System and method for determining a customized viewing window for a medical image - Patents.com

Abstract

A method for processing medical image information includes generating a first intensity distribution for a first image region, generating a second intensity distribution for a second image region, calculating a value based on the first intensity distribution and the second intensity distribution, and automatically determining a custom viewing window based on the calculated value. The custom viewing window is determined for viewing the first image region and the second image region present in the same image or different images. The images include brain scan images or other types of images.

Description

[Technical field]

[0001] FIELD OF THE DISCLOSURE

[0001] Exemplary embodiments disclosed herein relate to processing medical image information. [Background technology]

[0002]

[0002] When a medical image is displayed on a screen, a mapping from voxel values ​​to grayscale values ​​(e.g., brightness) is selected. Typically, the human eye can perceive a much smaller number of grayscale values ​​(e.g., up to 100) than are contained in a medical image. For example, a 12-bit CT may contain 4096 different voxel values. As a result, different standard mappings (or viewing windows) with predefined parameters have been established for different diagnostic tasks. Summary of the Invention [Problem to be solved by the invention]

[0003]

[0003] However, for some tasks, very slight differences between specific areas (or regions of interest) within one scan or image slice (or between different scans or image slices) must be evaluated. To perform this evaluation, a manual process is performed to select (or guess) which of the standard viewing windows can make the slight differences visible to the radiologist. Manually adjusting the viewing windows is time-consuming and error-prone. For example, important clinical findings are often missed because none of the standard viewing windows are sufficient to reveal those findings, and furthermore, the windows are not customized in any way for the particular patient being evaluated. [Means for solving the problem]

[0004]

[0004] A summary of various exemplary embodiments is presented below. Some simplifications and omissions may be made in the following summary, which are intended to highlight and introduce some aspects of various exemplary embodiments, and do not limit the scope of the present invention. A detailed description of exemplary embodiments suitable for enabling those skilled in the art to make and use the concept of the present invention is provided in the following section.

[0005]

[0005] Various embodiments relate to a method for processing medical image information comprising generating a first intensity distribution for a first image region, generating a second intensity distribution for a second image region, calculating a value based on the first intensity distribution and the second intensity distribution, and automatically determining a custom viewing window based on the calculated value, wherein the custom viewing window is determined for displaying the first image region and the second image region.

[0006] Various embodiments are described in which determining the custom viewing window comprises determining one or more parameters of the custom viewing window.

[0007]

[0007] Various embodiments are described in which determining one or more parameters includes calculating one or more parameters based on a difference value calculated between a first intensity distribution and a second intensity distribution.

[0008] Various embodiments are described in which determining the one or more parameters comprises applying at least one rule to the difference values ​​to determine the one or more parameters.

[0009]

[0009] Various embodiments are described in which one or more parameters of a custom viewing window enhance the perception of differences between a first image region and a second image region more than a viewing window used to display the first image region and the second image region using one or more other parameters.

[0010] Various embodiments are described in which the one or more parameters include at least one of a width or a level of the custom viewing window.

[0011]

[0011] Various embodiments are described in which the width is based on a maximum intensity value among the non-zero difference values ​​and a minimum intensity value among the non-zero difference values.

[0012] Various embodiments are described in which the value calculated based on the first intensity distribution and the second intensity distribution comprises a difference value between the first intensity distribution and the second intensity distribution.

[0013] Various embodiments are described in which the first image region and the second image region are in the same image.

[0014] Various embodiments are described in which the first image region is a first area in a first brain scan and the second image region is a second area in the first brain scan.

[0015] Various embodiments are described in which a first image region is obtained from a first bounding box and a second image region is obtained from a second bounding box.

[0016] Various embodiments are described that further include automatically identifying the first image region and the second image region based on a segmentation algorithm.

[0017] Various embodiments are described that further include automatically identifying a second image region based on the designation of the first image region.

[0018]

[0018] Further various embodiments relate to a medical image analyzer including a generator configured to generate a first intensity distribution for a first image region and a second intensity distribution for a second image region, an arithmetic logic unit configured to calculate a value based on the first intensity distribution and the second intensity distribution, and a rules engine configured to automatically determine a custom viewing window based on the calculated value, wherein the custom viewing window is determined for displaying the first image region and the second image region.

