Peripheral perfusion measurements

Abstract

The present invention relates to peripheral perfusion measurement. To provide more detailed peripheral perfusion characteristics for better understanding of the current situation, a device 10 for peripheral perfusion measurement is provided, comprising an image data input 12, a data processor 14 and an output interface 16. The image data input receives at least one perfusion angiography 2D X-ray image of a region of interest of a subject's foot and a 3D foot model including spatial perfusion related parameters. The data processor registers the 3D foot model with the foot in the at least one perfusion angiography X-ray image. The registering step includes pose estimation of the foot in the 2D X-ray image. Based on the pose estimation, information is mapped between the 2D image and the 3D foot model. Image processing correction instructions are identified based on the mapped information. Furthermore, at least one image signal is modified based on the image processing correction instructions. The output interface provides at least one modified image signal. In a first embodiment, a partial perfusion analysis is performed. In a second embodiment, normalization of the perfusion signal by a proportion of perfused tissue is performed. In a third embodiment, reporting of the 3D model is performed.

Description

[Technical field]

[0001] The present invention relates to peripheral perfusion measurement, and in particular to a device for peripheral perfusion measurement, a system for peripheral perfusion measurement, and a method for peripheral perfusion measurement. [Background technology]

[0002] The treatment of peripheral arterial disease involves monitoring perfusion status during interventional procedures to confirm the effectiveness of the selected therapy. One method to monitor perfusion changes, e.g., before and after interventional procedures, is angiography, i.e., projection X-ray imaging. For example, perfusion characteristics of the limbs can be efficiently and quickly measured by 2D angiography. In addition to being readily available in the catheterization laboratory, angiography has the advantage of providing immediate, high-resolution perfusion information in real time, with moderate radiation exposure. However, projection-type angiography inherently loses depth information in the projection direction. Therefore, even slight changes in the position of the foot in 3D space (relative to the radiation source and detector) can cause potentially complex spatial distortions of the anatomical structures in the 2D image. On the other hand, 3D imaging, e.g., computed tomography (angiography) (CT(A)) and magnetic resonance (angiography) (MR(A)), are time-consuming and costly, and in the case of CT, high radiation exposure is required. Furthermore, 3D images are not available during treatment and may not be obtained prior to interventional treatment, at least for the foot, which is the present target area for perfusion analysis, where perfusion defects may manifest as wounds, gangrene, ulcers, etc. However, quantitative comparisons between different regions on the same digital subtraction angiography (DSA) sequence or between different DSA sequences have been shown to be difficult with only measurements of accumulated absorbed X-ray direction. Summary of the Invention [Problem to be solved by the invention]

[0003] Therefore, to appreciate the current situation, it is necessary to provide detailed peripheral perfusion characteristics. [Means for solving the problem]

[0004] The object of the invention is solved by the subject matter of the independent claims, further embodiments are contained in the dependent claims. It should be noted that the aspects of the invention described below also apply to a device for peripheral perfusion measurement, a system for peripheral perfusion measurement and a method for peripheral perfusion measurement.

[0005] According to the present invention, there is provided a device for measuring peripheral perfusion. The device comprises an image data input, a data processor, and an output interface. The image data input is configured to receive at least one perfusion angiography two-dimensional x-ray image of a region of interest of a subject's foot. The image data input is also configured to receive a three-dimensional foot model comprising spatial perfusion related parameters. The data processor is configured to register the three-dimensional foot model with the foot in the at least one perfusion angiography x-ray image, the registration including a pose estimation of the foot in the two-dimensional x-ray image. The data processor is configured to map information between the two-dimensional image and the three-dimensional foot model based on the pose estimation. The data processor is also configured to identify image processing modification instructions based on the mapped information. The data processor is further configured to modify at least one image signal based on the image processing modification instructions. The output interface is configured to provide at least one modified image signal.

[0006] As a result, improved image data is provided that can be used for further analysis steps, leading to improved analysis results.

[0007] The term "spatial perfusion related parameters" relates to information of perfusion characteristics or perfusion properties related to a particular region or part of the 3D foot model. One example of spatial perfusion related parameters is information of the distribution ratio of bone and tissue structures in certain parts of the foot and the assumed blood perfusion in these parts. Another example of spatial perfusion related parameters is information on the classification of tissue parts that are recognized as well perfused and other tissue parts that are recognized as poorly perfused according to their tissue type and location. A further example of spatial perfusion related parameters is the perfusion area. A further example of spatial perfusion related parameters is the discrimination of different anatomical structures such as skin, tissue, muscle, bone, etc.

[0008] The term "spatial" in "spatial perfusion-related parameters" refers to the relative location in the spatial reference system of the 3D foot model. In other words, "spatial" refers to the location and orientation within the 3D foot model.

[0009] The expression "mapping information" refers to the transfer of spatially allocated information from one of the 2D image and the 3D foot model to the other of the 2D image and the 3D foot model. The term "information" refers to, for example, data, measurements, correction factors, classifications, area subdivisions, tissue types, bone structures, etc.

[0010] The term "image signal" relates to a signal constituting the basis of an image provided, e.g. presented, to a user. In one example, the image signal relates to a signal constituting a two-dimensional image (i.e. a two-dimensional X-ray image of a region of interest of a subject's foot). In another example, the image signal relates to a signal constituting a representation of a three-dimensional foot model.

[0011] In one example, the spatial perfusion related parameters include information about spatial location specific perfusion characteristics, ie, information about perfusion characteristics specific to spatial locations of the three-dimensional foot model.

[0012] In one example, the information mapped between a two-dimensional image (ie, a two-dimensional x-ray image of a region of interest of a subject's foot) and a three-dimensional foot model relates to peripheral perfusion.

[0013] In the first basic example, information is mapped from a 3D foot model to a 2D image.

[0014] In the second basic example, information is mapped from a 2D image onto a 3D foot model.

[0015] In a third basic example, information is mapped from the 2D image onto the 3D foot model and also from the 3D foot model back to the 2D image.

[0016] Based on the mapped information, for example, image processing modification instructions are identified for modifying an image signal of a two-dimensional image. In another example, image processing modification instructions are identified for modifying an image signal displaying a three-dimensional foot model. In a further example, image processing modification instructions are identified for modifying an image signal of a two-dimensional image and also for modifying an image signal of a display of the three-dimensional foot model.

[0017] Information mapping refers to the transfer of information between two worlds, a 2D image and a 3D foot model, in a spatially registered manner in order to use the information of one world in the respective other world. This makes it possible to give augmented data that combines the information of both worlds. Thus, by modification of the respective image signals, it is possible to provide this augmented result of data processing through an output, for example to a user.

[0018] In a first basic example, the two-dimensional image is corrected based on information from a three-dimensional foot model, for example spatial perfusion-related parameters of the three-dimensional foot model.

[0019] In a second basic example, the 3D foot model is modified based on information from the 2D images, for example perfusion measurements in the 2D images.

[0020] In a third basic example, the two-dimensional image is modified based on information from the three-dimensional foot model, e.g., spatial perfusion related parameters of the three-dimensional foot model, and the three-dimensional foot model is modified based on information from the two-dimensional image, e.g., perfusion measurements in the two-dimensional image.

[0021] In one example, at least one perfusion angiography two-dimensional x-ray image of a region of interest of a subject's foot includes information related to the perfusion present or prevailing in the region of interest of the subject's foot, and the three-dimensional foot model including spatial perfusion-related parameters includes information related to typical perfusion characteristics in the foot.

[0022] According to one example, for the mapping, the data processor is configured to map the perfusion regions from the three-dimensional foot model onto the two-dimensional X-ray image and project the perfusion regions in two dimensions. For the identification of the at least one image processing correction instruction, the data processor is also configured to sum the perfusion signal in each of the projected perfusion regions, and the data processor is configured to fit the perfusion regions to the estimated posture. Furthermore, for the correction, the data processor is configured to generate a set of time-density curves corresponding to these fitted perfusion regions.

[0023] According to one example, for the mapping, the data processor is configured to perform ray tracing. Furthermore, for the identification of the at least one image processing correction instruction, the data processor is configured to extract a correction factor from the three-dimensional model. Furthermore, for the correction, the data processor is configured to apply the extracted correction factor to at least one signal of the two-dimensional X-ray image. For providing the at least one corrected image signal, the data processor is also configured to report the corrected image signal in the image or to provide a quantity extracted from the at least one corrected image signal.

[0024] According to one example, for the mapping, the data processor is configured to perform ray tracing of at least one patch of pixels of the two-dimensional X-ray image and map the respective image signals to the perfusion regions in the three-dimensional model that contributed to the patch of the two-dimensional X-ray image. For the identification of the at least one image processing correction instruction, the data processor is configured to store the perfusion signals or extracted quantities of the two-dimensional X-ray image in the three-dimensional model. For the correction, the data processor is configured to generate a report by three-dimensional rendering of the three-dimensional model using the stored perfusion signals.

[0025] According to one example, the 3D foot model is an articulated model of the foot that can be adapted to a given bone constellation of the foot.

[0026] According to one example, the perfusion related parameters include perfusion analysis related three-dimensional structures including at least one of the skin of the foot, the parenchymal tissue below the skin, the bone, the division of the foot into clinically relevant regions, and the major blood vessel groups of the foot.

[0027] According to one example, the data processor is configured to calculate, as perfusion analysis, a regional perfusion analysis of the digital subtraction angiography sequence, taking into account distortions of the projection images for a given posture.

[0028] According to one example, a visualization unit is provided comprising a graphical user interface configured to present results of the data processor via the output interface. In one option, the graphical user interface is configured to compare perfusion parameters of different digital subtraction angiography sequences.

