Method, system and computer program unit for progression monitoring of an x-ray image series

Abstract

The present invention relates to a method, system and computer program element for progression monitoring of a series of X-ray images comprising the steps of receiving a series of X-ray images captured by an X-ray system from a patient, selecting a region of interest (ROI) from at least one image of the series of X-ray images, a selected ROI of each of the plurality of X-ray images is divided into spatial tiles using a processing unit, a respiratory cycle and a heartbeat cycle are determined from the tiles using the processing unit, the respiratory cycle and the heartbeat cycle are eliminated from the tiles using the processing unit, tiles of the ROI are compared to reference tiles, offset tiles that deviate from the reference tiles are determined, and the offset tiles are used to determine a selected ROI of each of the plurality of X-ray images. A visualization superimposed on the image of the series of X-ray images is generated from the offset tiles, and the visualization is displayed on an interface to the user.

Description

Field of the Invention

[0001] The present invention relates to the field of monitoring the temporal changes of a patient in X-ray images (i.e., monitoring the patient's progress). More specifically, the present invention relates to progress monitoring of a series of X-ray images, a system for progress monitoring of a series of X-ray images, and a computer program unit. Background Art

[0002] Despite modern imaging methods, patient progress monitoring remains a challenging task for physicians. For example, even for widely used X-ray equipment, due to changes in the internal parameters and conditions of image capture (e.g., the parameters that can be changed are the voltage and current of the emitter, exposure time, etc.), the patient's position, the inspiration phase, or due to technical reasons or hospital workload, the images obtained for the same patient may be different. These factors make it difficult to directly compare consecutively taken and / or images taken on different days. Therefore, different settings of the same device may complicate the monitoring of the patient's condition / progress during treatment. Modern digital X-ray detectors provide the ability to generate images in a short time, which enables the generation of a time series of X-ray images with high resolution at a rate of up to several frames per second. For such a time series of X-ray images, an objective and quantitative analysis would be highly desirable. Summary of the Invention

[0003] Therefore, there is a need to optimize the progress monitoring of a patient's X-ray images. There is a need to optimize the image capture process and to improve the visualization of X-ray images and / or a series of X-ray images during progress monitoring so that the functional status of the patient can be obtained.

[0004] It is an object of the present invention to provide an improved method, system, and computer program unit for progress monitoring of a series of X-ray images.

[0005] The object of the present invention is solved by the subject matter of the independent claims, wherein further embodiments are incorporated in the dependent claims.

[0006] It should be noted that any feature, function, and / or element described below with reference to the method applies equally to the system, and vice versa. Therefore, any feature, function, step, and / or element described below with reference to one aspect of the present disclosure applies equally to any other aspect of the present disclosure.

[0007] According to a first aspect of the present invention, a method for progress monitoring of a series of X-ray images is described. The method comprises the steps of: receiving a series of X-ray images captured from a patient by an X-ray system, selecting a region of interest (ROI) from at least one image in the series of X-ray images, dividing the selected ROI of each image in the plurality of X-ray images into spatial patches using a processing unit, determining a respiratory cycle and a cardiac cycle from the patches using the processing unit. The method further comprises the steps of: eliminating the respiratory cycle and the cardiac cycle from the patches using the processing unit, comparing the patches of the ROI with reference patches, determining deviated patches deviating from the reference patches, generating a visualization map from the deviated patches and superimposing the visualization map on the images of the series of X-ray images, and displaying the visualization map to a user on an interface.

[0008] In the context of the present invention, the term "series of X-ray images" should be understood as describing a sequence of X-ray images obtained by dynamic X-ray imaging, which describes functional imaging using sequential images and offers many advantages such as high temporal resolution and flexibility in body positioning. The series of images can be understood as a series of images generated during a time series. The series of images can be a sequence of images sequentially generated of the same object.

[0009] In the context of the present invention, the term "region of interest (ROI)" should be understood as describing a sample of particular interest within a medical image for use by a user in medical diagnosis. For example, the region of interest can be a tumor, bronchus, lung, heart, boundary of bronchus, lung region, where the list is not restrictive. The type and size of the ROI can depend on the medical image to be analyzed, the anatomical structures that can be found in the medical image, and the part of the patient's body that is undergoing medical image analysis.