[0019] Various embodiments are described in which a rules engine is configured to determine one or more parameters of a custom viewing window.

[0020]

[0020] Various embodiments are described in which determining one or more parameters includes calculating one or more parameters based on a difference value calculated between a first intensity distribution and a second intensity distribution.

[0021] Various embodiments are described in which the rules engine is configured to apply at least one rule to the difference values ​​to determine one or more parameters.

[0022] Various embodiments are described in which the one or more parameters include at least one of a width or a level of the custom viewing window.

[0023]

[0023] Various embodiments are described in which the width is based on a maximum intensity value in the difference values, a minimum intensity value in the difference values, and a scalar.

[0024] Various embodiments are described in which the first image region and the second image region are in the same image.

[0025]

[0025] Additional objects and features of the present invention will become more readily apparent from the following detailed description and the appended claims when considered in conjunction with the drawings in which:

[0025] Several illustrative embodiments are shown and described, in which like reference characters identify like parts in each of the figures. [Brief description of the drawings]

[0026] [Figure 1]

[0026] FIG. 1 illustrates an embodiment of a medical image analyzer. [Diagram 2]

[0027] FIG. 1 illustrates an embodiment of a method for automatically determining a custom viewing window for displaying a medical image. [Figure 3A]

[0028] FIG. 1 shows an exemplary image of a brain scan. [Figure 3B] FIG. 1 shows an exemplary image of a brain scan. [Figure 4A]

[0029] FIG. 13 illustrates an example of generating a difference histogram according to one embodiment. [Figure 4B] FIG. 13 shows an example of a difference histogram that is not generated as a curve. [Figure 5A]

[0030] FIG. 1 shows a brain scan displayed using a standard viewing window. [Figure 5B] FIG. 1 illustrates a brain scan image displayed using an embodiment of a custom viewing window. [Figure 6]

[0031] FIG. 1 illustrates an embodiment of a medical image analyzer. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0027]

[0032] It should be understood that the figures are only schematic and are not drawn to scale. It should also be understood that the same reference numbers are used throughout the figures to indicate the same or similar parts.

[0028]

[0033] The description and drawings illustrate the principles of various exemplary embodiments. Thus, it is understood that one skilled in the art can conceive of various configurations that embody and fall within the scope of the principles of the present invention, although not expressly described or presented herein. Furthermore, all examples described herein are expressly intended primarily for educational purposes to aid the reader in understanding the principles and concepts of the present invention, which have been contributed by the inventors to advance the art, and are not to be construed as being limited to such expressly described examples and conditions. Furthermore, the term "or" as used herein represents a non-exclusive or (i.e., and / or), unless otherwise indicated (e.g., "or as an exclusive or" or "or in the alternative sense"). Furthermore, the various exemplary embodiments described herein are not necessarily mutually exclusive, since some exemplary embodiments may be combined with one or more other exemplary embodiments to form new exemplary embodiments. Descriptors such as "first", "second", "third", etc., are not meant to limit the order of the elements described, but are used to distinguish one element from the next, and are generally interchangeable. Values ​​such as maximum or minimum values ​​are predefined and set to different values ​​based on the application.

[0029]

[0034] 1 shows a system for automatically determining a custom viewing window for displaying a medical image. The system includes a medical image analyzer 1 for processing medical images acquired from a computed tomography (CT) scanner or another type of medical imaging system. An example of a CT image is described in relation to the following embodiments.

[0030]

[0035] The medical image analyzer 1 is coupled between a medical image scanner 2 and a display 50. As shown so far, the scanner may be of various kinds, including but not limited to a CT scanner and a medical resonance imaging (MRI) scanner. In one embodiment, a segmenter and / or labeler 3 may be coupled between (or included in) the medical image analyzer to pre-process the image output from the scanner. The segmenter may, for example, automatically segment the image into various regions of interest or allow the specification of bounding boxes to specify the regions of interest. The automatic segmentation may be performed, for example, to identify Alberta Stroke Program Early (non-contrast) CT ASPECTS regions in an image slice of the brain. In such a case, the labeler may insert contours around the various ASPECTS regions in the image slice and label the various ASPECTS regions for further analysis.