[0029] According to the invention there is also provided a system for peripheral perfusion measurement, comprising an X-ray imaging device, a display device and a device for peripheral perfusion measurement according to one of the above examples. The X-ray imaging device is configured to generate at least one perfusion angiography X-ray image of a region of interest of the subject's body. The display device is configured to display at least one modified image signal.

[0030] There is also provided in accordance with the present invention a method for measuring peripheral perfusion, the method comprising: - receiving at least one perfusion angiography two-dimensional x-ray image of a region of interest of the subject's foot and receiving a three-dimensional foot model including spatial perfusion related parameters; - registering the 3D foot model with the foot in at least one perfusion angiography X-ray image, the registering including pose estimation of the foot in the 2D X-ray image; - mapping information between the 2D image and a 3D foot model based on the pose estimation; - identifying image processing modification instructions based on the mapped information; - modifying at least one image signal based on image processing modification instructions; - providing at least one modified image signal; has.

[0031] According to one embodiment of the method, the mapping step comprises mapping the perfusion regions from the 3D foot model onto the 2D X-ray image and projecting the perfusion regions in two dimensions. Furthermore, identifying the at least one image processing correction instruction comprises summing the perfusion signal in each of the projected perfusion regions, such that the perfusion regions are adapted to the estimated pose. Furthermore, the correcting step comprises generating a set of time-density curves corresponding to these adapted perfusion regions.

[0032] According to one embodiment of the method, the mapping step comprises ray tracing. Furthermore, the identifying at least one image processing modification instruction step comprises extracting a modification factor from the three-dimensional model. The modifying step comprises applying the extracted modification factor to at least one signal of the two-dimensional X-ray image. The providing at least one modified image signal step comprises reporting the modified image signal in the image or providing a quantity extracted from the at least one modified image signal.

[0033] According to one embodiment of the method, the mapping step comprises ray tracing at least one patch of pixels of the two-dimensional X-ray image and mapping each image signal to a perfusion region in the three-dimensional model that contributed to the patch of the two-dimensional X-ray image. Furthermore, the identified image processing correction instructions comprise storing the perfusion signals or extracted quantities of the two-dimensional X-ray image in the three-dimensional model. Furthermore, the correcting step comprises generating a report by three-dimensional rendering of the three-dimensional model using the stored perfusion signals.

[0034] According to an embodiment of the method, the three-dimensional foot model is a deformable three-dimensional foot model. For the mapping step, a step of detecting a foot pose in at least one X-ray image is performed. For the identifying step, a step of determining foot joint parameters based on the detected foot pose is performed. For the modifying step, a step of adapting the three-dimensional foot model and the spatial perfusion related parameters based on the detected pose and based on the determined joint parameters is performed, and a step of providing at least one X-ray image and the adapted spatial perfusion related parameters for further calculating a perfusion analysis for at least a portion of the subject's foot is performed. For the modifying step, a step of presenting the result of the perfusion analysis as a two-dimensional image is performed. Optionally, a step of performing a perfusion analysis for at least a portion of the subject's foot based on the at least one X-ray image and the adapted three-dimensional foot model is further performed. In one option, the analyzing step is performed by obtaining a time-density curve based on the adapted spatial perfusion related parameters.

[0035] According to one embodiment of the method, the perfusion related parameters include i) a perfusion analysis relevant three-dimensional structure comprising at least one of the skin of the foot, the parenchymal tissue below the skin, the bone, a division of the foot into clinically relevant regions and groups of major blood vessels, and / or ii) a perfusion division of the foot comprising regions for regional perfusion analysis.

[0036] According to an embodiment of the method, the 3D foot model is a deformable 3D foot model. For the mapping step, a step of detecting a foot pose in at least one X-ray image is performed. For the identifying step, a step of determining foot joint parameters based on the detected foot pose is performed. For the modifying step, a step of adapting the 3D foot model and the spatial perfusion related parameters based on the detected pose and based on the determined joint parameters is performed, and a step of providing the at least one X-ray image and the adapted spatial perfusion related parameters for further computing a perfusion analysis for at least a part of the subject's foot is performed.

[0037] According to one embodiment of the method, the perfusion analysis comprises a regional perfusion analysis of a Digital Subtraction Angiography (DSA) sequence that takes into account distortion of the projection images for a given posture.

[0038] According to one aspect, a device for peripheral perfusion measurement is provided, comprising an image data input, a data processor, and an output interface. The image data input receives at least one perfusion angiography two-dimensional x-ray image of a region of interest of a subject's foot and a three-dimensional foot model including spatial perfusion related parameters. The data processor registers the three-dimensional foot model with the foot in the at least one perfusion angiography x-ray image. The registering includes a pose estimation of the foot in the two-dimensional x-ray image. Information is mapped between the two-dimensional image and the three-dimensional foot model based on the pose estimation. Image processing correction instructions are identified based on the mapped information. Furthermore, at least one image signal is corrected based on the image processing correction instructions. The output interface provides the at least one corrected image signal. In a first embodiment, a partial perfusion analysis is provided. In a second embodiment, a normalization of the perfusion signal by a proportion of perfused tissue is performed. In a third embodiment, a report in the three-dimensional model is performed.

[0039] According to one embodiment, the foot model is provided as a generic 3D foot model, so that no model generation is required during the interventional treatment, the generic 3D foot model is provided and calculated in advance.

[0040] According to one embodiment, normalized signals are provided and assigned to two-dimensional image pixels for further two-dimensional image processing and presentation.

[0041] According to one embodiment, the normalized signals are provided and assigned to a three-dimensional model, which is then used for further three-dimensional image processing and presentation.

[0042] According to one embodiment, an analysis step of the 2D angiogram data is performed to compensate for the loss of depth information, while simultaneously pose- and posture-matching the angiogram with a 3D model containing internal parameters is performed. Advantageously, there is no need to acquire 3D images of the same patient.

[0043] According to one embodiment, a mechanism is provided to obtain time-density curves that are comparable within a patient, i.e. pre / post interventional treatment, and between patients. The time-density curves are normalized by considering the percentage of the parenchymal tissue that is penetrated as well as the 2D area of ​​the region of interest in one frame. The time-density curves can quantify the degree of comparability between two recorded curves. Furthermore, the time-density curves do not require 3D image data of the patient during development. The time-density curves also do not require ground truth of the pose of the 2D angiographic images during development.

[0044] According to one embodiment, the loss of depth information is partially compensated for by incorporating prior knowledge in the form of a generic 3D foot model and estimating the individual pose of this reference model for each angiography sequence, parameterized by the camera parameters of the projection as well as the model internal degrees of freedom, flexion, abduction / adduction, and anatomical variations.

[0045] The effect is that, to a certain extent, the measurement signal becomes less sensitive to changes in posture or position.

[0046] As an added benefit, regional perfusion measurements are reported on the angiograms of the 3D models by ray tracing and material decomposition. For example, 3D rendering of these results can help clinicians interpret the regional quantitative results and can be used for intra- and inter-patient comparisons. Furthermore, 3D rendering can also be used to guide the operator to interactively optimize the patient-detector positioning during post-interventional fluoroscopy to best match the corresponding pre-interventional fluoroscopy setup.

[0047] According to one aspect, the curve is provided in a normalized manner so that the clinician can compare the perfusion measurements with those observed in the past, regardless of the size or location of the region of interest, i.e. so that the clinician can interpret them based on experience. In simple terms, the contrast agent concentration per area of ​​the imaging plane takes into account the anatomy and the posture. Furthermore, obtaining this curve allows for quantitative intra- and inter-patient comparisons, i.e. the analysis ensures that the curve is calculated from the same region of different images.

[0048] These and other aspects of the invention will be apparent from and elucidated with reference to the embodiment(s) described hereinafter.

[0049] Exemplary embodiments of the present invention will now be described with reference to the following drawings, in which: [Brief description of the drawings]

[0050] [Figure 1] FIG. 1 shows a schematic diagram of an embodiment of a device for measuring peripheral perfusion. [Diagram 2] FIG. 1 shows an embodiment of a system for measuring peripheral perfusion. [Diagram 3] FIG. 1 illustrates the basic steps of one embodiment of a method for peripheral perfusion measurement. [Figure 3a] FIG. 1 shows a first further option of the method, performing a partial perfusion analysis. [Figure 3b] FIG. 13 shows a second further option of the method, where normalization of the perfusion signal is performed by the proportion of perfused tissue. [Figure 3c] FIG. 4 shows a third further option of the method for generating a report on a three-dimensional model. [Figure 4] FIG. 1 is an explanatory diagram of two different bends superimposed on one another in three-dimensional models. [Diagram 5] FIG. 2 illustrates a foot model with optional foot sections. [Figure 6]FIG. 1 shows a sequence of stages of perfusion analysis corresponding to input x-ray images, 3D foot model for registration, and superimposed anatomical regions on the piecewise map. [Figure 7] FIG. 13 is a schematic diagram illustrating the concept of an embodiment of peripheral perfusion measurement that decomposes transmitted material into tissue and bone and maps perfusion parameters onto a 3D foot model for normalizing the X-ray image signal. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0051] Next, specific embodiments will be described in more detail with reference to the accompanying drawings. In the following description, similar drawing reference numbers are used for similar elements even in different drawings. Matters defined in this specification, such as detailed configurations and elements, are provided to assist in a comprehensive understanding of the exemplary embodiments. Also, well-known functions or configurations will not be described in detail to obscure the embodiments due to unnecessary details. Furthermore, an expression such as "at least one of," when preceding a list of elements, modifies the entire list of elements and does not modify individual elements of the list.