[0010] In the context of the present invention, the term "patch" should be understood as describing a spatial region of a predetermined size, where the size can be defined by the user, predefined, or defined during image analysis. Specifically, a patch should be understood as describing at least one spatial region on an image that is homogeneous, in other words, includes similar image features. The minimum size of the patch can be at least one pixel, where preferably, the size of one or more patches can be greater than one pixel. The region of interest can be divided into a plurality of patches.

[0011] In other words, a method is described that allows the dynamics of a series of X-ray images (in particular a series of chest X-ray images) to be analyzed, observing one or several respiratory cycles with high temporal resolution and analyzing not only the intensities of individual chest snapshots collected at different times. On the other hand, an incomplete respiratory cycle can be used, which means not a complete respiratory cycle of inhalation and exhalation. A "globally uniform" contraction or dilation can be estimated and corrected. Thus, image data of less than one respiratory cycle can be used to estimate respiratory movement. The method includes selecting a region of interest (ROI) on the series of chest X-ray images and then performing a time series dynamic analysis. The time series dynamic analysis can include the following steps: dividing the selected region of interest into spatial sub-blocks and extracting the frequencies of the respiratory cycle and the heartbeat cycle. Comparing the sub-blocks with a reference region that serves as a reference sub-block. The extraction of the respiratory cycle and the heartbeat cycle can include an intensity-based time series analysis, which can include the extraction of the main frequency. To perform the method, a processing unit capable of performing one or more method steps can be used. For example, the series of X-ray images can be received at the processing unit and / or at a memory accessed by the processing unit. The images can be received from an X-ray imaging system. On the other hand, the images can be received from any computer that has received images from an X-ray system.

[0012] The elimination of the respiratory and heartbeat cycles allows for the compensation of drift, which may interfere with changes in anatomical structures in the series of X-ray images that are expected to be detected. For example, positional displacements of contrast objects such as the ribs and the diaphragm can be reduced or almost eliminated.

[0013] The step of determining the deviating sub-blocks that deviate from the reference sub-block can include automatically determining the deviating regions, i.e., the sub-blocks whose frequencies do not correspond to the respiratory cycle and / or the heartbeat cycle or adjacent sub-blocks.

[0014] The present invention can provide the following advantages: for modern machines, the effective radiation dose to the patient is relatively small, so several chest X-ray image scans do not pose a significant risk or radiation burden to the patient. From the perspective of clinical staff, there is no need to re-acquire images in the case of different equipment settings or incorrect inhalation phases. In addition, a physician can still select a suitable image from a series of images for, for example, canonical analysis. More accurate patient progress monitoring will help, for example, to estimate the effectiveness of treatment strategies and make necessary operational changes.

[0015] According to an exemplary embodiment of the present invention, the step of comparing the patch with the reference patch may include extracting amplitude and / or phase information from the X-ray image based on the real part and / or the imaginary part of the complex coherence. The extraction of the phase information may include any extraction of any image signal, in particular from any image of an image series. For example, for a given image, a pair of pixels or multiple pairs of pixels may be selected, and then a time series may be constructed based on using this image as a reference image. The pixel pair may refer to a part of the corresponding patch, in particular a patch that can be tracked along the image series. In addition, the extraction of the phase information may be an advanced mathematical model for extracting appearance changes in the image patch.

[0016] According to an exemplary embodiment of the present invention, determining the deviated patch may include at least one of independent component analysis, calculation of statistical moments, or the maximum pixel value change rate. For example, for this step, the respiratory cycle and the heartbeat cycle are removed by excluding the corresponding independent components and restoring the original image (i.e., the image signal). Therefore, it may be any one or even a combination of the above for determining the deviated patch.