[0031]

[0036] Referring to FIG. 1, the medical image analyzer includes an intensity distribution generator 10, an arithmetic logic unit 20, and a rule engine 30. The intensity distribution generator generates intensity distributions of a first image region and a second image region corresponding to an image output from a scanner. The first image region and the second image region are in the same image (e.g., when left and right ASPECTS regions are considered for analysis) or in different regions (e.g., when brain scans of the same patient are analyzed at different time points). The first image region is identified, for example, based on a bounding box superimposed on the image, automatically identified by a segmenter, or identified using different techniques. In some examples, more than two regions of interest are analyzed, as described in more detail below.

[0032]

[0037] In one embodiment, the intensity distribution generator is a histogram generator 10. For illustrative purposes, it is assumed that the intensity distribution generator is a histogram generator in view of the balance of this discussion. In operation, the histogram generator 10 extracts intensity values ​​from each of the regions of interest and generates a first histogram and a second histogram for corresponding ones of the regions of interest. Once generated, the histograms are further analyzed by the arithmetic logic unit.

[0033]

[0038] Arithmetic logic 20 calculates a number of values ​​based on the first and second histograms. In one embodiment, arithmetic logic 20 includes difference logic that calculates difference values ​​based on subtracting intensity values ​​in the first and second histograms. For purposes of illustration, arithmetic logic 20 is assumed to be a histogram generator in view of the balance of this discussion. A more detailed discussion of how these values ​​are generated is provided below.

[0034]

[0039] The rules engine 30 applies one or more predefined rules to the difference values ​​output from the difference logic. The rules are used to generate a custom viewing window that is used as a basis for displaying a region of interest in a manner that enhances recognition of features that would be hidden if a standard viewing window were used to view the region. In one embodiment, the rules applied by the rules engine determine one or more parameters of the viewing window. The parameters include, for example, one or both of a window width and a window level. Generating the parameters from the difference values ​​allows for customization of the viewing window for each patient. After the parameters for the custom viewing window are generated, information 40 defining the viewing window is output to a display 50 for viewing by a healthcare professional.

[0035]

[0040] 2 illustrates operations included in an embodiment of a method for automatically determining a custom viewing window for displaying a medical image. The method may be performed, for example, by any of the system (or medical image analyzer) embodiments described herein, or by a different system. For illustrative purposes, the method is described as being performed by the system of FIG.

[0036]

[0041] Referring to FIG. 2, the method includes receiving one or more images to be evaluated at 210. The one or more images correspond, for example, to the same patient. In such a case, the images include any part of the patient's body identified based on the purpose for which the images were acquired. In one non-limiting embodiment, the patient is one suspected of suffering from a brain abnormality, for example a lesion produced by an ischemic stroke, or the abnormality corresponds to another brain condition. In this case, the images produced by the scanning system (e.g. a computed tomography scanner) are non-contrast CT (NCCT) brain scan images. The CT scanner may be local (e.g. in the same hospital or clinical site) or may be located remotely from the processing logic of the system implementing the method. In this latter example, the system and method embodiments are used in outsourcing cases, for example where the images are received from a network.

[0037]

[0042] The brain scan image includes axial slices corresponding to a particular area of ​​the brain where one or more types of brain abnormalities occur, for example, given the patient's suspected symptoms. For example, when a patient is suspected of having an ischemic stroke, the slices corresponding to the brain scan image include the region in which the middle cerebral artery (MCA) is located. As will be described in more detail, such a stroke produces an obstruction (e.g., an ischemic lesion) in the image at least at the location of the MCA relative to the hemisphere of the brain in which the stroke occurred. The image slices are stored in the system's memory, with or without normalization, for processing according to the operations described herein. In one embodiment, two or more images are received for evaluation.