[0052] FIG. 1 shows a device 10 for measuring peripheral perfusion. The device 10 comprises an image data input 12, a data processor 14, and an output interface 16. The image data input 12 is configured to receive at least one perfusion angiography two-dimensional x-ray image of a region of interest of a subject's foot. The image data input 12 is also configured to receive a three-dimensional foot model including spatial perfusion related parameters. The data processor 14 is configured to register the three-dimensional foot model with the foot in the at least one perfusion angiography x-ray image. The registering includes a pose estimation of the foot in the two-dimensional x-ray image. The data processor 14 is also configured to map information between the two-dimensional image and the three-dimensional foot model based on the pose estimation. The data processor 14 is further configured to identify image processing modification instructions based on the mapped information. The data processor 14 is also configured to modify at least one image signal based on the image processing modification instructions. The output interface 16 is configured to provide at least one modified image signal.

[0053] The term "pose" relates to the positional relationship between the imaging arrangement and the object being imaged, i.e. the foot. The term pose thus includes at least one aspect or measure of the group of projection geometry, foot posture and further foot-related parameters. The term "pose estimation" thus relates to an estimation, or so to speak an inferential determination, of the projection geometry, foot posture and / or further foot-related parameters.

[0054] The term "projection geometry" refers to the "camera" parameters of an X-ray system, such as the geometrical setup and positional relationship of the X-ray source and X-ray detector in relation to the imaged subject, i.e., the foot. The term "projection geometry" also refers to internal and external camera parameters, i.e., internal and external radiography parameters. Projection geometry may therefore also be referred to as projection parameters, camera projection geometry, camera setup or projection setup. Projection geometry, in simple terms, refers to how an imaging system is configured.

[0055] The term "foot posture" refers to a particular arrangement, i.e. position and posture, of the foot. Posture reflects the kinematic state of the foot and defines which degrees of freedom of movement, such as the bending radius of the joints, are used and the resulting spatial anatomical arrangement of the foot. Foot posture refers to the anatomical joints present in a 2D X-ray image. Foot posture is also called "posture". Posture, in simple terms, refers to the anatomical "setup" of the foot.

[0056] The term "foot-related parameters" relates to further characteristic parameters of the foot, such as the morphology and shape of the individual bones.

[0057] The projective geometry is also called the camera pose, or simply pose. The term "pose estimation" refers to estimating both "pose" and "body orientation".

[0058] The term "joint" is used for the position of the foot in a 3D model. A joint is one of several internal parameters of the model, for example the shape of a bone.

[0059] In an embodiment, optionally provided in Fig. 1, a display 18 is provided for displaying the modified image signal. The data processor 14 provides the modified image signal as display data to a display in data connection to the output interface 16. The output of the display data and other data is indicated by a first dashed arrow 20. The input of image data, e.g. a perfusion angiography 2D X-ray image or a 3D foot model, is indicated by a second dashed arrow 22. Frame 24 illustrates the option of providing the image data input 12, the data processor 14 and the output interface 16 in an integrated manner, e.g. in a common housing. In an option, the image data input 12, the data processor 14 and the output interface 16 are provided as separate arrangements. In an option, the 3D foot model is provided by a data memory, e.g. in a data storage (not further shown).

[0060] In one embodiment, a display configured to display results of the analysis and comparison of perfusion parameters of various digital subtraction angiography sequences is provided.

[0061] In another embodiment, a visualization unit is provided which presents the results of the data processor 14 via an output interface 16 and optionally includes a graphical user interface adapted to compare perfusion parameters of different digital subtraction angiography sequences (see also FIG. 2 ).

[0062] The perfusion angiography device 10 is sometimes referred to as a perfusion imaging device.

[0063] "Image data input 12" refers to the supply of image data. The image data input 12 thus forms an interface for inputting image data. The image data input 12 may also be called an image data source, an image data supply, a data input, a data supply unit, a data input unit or a data supply module. The image data input 12 may be connected to a data storage or to an imaging source such as an X-ray imaging device.

[0064] The "data processor 14" relates to image data processing. The data processor 14 may also be called a processor, a processing unit, or a computing module.

[0065] The "output interface 16" relates to providing the processing result of the image data. The output interface 16 may also be called an output unit or an output module.

[0066] "Image processing modification instructions" relates to providing operating instructions, image data manipulation instructions, modification factors, correction calculations, and the like.

[0067] In one embodiment, the image data input 12 is configured to receive a sequence of perfusion angiographic x-ray images of a region of interest of a subject's body.

[0068] FIG. 2 shows an embodiment of a system 100 for peripheral perfusion measurement. The system 100 includes an X-ray imaging device 102. The X-ray imaging device 102 is configured to provide at least one perfusion angiography X-ray image of a region of interest of a subject's body. As an example, the X-ray imaging device 102 is shown as a movable C-arm structure 104 mounted on a ceiling, with an X-ray source 106 and an X-ray detector 108 at the end of the C-arm. For illustrative purposes, a subject 110, i.e. a subject, is also shown supported by a patient support 112 for examination, interventional, other treatment or imaging purposes. The system 100 further includes a display device 114, for example in the form of a monitor device shown movably suspended from the ceiling. The display device 114 is configured to display at least one modified image signal. The system 100 also includes an embodiment 116 of a device for peripheral perfusion angiography according to one of the preceding and following embodiments. Additionally, again for purposes of illustration, a control console with a monitor 118 and other user interface devices such as a mouse, keyboard, control panel, touch pad, etc. is shown on the right. A peripheral perfusion angiography device 116 is built into the control console. Line 120 indicates a data connection (wired or wireless) between the x-ray imaging device 102 and the peripheral perfusion angiography device 116.

[0069] In one embodiment, the display device 114 is configured to display the results of the perfusion analysis. It is noted that the display device 114 may be configured to display various outputs depending on the embodiments described above and below.

[0070] 3 shows the basic steps of one embodiment of a peripheral perfusion measurement method 200. The method 200 comprises the following steps: In a first step 202, at least one perfusion angiographic two-dimensional X-ray image of a region of interest of a subject's leg is received. - In a second step 204, a three-dimensional foot model comprising spatial perfusion related parameters is received. - In a third step 206, the 3D foot model is registered with the foot in at least one perfusion angiography X-ray image, the registration including a pose estimation of the foot in the 2D X-ray image. In a fourth step 208, information between the 2D image and the 3D foot model is mapped based on the pose estimation. In a fifth step 210, image processing modification instructions are identified based on the mapped information. - In a sixth step 212, the at least one image signal is modified based on the image processing modification instructions. In a seventh step 214, at least one modified image signal is provided. The two registration steps, the first and second step, may be performed simultaneously or in any other order.

[0071] Mapping, the fourth step 208, may involve mapping 2D to 3D, for example signal data from a perfusion angiography 2D x-ray image to a 3D foot model.

[0072] The mapping, fourth step 208, may also include mapping 3D to 2D, for example segmenting a 3D foot model to a perfusion angiography 2D x-ray image.

[0073] The mapping, or fourth step 208, may further include mapping two dimensions to three dimensions and three dimensions to two dimensions.

[0074] Optionally, as indicated by dashed frame 216, the modified image signal, ie the results of the perfusion analysis, is displayed.

[0075] This method of peripheral perfusion measurement may also be referred to as a method of perfusion angiography or a method of perfusion imaging. This method allows for adapted perfusion analysis, resulting in more accurate information.

[0076] In one embodiment, registering the three-dimensional foot model with the foot in at least one perfusion angiography x-ray image includes estimating anatomical joint parameters together with geometric x-ray imaging parameters.

[0077] In one embodiment of the method, a sequence of perfusion angiographic x-ray images of a region of interest in a subject's body is provided.

[0078] In one embodiment, the sequence of perfusion angiographic x-ray images of a region of interest of a subject's body is provided as a sequence of perfusion angiographic x-ray images of a region of the subject's leg.

[0079] The term "peripheral perfusion" relates to blood supply in the vasculature of peripheral parts of a subject's body, i.e. peripheral body regions, such as the foot and hand. In the case of the foot, the three-dimensional model is a three-dimensional foot model, and in the case of the hand, the three-dimensional model is a three-dimensional hand model. Blood supply includes blood flow, i.e. blood flow rate, in larger vessels, e.g., arteries and veins. Blood supply also includes flow and thus supply in smaller vasculature, e.g., to provide an adequate blood supply to a tissue region or area.

[0080] The term perfusion refers to the supply of blood to a tissue structure.

[0081] In one example, the region of interest is a peripheral body part, for example a distal limb part, such as a region of the foot, hand, lower arm or lower leg.

[0082] A perfusion angiogram x-ray image sequence is provided as a sequence of images showing the flow of contrast agent through the vasculature.

[0083] "Perfusion angiography x-ray image" refers to an x-ray image obtained upon application of a contrast agent that allows visualization of vascular structures within a subject.

[0084] A "region of interest on a subject's body" refers to a part of a subject's body that is subject to examination or treatment.

[0085] Image processing modification instructions may also be referred to as image modification processes or image modification procedures. Image processing modification instructions may relate to modification factors or to predefined tasks to be performed.

[0086] In one embodiment, the results of the pose estimation are used to map information from three dimensions to two dimensions. In another embodiment, the results of the pose estimation are used to map information from two dimensions to three dimensions. In a further embodiment, the results of the pose estimation are used to map information from two dimensions to three dimensions and from three dimensions to two dimensions.

[0087] For example, the mapping may include mapping at least one image signal of a two-dimensional x-ray image onto a three-dimensional foot model.

[0088] In one embodiment, the results of the pose estimation are used to map information from 3D to 2D X-ray image signals, which are image values, which may be compensation factors or predefined perfusion regions.

[0089] In one embodiment, the image signals are not mapped into three dimensions, but the results of the pose estimation are used to map the information from three dimensions to two dimensions.

[0090] For example, identifying the image processing correction instructions may be based on the two-dimensional signal mapped to the three-dimensional foot model. For example, the image processing correction instructions may include at least one correction factor.

[0091] In one embodiment, the pose estimation includes a pose measurement.

[0092] The identifying includes identifying at least one image processing modification instruction.