[0017] According to an exemplary embodiment of the present invention, the region of interest may be selected automatically or manually by the user. The region of interest may be such a region that is the anatomical structure for which progress monitoring should be performed. The region of interest may be determined by the user according to the anatomical structure to be studied. On the other hand, it may be determined automatically, for example, by artificial intelligence, which may be trained with X-ray image data including different anatomical structures such that the artificial intelligence can determine the corresponding anatomical structure shown in the X-ray image series. Additionally, the reference patch may be selected automatically or manually by the user, where the reference patch is a part of the X-ray image, and the part is determined as a normal element in the X-ray image. For example, the reference patch may be a patch within a bone, a heart, or an absolutely healthy tissue, depending on the task to be completed from the X-ray image series. If artificial intelligence is used to determine the patch, the artificial intelligence may be trained with healthy tissue data of the corresponding anatomical structure. The reference patch as a normal element in the X-ray image may be a healthy tissue or a healthy bone or a normal element in the sense of a normally functioning anatomical structure.

[0018] According to an exemplary embodiment of the present invention, the following steps may be performed for each image in a series of X-ray images: dividing a selected ROI of each image among a plurality of X-ray images, determining a respiratory cycle and a heartbeat cycle from the patches, eliminating the respiratory cycle and the heartbeat cycle from the patches, comparing the patches with a reference patch, determining a deviated patch deviating from the reference patch, and generating a visualization map from the deviated patch. Specifically, each of the above steps may be performed for each individual image of the X-ray series. Specifically, each step may be sequentially performed for each individual image of the time series, or these steps may be sequentially and simultaneously performed for each image of the time series of images. In other words, all images of the image series may have to undergo all of the above steps such that an estimation of dynamic parameters of tissues can be performed for and over the entire image series.

[0019] According to an exemplary embodiment of the present invention, determining a respiratory cycle and a heartbeat cycle may include determining the respiratory cycle and the heartbeat cycle based on at least one of a dynamic of a total patch intensity or a rate of change of pixel values. In other words, the patch intensity may not be measured in such a way, but the corresponding patches may be tracked spatially. The corresponding patches are the patches of each image in the image series. For example, a first image includes patches x1, x2, and x3, and a second image includes patches y1, y2, x3, where patch x1 anatomically corresponds to patch y1 of another image, that is, patches x1 and y1 depict the same anatomical structure (which may have moved relative to other anatomical structures due to some form of movement), and so on. The corresponding patches may be described as patches of different images that correspond in their positions in the images. The spatially tracked patches may be anatomical structures (bones, tissues, boundaries, etc.). The steps of determining different cycles may be performed by a processing unit or any computer element.

[0020] According to an exemplary embodiment of the present invention, in order to determine a respiratory cycle and / or a heartbeat cycle, one or more sensors may be additionally used. For example, it may include a respiratory sensor (spirometer) or data of a heart measurement device, a cardiac electrode, or sensor data for electrocardiogram. The sensor signal may be used to eliminate the respiratory cycle and / or the heartbeat cycle in the (one or more) images of the time series.

[0021] According to an exemplary embodiment of the present invention, the dynamic X-ray sequence may be a time series of chest X-ray images. In other words, the X-ray images are images taken from a patient's chest during X-ray image processing, where a sequence of X-ray images changing over time is generated.

[0022] According to an exemplary embodiment of the present invention, the method may further include the steps of: generating a total score of the ROI based on the ratio of normal patches and deviated patches, and comparing the total score with a previous X-ray image. The scores and spatial distributions of the deviated patches can be compared with previous observations, which means comparing with previous images in a previous time series.

[0023] The method may generally include dynamically evaluating the functional characteristics of tissue, where the tissue may be patches of the ROI. Thus, at least a short time series is required to generate an image sequence from which the functional characteristics can be determined.

[0024] According to an exemplary embodiment of the present invention, the method may further include the step of comparing patches of the received X-ray image series with patches received from another X-ray image series. Thus, the user can manually select the corresponding patches and then compare them with the patches in a previous session. For two patches, the fundamental frequencies such as respiration and heartbeat should be eliminated from the two time series so that a comparison between patches from different time series can be performed.

