferably correlate with a bone mineral density map, but may not be exactly equivalent to the reference values ​​described above.

[0127] Furthermore, according to other exemplary methods of the present invention, it is also possible to create / receive (one set / series) of synthetic CT data having a specific tube voltage. The water and / or fat content can be derived from the (corresponding) MRI data.

[0128] When the fat content and the number of CTs are known, the calculation of bone mineral density can be performed as follows:

[0129] For details, see CT count (CT n Generally, this can be determined based on a specific voltage by the following equation, where the parameter δ is a value of -1000 (i.e., a negative value) and the parameter ε refers to the number of water offsets.

number

[0130] Parameter V F and V TB It is generally unknown, and therefore V F This information can be collected, for example, from the results of Dixon analysis, IDEAL analysis, and other similar analyses.

[0131] Furthermore, the Ct parameter CT n This can be acquired from synthetic CT or conventional CT.

[0132] Therefore, based on equation (7), total bone volume V TB However, it can be calculated as follows.

number

[0133] Next, bone mineral density ρ BM This can be calculated in the same way as the method described by Nickoloff with respect to equation (9) as follows.

number

[0134] Therefore, the present invention also enables the identification and evaluation of bone mineral density based on the generated synthetic CT data in an efficient manner.

[0135] Figure 3 shows a schematic diagram of the method described in Figure 2, which includes a calibration step for the system 100, particularly the MRI scanning device 130.

[0136] In particular, the process shown in Figure 3 differs from the embodiment shown in Figure 2 by the learning of the first transfer element 206 and the calibration (260) of the system 100, preferably the MRI scanning device 130.

[0137] The method according to the present invention may include a learning step 206 of a first transfer element as shown in Figure 2, a calibration step 260 as shown in Figure 3, and / or the learning step 206 and the calibration step 260.

[0138] As shown in Figure 3, a calibration phantom element is provided (260), thereby enabling imaging and post-processing steps to more accurately generate images of fat volume, preferably in order to improve fat percentage estimation.

[0139] Calibration phantom elements may be supplied in the form of multiple tubes and / or balls (260). Different elements may have different moisture-fat percentages, preferably in the range of 0 to 100 percent, and particularly in the range of 20 to 80 percent. Calibration phantoms may comprise bone-mimicking materials such as hydroxyapatite, calcium, and collagen in concentrations ranging from 0 to 100 percent.

[0140] Furthermore, the resulting T1 and T2 signal weightings can be considered for the optimization of the resulting MRI patient scan data.

[0141] Based on the calibration of system 100 (270) using the provided calibration phantom element (260), MRI patient scan data of the first patient can be received / generated (210).

[0142] Alternatively, calibration of the system 100, particularly the MRI scanning device 130, may be provided in a later step of the method 200 shown in Figures 2 and / or 3, i.e., in terms of calibration of post-processing of MRI patient scan data.

[0143] In particular, calibration step 270 may be provided / performed / repeated within a (predetermined) time range, for example, once a day, once a week, once a month, once every six months, or it may be provided / performed for each patient individually.

[0144] Furthermore, one or more calibration phantoms may contain different target substances, such as calcium, collagen, etc.

[0145] The system 100, particularly the MRI scanning device 130 and / or image post-processing, can be calibrated in appropriate manner to enable optimized image data quality.

[0146] In summary, the present invention discloses, in particular, a system and method for enabling a single medical imaging technique, namely, bone volume assessment based on MRI patient scan data of a first patient.

[0147] Furthermore, by applying different transfer elements, particularly machine learning elements as the first transfer element, MRI patient scan data can be processed for the identification of specific information, such as fat volume and total bone volume information, in order to identify at least one bone volume parameter of a first patient, specifically by a combination of processed data. [Explanation of Symbols]

[0148] 100 Systems 110 Processor Module 120 storage modules 130 MRI scanning devices 140 Data Connections 150 Remote Data Storage / Servers 200 ways 202 Steps to receive MRI patient scan data from multiple patients 204 Steps to receive CT patient scan data from multiple patients 206 Learning Steps for the First Transfer Element 210 Step of receiving MRI patient scan data of the first patient 212 MRI patient scan data of the first patient 220 Step of applying the first transfer element 222 Composite CT Data 230 Step of applying the second transfer element 232 Patient data decreased 240 Step of applying the third transfer element 242 Total Bone Volume Information 250 Steps to identify bone volume parameters 260 Steps to provide phantom elements Steps to perform 270 system calibration

Claims

1. A system (100) for evaluating bone volume based on MRI patient scan data and / or at least one set of synthetic CT data, At least one processor module (110) for handling and processing MRI patient scan data, A storage module (120) for storing patient data, in particular MRI patient scan data, synthetic CT data and / or related data of at least one first patient, Equipped with, Between the processor module (110) and the at least one storage module (120), at least one data connection (140), in particular a bidirectional data connection (140), is provided so that data can be transferred for processing and / or storage. The processor module (110) further, The system receives at least one set of synthetic CT data from the first patient, and / or The system is configured to receive and / or request MRI patient scan data of the first patient from the storage module (120) or MRI scan device (130), and to apply a first transfer element to the MRI patient scan data of the first patient so that at least one set of composite CT data of the first patient is provided. A second transfer element is applied to at least one set of synthetic CT data and / or MRI patient scan data of the first patient, and / or a second transfer element is applied to the first patient's at least one set of synthetic CT data and / or MRI patient scan data, such that reduced patient data of the first patient is generated, which provides tissue volume information of the first patient, particularly information related to water and / or fat volume, and / or It is configured to receive tissue volume information, particularly water and / or fat volume-related information, from the MRI scan device (130). The system is configured to apply a third transfer element to the reduced patient data and / or MRI patient scan data of the first patient so that total bone volume information can be obtained. The processor module (110) is configured to process and / or analyze the MRI patient scan data, the at least one set of composite CT data and / or the reduced patient data, preferably in combination with each other, so that at least one bone volume parameter of the first patient, particularly bone mineral density, fat percentage, calcium concentration, total density intertrabecular space, etc., can be identified.