[0093] In one embodiment, a regional perfusion analysis is performed.

[0094] 3a illustrates a first further option of the method for performing a partial perfusion analysis. As shown, mapping includes mapping 220 perfusion regions from a 3D foot model onto a 2D X-ray image and projecting 222 the perfusion regions in two dimensions. Identifying image processing correction instructions includes summing 224 the perfusion signal in each of the projected perfusion regions, where the perfusion regions are adapted to the estimated pose. Correcting includes generating 226 a set of time-density curves corresponding to these adapted perfusion regions.

[0095] As an example, following pose estimation, the perfusion regions are mapped from the 3D model to the image. The perfusion signal is, for example, summed in each of the projected perfusion regions fitted to the measured pose. As a result, a set of time-density curves corresponding to these regions is given, along with potentially extracted quantities, such as peak heights, widths, etc.

[0096] In one embodiment, the modifying comprises extracting quantities such as peak height, width, etc. for the fitted perfusion regions.

[0097] In one embodiment, normalization of the perfusion signal by the percentage of perfused tissue is performed.

[0098] 3b illustrates a second further option of the method, normalizing the perfusion signal by the effective proportion of perfused material. As shown, mapping includes ray tracing 228. Identifying image processing correction instructions includes extracting 230 a correction factor from the three-dimensional model. Correcting includes applying 232 the extracted correction factor to at least one signal of the two-dimensional X-ray image. Providing at least one corrected image signal includes reporting 234 the corrected image signal in the image or providing 236 a quantity extracted from the at least one corrected image signal.

[0099] In one embodiment, pose estimation is followed by ray tracing, for example to extract correction factors from the 3D model without mapping the signal onto the 3D model. The correction factors are applied to the signal. The results are reported in images or extracted quantities, for example time-density curves.

[0100] 3c illustrates a third further option of the method, generating a report in a three-dimensional model. As shown, mapping includes ray tracing 238 at least one patch of pixels of the two-dimensional X-ray image and mapping 240 the respective image signals to perfusion regions in the three-dimensional model that contributed to the patch of the two-dimensional X-ray image. Furthermore, the identified image processing modification instructions include storing 242 the perfusion signals or extracted quantities of the two-dimensional X-ray image in the three-dimensional model. Further, modifying includes generating 244 the report by three-dimensional rendering of the three-dimensional model using the stored perfusion signals.

[0101] In one embodiment, pose estimation is followed by ray tracing, and the signal is mapped to the perfused region in the 3D model that contributed to the pixel or patch of pixels in the 2D image. The perfusion signal (or extracted quantities) are stored in the 3D model. The results are reported through 3D rendering and / or by comparison with other similar reports.

[0102] In one embodiment, the correction factors are extracted from the 3D model without mapping the signals of the 2D X-ray image onto the 3D model.

[0103] In one embodiment, a time-density curve extracted from at least one modified image signal is provided.

[0104] In one embodiment, reporting is done on a three-dimensional model.

[0105] In one embodiment, revising includes comparing the report to other similar reports.

[0106] In one embodiment, not shown in detail, for mapping, data processor 14 is configured to map the perfusion regions from the three-dimensional foot model onto a two-dimensional X-ray image and project the perfusion regions in two dimensions. For identification of at least one image processing correction instruction, data processor 14 is further configured to sum the perfusion signal in each of the projected perfusion regions. Data processor 14 is configured to fit the perfusion regions to the estimated posture. For correction, data processor 14 is configured to generate a set of time-density curves corresponding to these fitted perfusion regions.

[0107] In one embodiment, not shown in detail, for the mapping, the data processor 14 is configured to perform ray tracing. For the identification of the at least one image processing correction instruction, the data processor 14 is also configured to extract a correction factor from the three-dimensional model. For the correction, the data processor 14 is further configured to apply the extracted correction factor to at least one signal of the two-dimensional X-ray image. For the provision of the at least one corrected image signal, the data processor 14 is further configured to report the corrected image signal in the image or provide a quantity extracted from the at least one corrected image signal.

[0108] In one embodiment, not shown in detail, for the mapping, the data processor 14 is configured to perform ray tracing of at least one patch of pixels of the two-dimensional X-ray image and map the respective image signals to the perfusion regions in the three-dimensional model that contributed to the patch of the two-dimensional X-ray image. For the identification of at least one image processing correction instruction, the data processor 14 is configured to store the perfusion signals or extracted quantities of the two-dimensional X-ray image in the three-dimensional model. For the correction, the data processor 14 is configured to generate a report by a three-dimensional rendering of the three-dimensional model using the stored perfusion signals.

[0109] In one embodiment, a device for measuring peripheral perfusion is provided, comprising an image data input, a data processor, and an output interface. The image data input is configured to receive at least one perfusion angiography two-dimensional x-ray image of a region of interest of a subject's foot, and to receive a three-dimensional foot model including spatial perfusion related parameters. The data processor is configured to register the three-dimensional foot model with the foot in the at least one perfusion angiography x-ray image, map at least one image signal of the two-dimensional x-ray image to the three-dimensional foot model, identify at least one correction factor based on the two-dimensional signal mapped to the three-dimensional foot model, and correct the at least one image signal based on the at least one correction factor. The output interface is configured to provide the at least one corrected image signal.

[0110] In one embodiment, the 3D foot model is a deformable 3D foot model, and the data processor 14 - detecting a posture of the foot in at least one X-ray image for mapping; - determining foot joint parameters based on the detected foot pose for identification; - optionally adapting the bone shape of the 3D foot model, e.g. to take into account anatomical variations between different subjects; - adapting the three-dimensional foot model and the spatial perfusion related parameters based on the detected posture and based on the determined joint parameters for the correction and providing the at least one X-ray image and the adapted spatial perfusion related parameters for further calculating a perfusion analysis for at least a portion of the subject's foot. The present invention is configured to:

[0111] Optionally, the data processor 14 is configured to perform a perfusion analysis of at least a portion of the subject's foot based on the at least one X-ray image and the fitted three-dimensional foot model.

[0112] As a further option, for analysis, the data processor 14 is configured to obtain time-density curves based on the fitted spatial perfusion-related parameters.

[0113] In one embodiment, not shown in detail, the 3D foot model is an articulated model of the foot that can be adapted to a given bone arrangement in the foot.

[0114] The term "articulated model" refers to a model that exhibits similar degrees of freedom of movement as the respective anatomical structures of the foot, i.e., mimicking the possible positions of the foot in motion and in various postures.

[0115] In one example, not shown in detail, the perfusion related parameters include perfusion analysis related three-dimensional structures including at least one of the skin of the foot, the parenchymal tissue below the skin, the bone, the division of the foot into clinically relevant regions, and major blood vessel groups.

[0116] In one embodiment, a perfusion section of the foot is provided, including regions for regional perfusion analysis.

[0117] In one embodiment, not shown in detail, the data processor 14 is configured to calculate as perfusion analysis a partial perfusion analysis of a digital subtraction angiography sequence taking into account distortion of the projection images for a given posture.

[0118] The term digital subtraction angiography (DSA) refers to images obtained by subtracting a non-contrast background from an image of an injected contrast agent.

[0119] In one embodiment of the device, the data processor 14 is configured to report regional perfusion parameters extracted from digital subtraction angiography sequences in the three-dimensional foot model as a perfusion analysis. In one embodiment, not shown in detail, the data processor 14: - ray tracing image points of at least one perfusion angiography two-dimensional x-ray image for mapping, determining possible ray paths based on the ray tracing, and identifying intersections of the determined possible ray paths with the three-dimensional foot model; - assessing a portion of the perfusionable material of the three-dimensional foot model based on the identified intersection points to identify, and determining a transfer factor based on the assessed portion; - spatially transferring values ​​of the two-dimensional x-ray image to a three-dimensional foot model based on the transfer factor to correct the perfusion values ​​and generating an augmented three-dimensional foot model with the transferred perfusion values; - Present an extended 3D foot model with transferred perfusion values ​​to provide The present invention is configured to:

[0120] In one embodiment, to weight the perfusion measurements, the data processor 14 is configured to estimate posture parameters, identify at least one ray for the image points, calculate intersection points with the 3D foot model, decompose the intersection points into bones and structures to identify tissue region contributions, and normalize the portions to blood perfused tissue.

[0121] In one embodiment, the image data input is configured to receive a sequence of perfusion angiographic x-ray images of a region of interest of the subject's body.

[0122] The signals of the two-dimensional image are transferred to a three-dimensional model, for example by ray tracing. Signal contribution fraction factors are determined by evaluating the model. The results are then used to reweight the signal values ​​of the two-dimensional image. Values ​​(also called quantities) are obtained and normalized to the fraction of perfusable tissue of the foot.

[0123] The term "perfusable" refers to a material that is capable of exhibiting blood perfusion, such as a tissue material, as opposed to a material that is not perfusable, such as bone (compared to tissue). However, it should be noted that bone is of course also supplied with blood. Blood is supplied to and within bone, and the relative proportions of blood in bone and, for example, parenchymal tissue in bone, can vary significantly. In this sense, normalization can take these differences into account.

[0124] In the providing step, the at least one X-ray image and the matched spatial perfusion related parameters are provided as enhanced perfusion X-ray image data.

[0125] In one embodiment, the registration includes an estimate of the position of the detector relative to the subject, hi another embodiment, the registration includes an estimate of the foot joint parameters, e.g., flexion angle, and the contribution of the anatomical morphology.

[0126] The phrase "receiving at least one x-ray image of a region of the subject's foot" may also be expressed as "providing at least one x-ray image of a region of the subject's foot." The at least one x-ray image is provided to a further method step.

[0127] The expression "receiving a deformable three-dimensional foot model comprising spatial perfusion related parameters" may also be expressed as "providing a deformable three-dimensional foot model comprising spatial perfusion related parameters". The three-dimensional foot model is provided to a further method step.