[0025] According to an exemplary embodiment of the present invention, the visualization map may be selected from at least one of a respiration map and a perfusion map, where any visualization of the spatio-temporal map of the ROI can be applied. In addition, the average synchronization level or the relative change in the intensity of the selected ROI can be calculated. The value(s) can be compared with a threshold, such as the average value plus 3σ from the same region (ROI or only patches of the ROI) or from the selected reference patches. Then, for the selected region, the pixels can be plotted relative to this baseline. The visualization map can be selected by the user according to the selected region of interest and / or according to the diagnosis of the patient. In addition, the visualization map can be automatically selected according to the selected region of interest or the anatomical structure determined in the X-ray image (series) (by the processing unit using artificial intelligence or any other object detection unit).

[0026] According to an exemplary embodiment of the present invention, the generation of the visualization map may further include generating an ROI color map according to the selected visualization map, and the ROI color map is displayed as an overlay on a plurality of X-ray images. In particular, the visualization map visualizes the selected region of interest and indicates the region with a determined color. In addition, the ROI color map may be a spatial color map, which can distinguish patches by color indication in the region of interest.

[0027] According to an exemplary embodiment of the present invention, the generation of the visualization map may include a graphical representation of the functional status of the ROI. In particular, the functional status of the anatomical structure of interest can be visualized. For example, whether the two lung lobes of the lung are functioning properly, i.e., inhalation and exhalation can be derived from the visualization map, and the expansion of the lung during inhalation and exhalation can be derived. This also applies to other anatomical structures, such as the normal beating of the heart or the perfusion of the lung, etc.

[0028] According to an exemplary embodiment of the present invention, the received series of X-ray images is a time series of consecutive X-ray images. Therefore, the multiple images received during X-ray imaging are sequentially received images, where multiple X-ray images are generated during the execution of the imaging process.

[0029] According to a second aspect of the present invention, a system for progress monitoring of a series of X-ray images is described. The system may include a processing unit configured to: receive the series of X-ray images from a patient, select a region of interest (ROI) from at least one image in the series of X-ray images, divide the selected ROI of each image in the series of X-ray images into spatial patches using the processing unit, determine the respiratory cycle and the heartbeat cycle from the patches using the processing unit, eliminate the respiratory cycle and the heartbeat cycle from the patches using the processing unit, compare the patches with reference patches, determine the deviated patches deviating from the reference patches, and generate a visualization map superimposed on the images of the series of X-ray images from the deviated patches. In addition, the system may include an interface, where the interface is configured to display the visualization map generated by the processing unit to the user.

[0030] The system may be a stand-alone unit, which may be attached and / or integrated into any suitable X-ray system. On the other hand, the system may include an X-ray system that includes a corresponding X-ray source and an X-ray detector for performing X-ray imaging and includes a processing unit configured to perform the steps described above.

[0031] The proposed system and the proposed method can in principle be applied to any hospital having an X-ray machine (which can provide a high X-ray capture rate). All the required method steps and / or calculations can be provided in the cloud, which may also include automatic region of interest selection (e.g., based on lung segmentation) in the case of automatically uploaded chest X-ray image data. For example, user controls and result presentation can be provided on a display including the system interface, or can also be provided in a tablet application.

[0032] According to a third aspect of the present invention, a computer program unit for progress monitoring of a series of X-ray images. The computer program unit may be adapted to cause a system, when run by a processing unit of the system, to receive a series of X-ray images captured from a patient, select a region of interest (ROI) from at least one image in the series of X-ray images, divide the selected ROI of each image in the series of X-ray images into spatial patches using the processing unit, determine a respiratory cycle and a heartbeat cycle from the patches using the processing unit, eliminate the respiratory cycle and the heartbeat cycle from the patches using the processing unit, compare the patches with reference patches, determine deviated patches deviating from the reference patches, generate a visualization map from the deviated patches to be superimposed on the images of the series of X-ray images, and display the visualization map to a user on an interface.

[0033] The computer program unit may be part of a computer program, but it may also be the entire program itself. For example, the computer program unit can be used to update an existing computer program to implement the present invention.

[0034] The program unit may be stored on a computer-readable medium. The computer-readable medium can be regarded as a storage medium, such as a USB stick, a CD, a DVD, a data storage device, a hard disk, or any other medium on which the above program unit can be stored.