[0038]

[0043] At 220, the system receives information specifying at least two regions of interest in one or more images. When the one or more images include at least one image (e.g., a CT scan), the at least two regions of interest may be identified in a variety of ways. For example, the first and second regions of interest may be selected to correspond to areas enclosed within respective bounding boxes superimposed on the same image or on different images. The bounding boxes may be drawn, for example, by a radiologist. In another embodiment, the regions of interest correspond to areas specified by contours or polygons superimposed or drawn on the same or different images. An example in which the regions of interest are in the same image is described for illustrative purposes.

[0039]

[0044] In one embodiment, the region of interest corresponds to one of a plurality of areas generated, for example, by an image segmentation algorithm. For example, in a brain scan embodiment, the segmentation algorithm automatically identifies a plurality of regions defined by the ASPECTS protocol and segments the image into a plurality of regions defined by the ASPECTS protocol. When segmentation is performed according to the ASPECTS protocol, a total of 20 regions of interest are defined in two image slices, 10 regions in the left hemisphere of the brain and 10 regions in the right hemisphere of the brain. The 10 regions in the left hemisphere are of the same type as the regions in the right hemisphere, and therefore the 10 regions in the left hemisphere are considered to be complementary to each of the regions in the right hemisphere, thus forming complementary region pairs.

[0040]

[0045] Table 1 identifies 10 regions in each hemisphere of the brain that are compatible with the ASPECTS protocol. As shown for each hemisphere, seven of the 10 regions appear in the superior image slice (Figure 3A) and the remaining three regions appear in the inferior image slice (Figure 3B). Together, all 10 regions are considered candidate regions for lesions or other abnormalities affecting the brain.

[0041] [Table 1]

[0042]

[0046] In one embodiment, the segmentation results of the image slices are performed automatically in 3D using a model-based approach. The model processes the image to identify (extract) and then generate an overlay graphic that outlines each of the 10 regions of interest. The model then labels each of the segmented regions as shown in Figures 3A and 3B. The outlines are matched to each other using, for example, multi-plane reformatting. Automatic segmentation of image slices may be performed, for example, according to the techniques described in WO2020 / 109006 or as described in the article entitled "Automatic model-based segmentation of the heart in CT images" by Ecabert O, Peters J, Schramm H, Lorenz C, von Berg J, Walker MJ, Vembar M, Olszewski ME, Subramanyan K, Lavi G, Weese J. et al. (disclosing various techniques including, but not limited to, applying a shape-constrained anatomical active contour model using a triangular mesh to the patient scan), IEEE Trans Med Imaging, September 2008;27(9):1189-201, the entire contents of which are incorporated herein by reference.

[0043]

[0047] In the case of the brain scan embodiment described above, step 220 includes receiving information designating any two of the ASPECTS regions. For example, the two designated regions may be the same region on complementary sides of the brain, or may be different ones of the ASPECTS regions. In either case, the two regions may be designated by manual selection, for example by a radiologist or other health care professional, or an algorithm may select the two regions of interest based on preferences, for example built into the control software implementing the algorithm. In one embodiment, the selection of one ASPECTS region as a region of interest automatically triggers (e.g., by the control software of segmenter 3) the selection of a second region of interest as a complementary APSECTS region on the other side of the brain. When multiple images are received in step 210, for example when one of the images is a follow-up image, the regions of interest are in different ones of the images.

[0044]

[0048] At 230, intensity values ​​are extracted for each of the regions of interest designated by the information received in step 220. The intensity values ​​are expressed in Hounsfield Units (HU), which indicate how strongly the radiation is attenuated during scanning. The degree of attenuation (e.g., as indicated by the CT attenuation coefficient) correlates to tissue density. Thus, the HU values ​​provide a quantitative indication of the density (and thus the type of tissue) located in the designated regions of interest.

[0045]

[0049] For image visualization purposes, the HU values ​​correspond to voxel values ​​represented in a color or gray scale, e.g., various shades between white and black, inclusive. Lower density tissues are represented with darker shades (or intensities), whereas higher density tissues are represented with lighter shades (or intensities). In some examples, the HU or voxel values ​​are normalized or scaled, e.g., by shifting them by a predetermined constant value. After the HU values ​​are extracted for each of the two regions of interest in step 230, the HU values ​​are stored in corresponding tables, e.g., a first table contains the HU values ​​for the first region of interest and a second table contains the HU values ​​for the second region of interest.