[0128] The phrase "providing at least one X-ray image and the matched spatial perfusion related parameters for calculating a perfusion analysis for at least a portion of the subject's foot" may also be expressed as "outputting at least one X-ray image and the matched spatial perfusion related parameters for calculating a perfusion analysis for at least a portion of the subject's foot."

[0129] In one embodiment, the term clinically relevant region relates to the blood supply area of ​​a major blood vessel, for example an angiosome.

[0130] Diagnosis or evaluation in different regions allows for the signal to be collected in a targeted manner. The signal density of the contrast agent is acquired over time.

[0131] In one embodiment, the three-dimensional foot model includes a foot perfusion section that includes the region of the regional perfusion analysis.

[0132] In one embodiment, the region is a region of interest for regional perfusion analysis.

[0133] In one embodiment, the 3D foot model has a structure divided into a bone portion and a tissue portion.

[0134] When the 3D foot model is fitted, the segmental structures are displaced and therefore deformed. The segmental structures may be provided as a 2D map, i.e., 2D data, which is applied, i.e., transformed, into the 3D world of the 3D foot model. After fitting in the 3D world, the information is transferred back to 2D by mapping the information into the 2D world of at least one X-ray image used for perfusion analysis.

[0135] In one embodiment, the segments are integrated into a three-dimensional model, for example by providing overlapping regions within the projection layers in which signals can be weighted.

[0136] The information is given by a 3D model but is for 2D use.

[0137] In the embodiment of the foot model, angiographic perspective distortion and postural variations are realized in the perfusion analysis of the DSA sequence. Depth information lost during projection imaging is partially restored by a 3D foot model aligned to a given X-ray image. Such a model incorporates structures relevant for analyzing perfusion. These structures include the foot skin, the parenchymal tissue below the skin, all bones, clinically relevant regions such as the vascular territories of the major vessels, subdivisions of the foot such as the angiosomes, and the major vessels.

[0138] In order to fit the 3D foot model to a given X-ray image or to X-ray images of an X-ray image sequence, the model is equipped with the relevant degrees of freedom of movement and deformation. For the analysis of foot angiograms, flexion (plantar and dorsal flexion) is incorporated since it constitutes a mode of movement that causes a relatively large postural variability. The joints of the 3D foot model also include other modes such as adduction and abduction. Furthermore, anatomical variability (e.g. bone and skin shape) is statically incorporated by vertex-displacement vectors obtained, for example, from a Procrustes analysis of a set of foot meshes taken from different patients. Mathematically, it is assumed that the discussed degrees of freedom are described by the mapping of certain independent model internal parameters, e.g. flexion angles, which are a mixture of certain anatomical modes of the Procrustes analysis, onto the 3D mesh, for example by adapting the vertex positions of a reference mesh.

[0139] In one embodiment, a 3D foot model is computed by segmenting the relevant structures (skin, bone, ...) in a set of CT or MR images (corresponding to one or more patients), meshing these segmentations, registering them to a fixed reference, analyzing the residual variability, and modeling it, for example, by mesh blending, similarity transformation, or statistical analysis (e.g. Procrustes) and dimensionality reduction. The 3D images are necessary for the initial modeling of the anatomical shape, shape variability, and dynamics. They are not required for the later development of the algorithms and therefore may be obtained from other clinical research fields unrelated to perfusion analysis and peripheral vascular disease (e.g. orthopedics).

[0140] In one embodiment of registration, various techniques are provided to register a given X-ray image (e.g., angiographic fluoroscopic frame) to a 3D model. Some of these methods iteratively solve an optimization problem in which the 3D foot model projected onto the X-ray detector plane finds the optimal pose that best matches the given X-ray image features with some given metric. These image features include, for example, detected intensity patterns, contours or landmarks. In contrast to solutions that compute 3D pose information from optical camera images, the required registration module cannot rely on the assumption that the detected patterns or landmarks are present on the object's surface. This makes standard triangulation approaches in particular useless. In that case, the corresponding features must be derived from the 3D foot model (e.g., by projecting the model's landmarks onto the detector plane). In some embodiments, the X-ray image feature detection model is trained on the 3D model or synthetic X-rays obtained from the underlying 3D image (e.g., CT).

[0141] Instead of solving such optimization problems, another example aims to directly map 2D images to pose parameters (and possibly even to a 3D mesh) using forward models. Often, these models operate directly on 2D images and therefore require labeled pairs (2D image, pose parameters / 3D mesh) during the training phase. These data requirements are cost-prohibitive to meet, unless 2D X-ray images are synthetically derived from 3D images (e.g. CT volumetrics) by projection with known sampled camera parameters.

[0142] "Time-density curve" refers to a curve showing the evolution of a contrast agent over time as it flows through a particular location in an image.

[0143] Perfusion analysis is suitable for example for outcome prediction and outcome assessment: perfusion can therefore be compared at pre- and post-intervention stages.

[0144] The 3D model can be adapted to the real foot by deformation. Size adaptation is also possible. The model is adapted for angiography.

[0145] In a further embodiment of the method, the perfusion analysis includes reporting regional perfusion parameters extracted from the digital subtraction angiography sequences in the three-dimensional foot model.

[0146] In one embodiment of the analysis, after estimating the projection pose and associated model parameters for a given angiogram, the corresponding DSA sequence can be analyzed in both the projection plane and the 3D foot model.

[0147] According to one embodiment of the invention, in a first possible sub-step of the analysis, a normalization of the extensive perfusion parameters by the local tissue proportions is performed. The signal of a DSA image of one pixel is constituted by the sum of the absorption of the contrast medium along the X-ray direction (i.e. the logarithm of the concentration). Therefore, all extensive quantities directly obtained from the DSA image / sequence, such as the time-density curve of a region, inherit this summation. Since the blood vessels and parenchymal tissues penetrated by the X-rays of interest mainly contribute to the local signal, such extensive quantities are sensitive to the relative proportions of penetrated bone and parenchymal tissue, limiting the quantitative comparability of extensive quantities in different regions of the foot within the same DSA image / sequence. Since the relative proportion of penetrated bone also depends on the camera pose, this sensitivity affects the quantitative comparability of DSA images / sequences (e.g. pre- and post-interventional angiograms).

[0148] This mitigates the lack of quantitative comparability by ray tracing between different regions or (small) pose variations, as shown in Fig. 6 and Fig. 7. With the estimation of the camera pose parameters as well as the model internal parameters, the 3D foot model is articulated and deformed accordingly. For each pixel of interest in the DSA image / sequence, the corresponding ray of the X-ray source is traced and its intersection with the outer surface of the foot model is determined. The transmission path length L is then measured. Similarly, the total transmission path length LB of the bones intersecting with this ray is calculated. The local fraction of the transmitted bone is finally compensated by multiplying the quantity of extensiveness by a position and pose dependent factor (1-LB / L). In a more sophisticated normalization, the weighting factor is generated by estimating the typical perfusion intensity IP of the parenchyma and the typical perfusion intensity IB of the bone and weighting them by the local fraction of transmitted bone and parenchyma, resulting in an overall weighting factor IP(1-LB / L)+IB·LB / L. For more efficient calculations, this compensation factor is also calculated for a small patch of pixels in the DSA image / sequence by intersecting the 3D foot model with a cone formed by the x-rays of the pixels outside this patch.

[0149] This normalization procedure applies to intensive quantities, i.e. quantities whose values ​​increase proportionally to the size of the domain they are evaluated / accumulated in. No normalization is applied to intensive quantities such as durations, e.g. arrival times.

[0150] In a second possible sub-step of the analysis, a perfusion analysis of the pose-adapted regions is performed. Although angiography provides a comprehensive view of the perfusion of the whole foot, information compression is desirable for the presentation of data to the clinician, for comparison of perfusion (e.g. before and after interventional treatment), for example for score extraction for interventional treatment outcome evaluation, and for inter-patient comparison. Compression of the spatial information of angiography can be achieved by evaluating the perfusion in well-defined perfusion regions, for example by summing the DSA signals of these regions and displaying them as time-density curves (or displaying characteristic parameters of these curves). Such a definition for a given X-ray image cannot be adequately defined on the projection images alone, since a simple transfer (e.g. by image-to-image registration) to another X-ray image with a different camera pose would not correspond to the same volume of the foot. A purely two-dimensional definition of the perfusion region does not guarantee an anatomical correspondence when transferred to another X-ray image. To mitigate the lack of anatomical correspondence, the following is done (see Fig. 6 for illustration):

[0151] With the camera pose and model internal parameters estimated, a given section of the 3D leg model into meaningful perfusion regions motivated by angiosomes, e.g., the blood supply regions of major vessels, can be projected onto the angiogram. With the DSA signal normalized according to the above-mentioned embodiment, the contrast agent concentration can be summed in the projected perfusion regions. Here, care must be taken in handling the overlapping regions of two adjacent perfusion regions (see Fig. 6), since the measured contrast agent concentration needs to be properly distributed between the adjacent perfusion regions.

[0152] The method to weight the measured contrast agent concentration consists in ray tracing (see above) and taking as a weighting factor the ratio of the transmission path length in the tissue of the adjacent perfused area. This setup assumes spatial homogeneity of the perfusion in the tissue intersected by the X-ray, which is compromised by the large vessels intersecting the perfused area. Therefore, this approach is beneficial when applied to the microvascular perfusion signal, which can be automatically separated from the large vessel perfusion in DSA images / sequences.

[0153] The contribution of large vessels to the overlap of perfusion regions can be appropriately weighted in a dedicated manner.

[0154] Major vessels can be segmented and labeled in the DSA images / sequences, and detected major vessels can be matched to their 3D counterparts using prior knowledge, such as the average vascular tree of a 3D foot model, so that their signals can be uniquely assigned to one and only one of the perfusion regions.