[0035] According to various embodiments of the present disclosure, the methods described herein can be implemented using a hardware computer system executing a software program. In addition, in an exemplary non-limiting embodiment, the implementation can include distributed processing, component / object distributed processing, and parallel processing. Virtual computer system processing can implement one or more of the methods or functions described herein, and the processors described herein can be used to support a virtual processing environment.

[0036] It must be noted that embodiments of the present invention have been described with reference to different subjects. In particular, some embodiments have been described with reference to apparatus / system type claims, while other embodiments have been described with reference to method type claims. However, those skilled in the art will understand from the above and the following description that, unless otherwise stated, any combination between features related to different subjects, in particular any combination between features of apparatus type claims and features of method type claims, is also considered to be disclosed together with this application, in addition to any combination of features belonging to one type of subject. Description of the Drawings

[0037] From the examples of the embodiments to be described below, the above-defined aspects and other aspects of the present invention will be apparent and will be explained with reference to the examples of the embodiments. The present invention will be described in more detail below with reference to the examples of the embodiments, but the present invention is not limited thereto.

[0038] Figure 1 Shows a flowchart of a method according to an embodiment of the present invention.

[0039] Figure 2 Shows different images of an image series according to an embodiment of the present invention.

[0040] Figure 3 Shows the determination of a deviation patch according to an embodiment of the present invention.

[0041] Figure 4 Shows the generation of a visualization diagram according to an embodiment of the present invention.

[0042] List of reference numerals:

[0043] S1 to S9 Each step of the method 101, 102, 103 X-ray images

[0044] 105 X-ray image

[0045] 310 Visualization diagram

[0046] 320 Respiratory cycle

[0047] 321 Region of interest

[0048] 410 Visualization diagram

[0049] 411 Inverted cardiac signal

[0050] 412 Cardiac signal

[0051] 413 Lung signal

[0052] 421 Region of interest Detailed implementation manner

[0053] The illustrations in the drawings are schematic. Note that in different figures, similar or identical elements are provided with the same reference numerals.

[0054] Figure 1A flowchart showing method steps according to an embodiment of the present invention is presented. A method for progress monitoring of a series of X-ray images, the method comprising steps S1 to S9, where more method steps may also be applied. The order of the steps is merely exemplary and is not limited to the described order. For example, the order of one or more steps may be changed with respect to each other, or one or more steps may be repeated before performing another step. The method includes: step S1, receiving a series of X-ray images captured from a patient by an X-ray system; step S2, selecting a region of interest (ROI) from at least one image in the series of X-ray images; step S3, using the processing unit to divide the selected ROI of each image in the plurality of X-ray images into spatial patches. Step S3 may include further sub-steps, where, for each image of the image series, the ROI is divided into patches. On the other hand, in step S3, for each image of the image series, the ROI is divided into patches simultaneously. In addition, the other described method steps may be performed sequentially for each image of the image series, or may be performed simultaneously for each image of the image series. Further, the method includes: step S4, using the processing unit to determine the respiratory cycle and the heartbeat cycle from the patches; step S5, using the processing unit to eliminate the respiratory cycle and the heartbeat cycle from the patches; step S6, comparing the patches of the ROI with reference patches; step S7, determining patches deviating from the reference patches; step S8, generating a visualization map superimposed on the images of the series of X-ray images from the deviating patches; and step S9, displaying the visualization map to a user on an interface. The display of the visualization map may be displayed to the user on an interface, or on a display of the interface, on a display of an X-ray device or an X-ray system. Alternatively, the display may be performed on a mobile device (such as a tablet computer).

[0055] Step S6 may further include comparing the patches with reference patches, including extracting amplitude and / or phase information from the X-ray images respectively based on the real part and / or the imaginary part of the complex coherence, which may be calculated as a correlation function between spectral components of two time series for a given set of frequencies, where this may be performed in a sub-step of step S6. In addition, step S7 may include at least one of independent component analysis, calculation of statistical moments, or the rate of change of the maximum pixel value, where each of these methods may be performed independently of each other as sub-steps of step S7, or may be additional steps after step S7.