2. The system (100) according to claim 1, wherein the first transfer element is learned based on the CT patient scan data and MRI patient scan data of a plurality of patients to generate synthetic CT data.

3. The system (100) according to claim 2, wherein learning data of multiple patients for learning and / or adapting the first transfer element, particularly CT patient scan data and / or MRI patient scan data, are stored in the storage module (120) of the system (100) and / or received by the system (100) from remote data storage (150).

4. The system (100) according to any one of claims 1 to 3, wherein the first transfer element is provided in the form of a machine learning element, in particular a deep learning element, a neural network element, preferably a transformer, encoder-decoder, spread, U-net, V-net, recurrent neural network, recurrent interface network (RIN), or perseverer I / O type neural network, or any other type of machine learning element.

5. The system (100) according to any one of claims 2 to 4, wherein the CT patient scan data of multiple patients is provided in the form of two sets of CT patient scan data each referencing different energy levels, particularly dual-energy CT patient scan data or unenhanced multi-energy scan data.

6. The system (100) according to any one of claims 1 to 5, wherein the second transfer element is provided in the form of an analytical element of type Dixon, IDEAL, MAGO, MAGORINO, such as machine learning-based water-lipolysis, water-lipolysis based on in-phase and / or out-of-phase acquisition, for identifying fat and / or water volume-related information from the MRI patient scan data and / or the synthetic CT data.

7. The system (100) according to any one of claims 1 to 6, wherein the third transfer element is provided in the form of a Nickoloff model element for identifying total bone volume information, particularly based on the MRI patient scan data and / or reduced patient data of the first patient.

8. The MRI patient scan data is provided, in particular, by the application of a (multi-echo) gradient echo sequence, a (turbo) spin echo sequence, a dual-echo steady-state (DESS) multi-echo steady-state (MESS) sequence, an equilibrium sequence (e.g., bSSFP), a (multi-echo) ultrashort echo time (UTE) sequence, a zero TE (ZTE) sequence, a synthetic MRI (MR fingerprinting-based, MR-STAT-based), etc., such that the transfer element provides enhanced identification of the fat volume information of the first patient, according to the system (100) of any one of claims 1 to 7.

9. The MRI patient scan data and / or CT patient scan data are provided and / or stored in DICOM format, HDF5 format, NifTI format, etc., according to the system (100) of any one of claims 1 to 8.

10. The system comprises at least one MRI scanning device (130), The system (100) according to any one of claims 1 to 9, wherein the data connection unit (140), in particular the one-way data connection unit (140), is provided between the MRI scan device, the at least one processor module and / or the at least one storage module.

11. With additional proofreading phantom elements, The system (100) according to any one of claims 1 to 10, wherein the calibration phantom element is applied once to the determination of a general calibration coefficient for each of the MRI patient scan data of the multiple patients or for each of the MRI patient scan data of the multiple patients in order to identify individual calibration coefficients for the multiple patients.

12. The calibration phantom elements are preferably provided in the form of a plurality of tubes and / or balls filled with a homogeneous or fluctuating moisture-fat ratio within a moisture-fat ratio range of 20 to 80 percent, according to the system (100) of claim 10.

13. A method for evaluating bone volume based on MRI patient scan data and / or at least one set of synthetic CT data by applying the system (100) according to any one of claims 1 to 12, the following: To receive and / or request at least one set of composite CT data of the first patient, or The steps include receiving and / or requesting (210) MRI patient scan data (212) of the first patient from the storage module or MRI scan device, and applying a first transfer element (220) to the MRI patient scan data of the first patient so that at least one set of composite CT data (222) of the first patient is generated, The second transfer element (230) is applied to the at least one set of synthesized CT data (222) and / or MRI patient scan data (210) of the first patient so as to generate reduced patient data (232) of the first patient that provides tissue volume information of the first patient, particularly water and / or fat volume information, and / or The steps include receiving tissue volume information, particularly water and / or fat volume-related information, from the MRI scan device (130), The steps include applying a third transfer element (240) to the reduced patient data (232) and / or the MRI patient scan data (210) of the first patient so that total bone volume information (242) is generated, Equipped with, Method (200) to identify at least one bone volume parameter (250) of the first patient, particularly bone mineral density, fat percentage, calcium concentration, total density intertrabecular space, etc., based on the MRI patient scan data (210) of the first patient, the MRI patient scan data (210), the at least one set of composite CT data (222), and / or the reduced patient data (232), particularly combined with each other, and processed, analyzed, and / or evaluated by the processor module.

14. below, The steps include receiving MRI patient scan data (202) from multiple patients, The steps include receiving CT patient scan data (204) from multiple patients, The steps include: applying the MRI patient scan data (202) and the CT patient scan data (204) to the first transfer element (206), preferably in the form of deep learning elements, neural network elements, etc., to generate at least one set of synthesized CT data (222) based on the MRI patient scan data (210) of the first patient; The method according to claim 13 (200), further comprising the above.

15. A computer-readable medium comprising instructions for causing at least one computer or the like to perform the method according to claim 13 or 14.

Images