[0155] Given the camera pose and model intrinsic parameters, as well as a segmentation of the major vessels in the angiogram, the 3D vascular tree is deformed to best match the segmentation. This adapted 3D vascular tree can be used either to assign vessels in the projection plane to specific perfusion regions (see above) or to individually adjust the perfusion region boundaries in the 3D model, so that each region is supplied by only one major vessel.

[0156] Additionally, the overlapping areas of adjacent perfusion regions are also used to generate confidence scores and error bars for the time-concentration curves, which reflect the relative portion of the overlapping regions to the non-overlapping regions and reflect the residual uncertainty due to the lack of information on the distribution of microvascular perfusion in the projection direction.

[0157] The 3D foot model is used to identify the radiation path through the object in order to normalize the detected signal of the 2D X-ray image.

[0158] Starting from the 2D image, ray tracing is performed for each pixel: Which X-ray radiation path leads to the respective pixel? Where does the radiation path pass through the 3D foot model? Instead of radiation paths, a pyramid is generated, i.e. a pyramid space starting from the image point is expanded to show possible origins of the X-ray radiation. In one embodiment, for each patch of image points, a truncated pyramidal radiation path is generated. This is also called ray tracing. A patch can refer to the size of one or several pixels in the 2D image.

[0159] The overlap is determined using the volume of the 3D foot model, where a layer model can be used: the 3D foot model provides a structural distinction between areas where perfusion may be excluded and areas where perfusion is expected, so that the pyramid-like shape of the radiation path overlaps with the structure of the 3D foot model, making it possible to determine the number of areas where perfusion is expected at a particular image value.

[0160] The signal (of the 2D patch) is divided into parts suitable for perfusion, thus obscuring the signal for this region, and the resulting value is therefore given a larger value since the otherwise missing depth information is taken into account at least in the fundamental aspect of how much perfusable tissue has actually been passed or penetrated.

[0161] In one embodiment, before actually registering the 2D image and the 3D model, an attempt is made to place or pre-position them with a similar spatial rotation as a sort of preparatory step in order to achieve improved alignment and therefore improved ray tracing.

[0162] In other words, it is done to normalize the image values.

[0163] Materials that typically exhibit perfusion, such as blood vessels and tissue, i.e., portions of the foot, may be referred to as perfusable materials. Materials that typically exhibit less perfusion, such as bone, i.e., portions of the foot, may be referred to as non-perfusable materials.

[0164] The angiographic signal is thus captured in three dimensions, i.e. mapped, for correction.

[0165] In one embodiment, the three-dimensional foot model is augmented with signals from the two-dimensional image to generate a three-dimensional perfused foot image.

[0166] For example, a first three-dimensional perfused foot image is calculated based on the subject's two-dimensional image data before treatment, resulting in a three-dimensional perfused foot pre-image. The three-dimensional perfused foot pre-image is sometimes referred to as a three-dimensional perfusion pre-model. A second three-dimensional perfused foot image is calculated based on the subject's two-dimensional image data after treatment, resulting in a three-dimensional perfused foot post-image. The three-dimensional perfused foot post-image is sometimes referred to as a three-dimensional perfusion post-model. The three-dimensional perfused pre-model and the three-dimensional perfusion post-model can be co-registered to allow for better comparison between the pre-treatment and post-treatment conditions. Optionally, a user can toggle between the two three-dimensional perfusion models to visualize the results in three dimensions.

[0167] Briefly, normalization of 2D image information is performed by registering 2D X-ray image signals with a 3D model. In a first option, the normalization is used for visualization of (2D) angiograms. Pose is estimated for the 2D normalization. In a second option, the normalization is used for reporting in the context of the (3D) model.

[0168] The spatial allocation to the model strengthens the signal, which is thus transferred to the next level.

[0169] A further option, for example in a third possible sub-step of the analysis, is the reporting of angiographically based perfusion measurements in the 3D foot model. Elucidating the sub-volumes of the 3D object that contribute to the local signal in the projection images is performed for a complete interpretation. Applying this inference to new images for comparison with previous images (e.g. before and after interventional treatment, different patients) is tedious and requires significant mental effort.

[0170] These issues can be addressed by mapping the angiographic information onto a 3D foot model. If such a mapping is established for other imaging modalities, their measurements can be directly compared to the angiographic measurements in 3D space. Furthermore, reporting angiographic measurements allows for a direct comparison of different DSA images / sequences in 3D, taking into account variations in camera pose and joints (e.g., flexion).

[0171] A 2D to 3D mapping for such angiographic reporting can be done as follows (see FIG. 7): After estimating the camera pose and the model internal parameters, the x-rays of all pixels are traced. If the quantities to be reported are extensive, the normalization scheme of the first possible substep of the analysis described above (normalization of extensive perfusion parameters by local tissue ratios) can be applied first. The traced x-rays are then thickened into small rods. The rods of x-rays originating from adjacent pixels of the angiogram should not cross each other. This can be achieved by choosing cylinders of sufficiently small radius as rods. An even better construction involves tracing four x-rays that end at the vertex of the pixel under investigation, and using as rods the pyramid formed by the pixel area, the four x-rays and the focal point. The latter configuration results in a space-filling partitioning of the 3D foot model into rods (at least for a given angiographic field of view). For each pixel of the angiogram (sequence), the corresponding (potentially normalized) perfusion parameters are assigned to all positions in the 3D foot model where the corresponding rod overlaps with tissue. This mapping procedure may also be applied to small patches of angiograms (rather than pixel-based), in which case the pyramid-rod structure may be applied directly to the corners of the patch.

[0172] In this mapping, the local information obtained from angiography is smeared along the x-ray direction, so that the contribution of major vessels to the local DSA signal is also distributed to the corresponding rods, which may lead to misinterpretation.

[0173] Using the separation of microvessels and macrovessels in the local DSA signal and prior knowledge of the 3D foot model (either an average vascular tree or a fit of the vascular tree that best matches the angiographically segmented vessels), it is possible to assign microvascular properties to the parenchyma and macrovascular properties to the vascular tree of the 3D foot model.

[0174] Reporting the angiographic data in this way results in fitted 3D models (e.g., with tortuosity adjusted to best match the angiogram). To better compare two different angiograms (pre-intervention vs. post-intervention, different patients), the fitting of these foot models can be reversed, e.g., by mesh-based registration or morphing the fitted foot models to a standard configuration of model internal parameters (e.g., average anatomical shape, reference tortuosity, etc.).

[0175] According to one embodiment of the method, for mapping, - ray tracing image points of at least one perfusion angiography 2D X-ray image; - determining possible radiation paths based on ray tracing; - identifying points of intersection of the determined possible radiation paths with the three-dimensional foot model; will be carried out.

[0176] In addition, for identification purposes, - evaluating the portion of perfusable material of the three-dimensional foot model based on the identified intersection points; - determining a transfer factor based on the evaluated portions; will be carried out.

[0177] Furthermore, for correction, - spatially transferring values ​​of the two-dimensional x-ray image to the three-dimensional foot model based on a transfer factor to generate an augmented three-dimensional foot model with the transferred perfusion values. will be carried out.

[0178] In addition, for the provision of - presenting the augmented three-dimensional foot model with the transferred perfusion values. will be carried out.

[0179] In a further embodiment of the method, further weighting the perfusion measurements may include: - estimating pose parameters; - identifying at least one X-ray for an image point and calculating an intersection point with a 3D foot model; - Decomposing the intersection points into bones and structures to identify the contribution of tissue regions; - Normalizing the sections to blood-perfused tissue; will be carried out.

[0180] The amount of tissue contributing to the signal is thus determined. By determining the length of the radiation passing through the foot and the length through the portion of the tissue that can be perfused by blood, the ratio of each is given.

[0181] Alternatively, starting from an image, a pyramid or cone-like structure representing each image ray used in the image is identified and it is determined how the pyramid intersects the 3D foot model.

[0182] The data, i.e. signals, are transferred from 2D to a 3D model. The features seen in 2D are transferred to the 3D world of the model. The model allows this to be transferred to the actual perfused tissue. This can be used for reporting.

[0183] In a further embodiment of the method, the three-dimensional foot model is an articulated model of the foot that can be adapted to a predetermined bone arrangement of the foot.

[0184] In one embodiment, the model is configured with a predefined degree of plantar flexion and dorsiflexion.

[0185] In one embodiment, the 3D foot model statistically represents anatomical variability.

[0186] In one embodiment of the method, fitting the three-dimensional foot model is performed by fitting a model bone structure to a predetermined bone arrangement in at least one X-ray image.

[0187] In a further embodiment of the method, steps of displaying the analysis results and comparing the perfusion parameters of different digital subtraction angiography sequences are performed.

[0188] As an example, a pre-interventional / post-interventional angiogram is shown.

[0189] In one embodiment of visualization, the results of the analysis unit may be displayed to the user in a variety of ways, as explained in the order of the analysis aspects above.

[0190] The first possible sub-step of visualization involves the visualization of tissue ratio normalized extensive perfusion parameters. The result of the normalization step described above can be visualized in the form of time-density curves corresponding to manually or automatically selected regions. Alternatively, the normalized regional perfusion parameters can be presented to the clinician as an overlay on the angiogram, in which case a toggle user interface for switching between the overlay of "raw" and tissue ratio normalized perfusion parameters is useful for building confidence in this regional normalization.

[0191] In a second possible sub-step of the visualization, the visualization of the analysis in the pose-adapted perfusion regions is performed. This analysis step aims at a partial measurement of anatomically well-defined perfusion characteristics and can be presented in the form of a visualization of time-density curves of a given region and of the inferred parameters (e.g. peak height, peak time). To link these aggregated information to the anatomical, such visualization is accompanied by a three-dimensional rendering of an articulated model of the foot and its segments. The rendering camera can be adapted to the estimated pose and the projection of the segments onto the angiogram is performed, for example in the form of a colored overlay. Clicking on a segment on the angiogram triggers the selection of this perfusion region and the corresponding update of the time-density curve plot.