[0056] Steps S3 to S8 are performed sequentially for each image in the series of X-ray images, or steps S3 to S8 are performed simultaneously for each image of the series of images.

[0057] The method may further include additional steps, such as generating an overall score of the ROI based on the ratio of normal patches and deviated patches, and comparing the overall score with a previous X-ray image. Thus, a current X-ray image series can be compared with another X-ray image series (e.g., a historical image series) to analyze changes in the patient's condition or the progression of the condition of the functional characteristics of the analyzed anatomical structure.

[0058] Figure 2 Three different images 101 to 103 of a patient's chest are shown. An example of a series of lateral chest X-ray images observed for a patient during treatment is shown. Different positions of the patient and capture settings of the same device may complicate patient condition monitoring, wherein when the method as described above is applied, these complexities can be overcome because for each image, a region of interest can be determined and the region of interest can be analyzed using the method as described herein.

[0059] Figure 3 An illustration of the determination and visualization of deviated patches of FIG. 310 according to an embodiment of the present invention is shown. The region of interest 321 is selected manually or automatically by a user, a physician. In Figure 3 , the region of interest 321 is the lungs, wherein two lung lobes of the lungs are indicated as the region of interest 321. It can be seen that at least five X-ray images 101-105 are shown from the X-ray image series. Thus, it can be said that Figure 3 the image series of includes at least five chest X-ray images 101 to 105. For each of the images 1 to 105, the region of interest 321 is indicated. In addition, in each of the images 1 to 105, the region of interest 321 is divided into patches. Starting from image 101, the right lobe of the lungs is fully indicated in the region of interest. In contrast, the left lobe of the lungs is not fully visible. When comparing image 101 with 105, the differences between the indicated regions of interest can be compared, thus comparing the differences between the patches of the left and right lobes of the lungs. In addition, Figure 3 A respiratory cycle 320 determined from the patches of the region of interest 321 is shown, wherein the respiratory cycle 320 is determined from / on all of the images of the image series 101 to 105. The respiratory cycle 320 is shown as a frequency plot and is generated for the selected patches. The respiratory cycle 320 can be divided into an inhalation part and an exhalation part, wherein the determination of the rate of change of pixel values can be performed to determine deviated patches in the region of interest. For example, an increasing pixel value can be determined during exhalation, and a decreasing pixel value can be determined for inhalation. Thus, the functional condition of the lungs can be shown via the visualization diagram 310, which can be displayed to the user via a display on the interface, and the functional condition of the patient observed within a time series can be displayed. The shown chest X-ray images can also be used for the user to perform independent observation / analysis. [[ID=I5]]

[0060] Figure 4 Shows the generation of the visualization graph 410 according to an embodiment of the present invention. In particular, as the visualization graph 410, a perfusion graph generation process is shown, which is based on the dynamic X-ray time series of the images 101 to 105 of the lungs. According to the time series of the X-ray images 101 to 105, the lung signal 413 (which may be the respiratory cycle) and the cardiac signal 412 (i.e., the heartbeat cycle) can be determined. Thereafter, these signals can be eliminated. For example, the anti-phase cardiac signal 411 can be applied to eliminate the cardiac signal 412. Additionally, according to the signal to be eliminated and the signal to be obtained, filters such as high-pass or low-pass filters can be applied.

[0061] Although the present invention has been illustrated and described in detail in the drawings and the foregoing description, such illustration and description are to be considered illustrative or exemplary and not restrictive; the present invention is not limited to the disclosed embodiments. Other variations of 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 appended claims. It should be noted that the term "comprising" does not exclude other elements or steps, and "a" or "an" does not exclude a plurality. Further, elements described in connection with different embodiments can be combined. It should also be noted that the reference signs in the claims should not be construed as limiting the scope of the claims.

Claims

1. A method for progress monitoring of a series of X-ray images, the method comprising the following steps: Receiving the series of X-ray images captured from a patient by an X-ray system, Selecting a region of interest (ROI) from at least one image in the series of X-ray images, Using a processing unit to divide the selected ROI of each image in the plurality of X-ray images into spatial patches, Using the processing unit to determine a respiratory cycle and a heartbeat cycle from the patches, Using the processing unit to eliminate the respiratory cycle and the heartbeat cycle from the patches, Comparing the patches of the ROI with reference patches, Determining patches deviating from the reference patches, Generating a visualization map from the deviating patches and superimposing the visualization map on the images in the series of X-ray images, Displaying the visualization map to a user on an interface.