[0192] A third possible sub-step of the visualization involves the reporting of perfusion parameters in the 3D model. After mapping the angiographically measured perfusion characteristics onto the 3D foot model in a space-filling manner, the results can be presented to the clinician in the form of a 3D rendering, where the values ​​of the perfusion parameters (e.g. tissue-normalized time-concentration peak) are color-encoded. Besides the 3D rendering with trackback navigation, orthogonal cuts of the model (e.g. cuts parallel to the sagittal plane) are also displayed. The 2D angiograms are further integrated into the 3D plot as a plane texture oriented and placed at the actual X-ray detector position. Furthermore, the projection direction of a given field of view of the 3D foot model is displayed to the user on request in the form of several arrows pointing from a position in the foot model to the direction of the detector pixel onto which this position is projected. To compare the perfusion of two angiograms (e.g. of different patients or before and after an interventional treatment), the graphical user interface is equipped with a button that, when clicked, triggers the registration of the two fitted foot models. The user then toggles between the two angiographic perfusion measurements, for example while keeping the 3D rendering camera fixed.

[0193] Thus, the embodiments described herein support both intra-patient (e.g., pre-intervention vs. post-intervention) and inter-patient comparisons. Advantageously, with components that contribute to the perfusion signal (parenchymal tissue, blood vessels) and those that do not or have significantly less contributions (bone), sensitivity to posture (patient-detector positioning) and inhomogeneity of the penetrating material are considered important properties for such quantitative comparisons.

[0194] Thus, prior knowledge in the form of a generic 3D model of the limb is often utilized to compensate for the lack of depth information of the perfusion signal. By registering a compatible 3D model to the individual 2D angiographic sequence (estimates posture, limb position, and model internal parameters), the volumetric perfusion area can be mapped from the 3D model to this patient-specific 2D domain, and the regional perfusion signal can be normalized by the actual contributing regional material component fraction.

[0195] In one embodiment, a device for measuring peripheral perfusion comprises an image data input, a data processor, and an output interface. The image data input is configured to receive at least one perfusion angiography two-dimensional x-ray image of a region of interest of a subject's foot and to receive a three-dimensional foot model including spatial perfusion related parameters. The data processor is configured to register the three-dimensional foot model with the foot in the at least one perfusion angiography x-ray image, establish a spatial correspondence between locations in the at least one perfusion angiography x-ray image and related regions in the three-dimensional foot model, map the spatial perfusion related parameters from the three-dimensional model to the at least one perfusion angiography two-dimensional x-ray image, and use the correspondence to modify image signals based on the mapped spatial perfusion related parameters or to map image signals to the three-dimensional foot model. The output interface is configured to provide at least one modified image signal.

[0196] In one embodiment, to improve perfusion analysis, the model possesses anatomically relevant definitions of 3D perfusion regions, such as standard or customized vascularized territories of major blood vessels of the foot, such as angiosoms. Once the model pose is known for the current 2D angiogram, it can be used as a forward model to project these region definitions onto the current 2D region. This maximizes the contours of comparable regions between different angiograms. It also results in a natural approach to normalize the intensity sum of each region, where knowledge of comparable extents is readily available. The forward projection provides information about the overlap of 3D regions in that 2D projection, allowing an estimate of the proportion of parenchymal tissue and bone that is x-ray-penetrated in the projection direction. Thus, on the one hand, the model compensates for the missing depth information of the projection with prior knowledge of the foot anatomy, but also provides a tool to report perfusion measurements in a 3D reference using the posterior pathway.

[0197] In the following, further figures are provided for explanation and illustration of some of the above-mentioned aspects.

[0198] Figure 4 shows an example 300 in which a 3D model of a foot 302 is superimposed in two different bends, i.e. an illustration of how a 3D model can be adapted to different bends. The model foot is displayed with basic structures such as bone silhouettes. A first bend 304 and a second bend 306 are shown. The difference in bends is most visible in the distal end, i.e. in the toe area.

[0199] 5 shows a three-dimensional foot model 308 with an optional foot section 310. The three-dimensional foot model 308 may be articulated. FIG. 5 shows an exemplary foot model 308 including a skin sub-mesh, sub-meshes for individual bones (not further shown), and a division 310 of the foot into perfusion regions, such as a first region 312a associated with the toes, a second region 312b, a third region 312c, a fourth region 312d, a fifth region 312e, and a sixth region 312f. The articulation of the model is illustrated by superimposing the model configurations of the two flexion angles of FIG. 4. An exemplary division into perfusion regions is shown in FIG. 5.

[0200] FIG. 6 shows the sequence of stages of the perfusion analysis using an input X-ray image (left), a 3D foot model for registration (middle), and superimposed anatomical regions as a piecewise map (right).

[0201] FIG. 6 shows an X-ray image 350 on the left, showing the subject's foot in a first pose 352. In one embodiment, the image is a pre-intervention image. In another embodiment, the image is a post-intervention image. For comparison, a pre-intervention overlay image (as shown in the right) is also presented next to each post-intervention image. In the center of FIG. 6, a 3D foot model 354 for registration is shown, with the joints aligned, i.e. aligned to the first pose 356. In the right part of FIG. 6, anatomical regions, such as a piece map 358, based on the 3D foot model 354 in the first pose 356 are superimposed on the pre-intervention X-ray image 350. In the right part of FIG. 6, anatomical regions, such as a piece map 356, based on the 3D foot model 354 in the second pose 362 are also superimposed on the post-intervention X-ray image 360. Each of the partition maps 356 in FIG. 6 is shown with a number of sub-regions 356a, 356b, 356c, 356d, and 356e.

[0202] A first arrow 366 between the left x-ray image and the central 3D foot model indicates the 2D-3D registration (FIG. 6). A second arrow 368 between the central 3D foot model and the projection on the right indicates the projection of the 3D model onto the 2D image plane (FIG. 6).

[0203] In one embodiment (not shown), an exemplary illustration of posture adaptation of perfusion regions for two different input images acquired in different postures is provided: first, the input X-ray image is registered to a 3D foot model. Using the estimated posture parameters, the perfusion regions defined in the 3D model are projected onto the X-ray image. These posture adapted perfusion regions are used to estimate the perfusion tissue density curves of the anatomically corresponding regions, also taking into account the overlap of adjacent perfusion regions in the X-ray image. The posture adaptation of perfusion regions enables quantitative regional perfusion comparison of pre- and post-intervention DSA acquired in different postures.

[0204] FIG. 7 shows a conceptual schematic of an embodiment of peripheral perfusion measurement that decomposes transmitted material into tissue and bone and maps perfusion parameters onto a 3D foot model in order to normalize the X-ray image signal.

[0205] FIG. 7 shows a first X-ray image 370 and a second X-ray image 372 on the left, both showing the same foot 373 from different orientations, as indicated by a first X-ray source position 374 and a second X-ray source position 376. Ray tracing is applied using the estimated pose parameters. A first line 378 shows a first path of radiation from the first X-ray source position 374, respectively. A second line 380 shows a second path of radiation from the second X-ray source position 376, respectively. Furthermore, in the center of FIG. 7, the state of a three-dimensional foot model 382 is shown. The three-dimensional foot model 382 is used to identify the intersection of the foot model with the X-ray. A first intersection 384 of the first line 378 with the three-dimensional foot model 382 in the spatial relationship, i.e. in three dimensions, is shown by a dotted pattern. A second intersection 386 of the second line 380 with the 3D foot model 382 in a spatial relationship, ie, in three dimensions, is also shown in a dotted pattern.

[0206] A first decomposition 390 is provided for x-rays passing through the three-dimensional foot model 382 in a first x-ray direction 392 as indicated by a first dashed arrow 388. The first decomposition 390 shows a first portion 394 for a first category of materials, e.g., tissue structures, as indicated by a first pattern 396, a second portion 398 for a second category of materials, e.g., bone structures, as indicated by a second pattern 400, and a third portion 402 for the first category of materials. A second decomposition 406 is provided for x-rays passing through the three-dimensional foot model 382 in a second x-ray direction 408 as indicated by a second dashed arrow 404. The second decomposition 406 shows a first portion 410 for the first category of materials, a second portion 412 for the second category of materials, and a third portion 414 for the first category of materials. This information may be utilized for normalization of the x-ray signal. For example, the second X-ray penetrates more bone structure than the first X-ray. The signal can then be adapted accordingly, i.e. normalized at least to the first degree.

[0207] Figure 7 thus shows the mapping of perfusion parameters onto a 3D foot model for reporting and comparing pre- and post-intervention images. Given an X-ray image and corresponding DSA sequence, the posture parameters are estimated. For each pixel of the X-ray radiographic image data, the corresponding X-ray can thereby be identified. If the intersection points are calculated with the 3D foot model, the penetrated material can be decomposed into tissue and bone. The DSA perfusion parameters measured at this pixel are mapped onto the 3D model as follows: all positions within the intersection points of the small rod and tissue around the considered X-ray are assigned the same parameter value. Here, the intrinsic parameters, e.g. arrival times, are assigned to these positions without post-processing, and the extrinsic parameters, e.g. time-averaged contrast agent concentration, are normalized by the tissue penetration depth.

[0208] In one embodiment to be claimed, a processor for peripheral perfusion measurement is provided, the processor being configured to perform the steps of one of the method embodiments.

[0209] In one embodiment to be claimed, a computer-implemented method is provided that enables a processor to perform the method of one of the above-mentioned embodiments.

[0210] In one embodiment to be claimed, a computer program or program element for controlling an apparatus according to one of the above-mentioned embodiments is provided, which program or program element is adapted to perform the method steps of one of the above-mentioned method embodiments when executed by a processing device.

[0211] In one embodiment to be claimed, a computer program is provided which enables a processor to carry out the method of one of the method claims.

[0212] The term "subject" may also be referred to as an individual. A "subject" may also be referred to as a patient, although it should be noted that this term does not imply the actual presence or absence of any disease or disorder in the subject.

[0213] In another exemplary embodiment of the invention, a computer program or a computer program element is provided, characterized in that it is adapted to carry out, on a suitable system, the method steps of the method according to one of the preceding embodiments.

[0214] Thus, the computer program elements may be stored in one computing unit or distributed among several computing units, which may also be part of the embodiments of the present invention. The computing unit is adapted to execute or direct the execution of the steps of the above-mentioned method. Furthermore, the computing unit is adapted to operate the above-mentioned equipment components. The computing unit may be adapted to operate automatically and / or to execute user instructions. The computer program is loaded into the working memory of a data processor. In this way, the data processor is equipped to execute the method of the present invention.

[0215] Aspects of the invention may be implemented in a computer program product, which is a collection of computer program instructions stored on a computer readable storage device for execution by a computer. Instructions of the invention may be any interpretable or executable code mechanism, including, but not limited to, a script, an interpretable program, a dynamic link library (DLL), or a Java class. Instructions may be provided as a complete executable program, a partial executable program, a modification (e.g., an update) of an existing program, or an extension (e.g., a plug-in) of an existing program. Additionally, parts of the processing of the invention may be distributed across multiple computers or processors.

[0216] As described above, a processing device, e.g., a controller, executes the control method. The controller can be implemented in numerous ways, with software and / or hardware, to perform the various functions required. The processor is one example of a controller employing one or more microprocessors that are programmed with software (e.g., microcode) to perform the required functions. However, the controller may be realized with or without a processor, or as a combination of dedicated hardware to perform some functions and a processor (e.g., one or more programmed microprocessors and associated circuitry) to perform other functions.

[0217] Examples of controller components employed in various embodiments of the present disclosure include, but are not limited to, conventional microprocessors, application specific integrated circuits (ASICs), and field programmable gate arrays (FPGAs).

[0218] This exemplary embodiment of the present invention covers both computer programs that utilize the present invention from the beginning, and computer programs that have been updated to convert existing programs into programs that utilize the present invention.

[0219] Much further ahead, a computer program element may be capable of providing all the steps required to carry out the procedures of the exemplary embodiments of the method as described above.

[0220] According to a further exemplary embodiment of the present invention, a computer readable medium, such as a CD-ROM, is presented, which has stored thereon a computer program element, which is described in the previous paragraphs. The computer program may be stored and / or distributed on a suitable medium, such as an optical storage medium or a solid-state medium, supplied together with or as part of other hardware, but also in other forms, such as via the Internet or other wired or wireless telecommunications systems.

[0221] However, the computer program may also be presented via a network, such as the World Wide Web, and may be downloaded from such a network into the working memory of a data processor. According to a further exemplary embodiment of the invention, a medium for making available for downloading a computer program element is provided, this computer program element being arranged to carry out a method according to one of the embodiments of the invention described above.

[0222] It should be noted that the embodiments of the present invention are described with reference to different subject matters. In particular, some embodiments are described with reference to method claims, while other embodiments are described with reference to device claims. However, a person skilled in the art will understand from the above and the following description that, unless otherwise specified, in addition to combinations of features belonging to one subject matter, combinations between features related to different subjects are also considered to be disclosed in the present application. However, by combining all features, a synergistic effect can be obtained that is greater than the simple sum of the features.

[0223] While the invention has been illustrated and described in detail in the drawings and foregoing description, such illustration and description are to be considered illustrative or exemplary and not restrictive. The invention is not limited to the disclosed embodiments. Other variations to the disclosed embodiments can be understood and effected by those skilled in the art in practicing the claimed invention, from a study of the drawings, the disclosure, and the dependent claims.

[0224] In the claims, the word "comprises / including / having" does not exclude other elements or steps, and the singular elements do not exclude a plurality. A single processor or other unit may fulfill the functions of several items recited in the claims. The mere fact that certain measures are recited in mutually different dependent claims does not indicate that a combination of these measures cannot be used to advantage. Any reference signs in the claims should not be interpreted as limiting the scope.

Claims

1. An image data input unit; A data processor; Output interface and A device for peripheral perfusion measurement comprising: the image data input receiving at least one perfusion angiographic two-dimensional x-ray image of a region of interest of the subject's foot and receiving a three-dimensional foot model including spatial perfusion-related parameters; the data processor registers the three-dimensional foot model with the subject's foot in the at least one perfusion angiography x-ray image, the registering including estimating a pose of the foot in the two-dimensional x-ray image, mapping information between the two-dimensional x-ray image and the three-dimensional foot model based on the pose estimation, identifying image processing modification instructions based on the mapped information, and modifying at least one image signal based on the image processing modification instructions; The device, wherein the output interface provides the at least one modified image signal.

2. for said mapping, said data processor maps perfusion regions from said three-dimensional foot model onto said two-dimensional x-ray image and projects said perfusion regions in two dimensions; for said identification of said at least one image processing modification instruction, said data processor summing perfusion signals in each of said projected perfusion regions, said data processor fitting said perfusion regions to an estimated pose; The device of claim 1 , wherein for said modification, said data processor generates a set of time-density curves corresponding to these adapted perfusion regions.

3. for said mapping, said data processor performs ray tracing; extracting correction factors from the three-dimensional foot model for said identification of said at least one image processing correction instruction; for said correction, said data processor applies said extracted correction factor to said at least one signal of said two-dimensional X-ray image; 3. A device as claimed in claim 1 or 2, wherein for said providing of the at least one modified image signal, the data processor reports the modified image signal in an image or provides a quantity extracted from the at least one modified image signal.

4. for said mapping, said data processor performs ray tracing of at least one patch of pixels of said two-dimensional X-ray image and maps each image signal to a perfusion region in said three-dimensional foot model that contributed to said patch of said two-dimensional X-ray image; storing perfusion signals or extracted quantities of the two-dimensional X-ray images in the three-dimensional foot model for the identification of the at least one image processing correction instruction; The device according to claim 1 , wherein for said correction, said data processor generates a report by 3D rendering of said 3D foot model using said stored perfusion signals.

5. The device of claim 1 , wherein the three-dimensional foot model is a foot joint model that can be adapted to a predetermined bone arrangement of the subject's foot.

6. 6. The device of claim 1, wherein the spatial perfusion-related parameters include a perfusion analysis-related three-dimensional structure including at least one of the skin of the foot, the parenchymal tissue below the skin, the bone, the division of the foot into clinically relevant regions, and major blood vessel groups.

7. 7. The device according to claim 1, wherein the data processor calculates as perfusion analysis a partial perfusion analysis of a digital subtraction angiography sequence taking into account distortion of the projection images for a given posture.

8. The data processor, for said mapping, ray tracing image points of said at least one perfusion angiography two-dimensional X-ray image, determining possible ray paths based on said ray tracing, and identifying intersections of said determined possible ray paths with said three-dimensional foot model; for said identification, evaluating a portion of perfusable material of said three-dimensional foot model based on said identified intersection points, and determining a transfer factor based on said evaluated portion; spatially transferring values ​​of said two-dimensional x-ray image to said three-dimensional foot model based on said transfer factor for said correction, and generating an augmented three-dimensional foot model with the transferred perfusion values; A device according to claim 1 , which for said providing presents the augmented three-dimensional foot model together with the transferred perfusion values.

9. 9. A device according to any one of claims 1 to 8, wherein a visualization unit is provided with a graphical user interface for presenting results of the data processor via the output interface and for comparing perfusion parameters of different digital subtraction angiography sequences.

10. an X-ray imaging device; A display device; A device according to any one of claims 1 to 9 for peripheral perfusion measurement. A system for peripheral perfusion measurement comprising: said x-ray imaging device providing at least one perfusion angiography x-ray image of a region of interest of a subject's body; The system, wherein the display device displays the at least one modified image signal.

11. receiving at least one perfusion angiographic two-dimensional x-ray image of a region of interest of the subject's foot and receiving a three-dimensional foot model including spatial perfusion-related parameters; registering the three-dimensional foot model with the subject's foot in the at least one perfusion angiography x-ray image, said registering comprising a pose estimation of the foot in the two-dimensional x-ray image; mapping information between the 2D x-ray image and the 3D foot model based on the pose estimation; identifying image processing modification instructions based on the mapped information; modifying the at least one image signal based on the image processing modification instructions; providing said at least one modified image signal; 13. A method for measuring peripheral perfusion comprising:

12. the mapping step comprises the steps of: mapping a perfusion region from the three-dimensional foot model onto the two-dimensional x-ray image; and projecting the perfusion region in two dimensions; and identifying the at least one image processing modification instruction comprises summing perfusion signals in each of the projected perfusion regions, the perfusion regions being adapted to the estimated pose; The method of claim 11 , wherein the modifying step comprises generating a set of time-density curves corresponding to these fitted perfusion regions.

13. the mapping step comprises a ray tracing step; wherein identifying the at least one image processing correction instruction comprises extracting a correction factor from the three-dimensional foot model; the modifying step comprises applying the extracted modification factor to the at least one signal of the two-dimensional X-ray image; 13. The method of claim 11 or 12, wherein the step of providing the at least one modified image signal comprises the step of reporting the modified image signal in the image or the step of providing a quantity extracted from the at least one modified image signal.

14. A computer implemented method enabling a processor to carry out the method of any one of claims 11 to 13.

15. A computer readable medium storing a program for the method according to claim 14.

Images