2. The method according to claim 1, Among them, The step of comparing the patches with the reference patches includes extracting amplitude and / or phase information from the X-ray images respectively based on the real part and / or the imaginary part of the complex coherence.

3. The method according to claim 1 or 2, Among them, Determining the patches deviating from the reference patches includes at least one of independent component analysis, calculation of statistical moments, or rate of change of maximum pixel value.

4. The method according to any one of the preceding claims, Among them, The ROI is selected automatically or manually by the user, and / or wherein, the reference patches are selected automatically or manually by the user, wherein, the reference patches are a part of the X-ray images, and the part is determined as a normal element in the X-ray images.

5. The method according to any one of the preceding claims, Among them, Performing the following steps for each image in the series of X-ray images: dividing the selected ROI of each image in the plurality of X-ray images, determining a respiratory cycle and a heartbeat cycle from the patches, determining a respiratory cycle and a heartbeat cycle from the patches, eliminating the respiratory cycle and the heartbeat cycle from the patches, comparing the patches with reference patches, determining patches deviating from the reference patches, and generating a visualization map from the deviating patches.

6. The method according to any one of the preceding claims, Among them, Determining the respiratory cycle and the heartbeat cycle includes determining the respiratory cycle and the heartbeat cycle according to at least one of the dynamics of the total patch intensity or the rate of change of pixel values.

7. The method according to any one of the preceding claims, Among them, The dynamic X-ray sequence is a time series of chest X-ray images.

8. The method according to any one of the preceding claims, wherein, The method further comprises the following steps: Generating a total score of the ROI based on the ratio of normal patches and deviating patches, and Comparing the total score with a previous X-ray image.

9. The method according to any one of the preceding claims, wherein, The method further comprises the following steps: Comparing the patches of the received series of X-ray images with patches received from another series of X-ray images.

10. The method according to any one of the preceding claims, Among them, The visualization map is selected from at least one of a respiratory map and a perfusion map.

11. The method according to any one of the preceding claims, Among them, The generation of the visualization map further includes generating an ROI color map according to the selected visualization map, and the ROI color map is displayed as an overlay on the plurality of X-ray images.

12. The method according to any one of the preceding claims, Among them, The generation of the visualization map includes the illustration of the functional status of the ROI.

13. The method according to any one of the preceding claims, Among them, The received X-ray image series is a time series of consecutive X-ray images.

14. A system for progress monitoring of an X-ray image series, comprising: A processing unit configured to: Receive the X-ray image series from a patient, Select a region of interest (ROI) from at least one image in the X-ray image series, Use the processing unit to divide the selected ROI of each image in the X-ray image series into spatial patches, Use the processing unit to determine the respiratory cycle and the heartbeat cycle from the patches, Use the processing unit to eliminate the respiratory cycle and the heartbeat cycle from the patches, Compare the patches with a reference patch, Determine the deviated patches that deviate from the reference patch, Generate a visualization map from the deviated patches and overlay it on the images of the X-ray image series, The system further includes an interface, wherein the interface is configured to display the visualization map generated by the processing unit to the user.

15. A computer program unit for progress monitoring of an X-ray image series, Among them, When run by the processing unit of the system, the computer program unit is adapted to cause the system to: Receive the X-ray image series captured from a patient, Select a region of interest (ROI) from at least one image in the X-ray image series, Use the processing unit to divide the selected ROI of each image in the X-ray image series into spatial patches, Use the processing unit to determine the respiratory cycle and the heartbeat cycle from the patches, Use the processing unit to eliminate the respiratory cycle and the heartbeat cycle from the patches, Compare the patches with a reference patch, Determine the deviated patches that deviate from the reference patch, Generate a visualization map from the deviated patches and overlay it on the images of the X-ray image series, Display the visualization map to the user on the interface.

Images