[0046]

[0050] At 240, an intensity distribution is generated based on the extracted intensity values ​​for the regions of interest. In one embodiment, the intensity distribution is normalized and generated in the form of a histogram based on the extracted HU values ​​at step 230. For example, a first histogram is generated for the first region of interest and a second histogram is generated for the second region of interest. Thus, the histograms represent the distribution of all voxels within each one of the regions of interest. Each histogram is generated such that the values ​​along the x-axis correspond to the HU values ​​and the values ​​along the y-axis correspond to the number of voxels with the corresponding HU value among the HU values ​​in the region of interest.

[0047]

[0051] In one embodiment related to brain scans, the intensity histogram is generated on a bilateral basis, where a first region of interest includes an ASPECTS region in the left lateral portion of the brain, and a second region of interest includes the same ASPECTS region in the right lateral portion of the brain. In this case, the histogram provides an indication of the voxel content in the complementary regions of the brain. In generating the histogram, the range of HU values ​​is the entire range of HU values, or a predefined subset of values ​​within the entire range that are believed to be associated with, for example, a particular type of brain abnormality of interest.

[0048]

[0052] In one embodiment, the HU values ​​are optionally binned. In such a case, in each histogram, a range of HU values ​​is shown on the x-axis, and the values ​​on the y-axis correspond to the number of voxels in that range. In another embodiment, the intensity histogram has values ​​restricted to a range between HU values ​​of 10 and 60, spread over 25 bins, with each bin having a size of 2 HU. This range may be suitable for some applications, for example, a lower HU boundary of 10 does not include most of the CSF, and an upper HU boundary of 60 includes the entire gray matter but does not include calcified and hemorrhagic lesions. The histogram data is set based on different ranges of HU values ​​in another embodiment.

[0049]

[0053] After the histograms are generated, the values ​​in each histogram are subjected to a normalization process, which may involve, for example, dividing each y-value in each histogram by the total number of voxels in order to normalize the histogram voxel values ​​to be between 0 and 1, and so that the sum of all histogram voxel values ​​is 1.

[0050]

[0054] At 250, derivative histogram data is generated based on the histograms formed at step 240. The derivative histogram data includes, for example, difference histograms generated for the first and second regions of interest. For example, the difference histogram is generated based on the difference between the HU values ​​in the histogram of the first region of interest and the HU values ​​in the histogram of the second region of interest. When the two regions of interest are complementary ASPECTS regions of a brain scan image, the difference histogram provides a substantial (quantitative and qualitative) indication of how the intensity values ​​differ in those regions, and further characterizes the tissue type and other artifacts, if any, at their locations. This then allows for automatic selection of a viewing window for the image, by techniques described in more detail below.

[0051]

[0055] 4A shows an example of a difference histogram 430 (generalized to a curve) generated based on a histogram 410 (generalized to a curve) corresponding to a first region of interest in an image and a histogram 420 (generalized to a curve) corresponding to a second region of interest in an image. For illustrative purposes, the histogram 410 is labeled R1 and the histogram 420 is labeled R2. Thus, the difference histogram 430 is calculated by subtracting the intensity values ​​in the first histogram from the intensity values ​​in the first reference histogram, for example by calculating the formula R1-R2, thus considering the negative difference values ​​in the histogram 430.

[0052]

[0056] The resulting difference histogram 430 includes difference values ​​that fall into a range 440. It serves as a basis for characterizing tissue types or other artifacts in the first and second regions, and also provides a basis for determining window selection. An exemplary difference histogram (not generalized to a curve and) with normalized values ​​to difference values ​​within a given range on the y-axis versus HU values ​​on the x-axis is shown in FIG. 4B.

[0053]

[0057] At 260, a customized viewing window is automatically determined based on the difference histogram. Customization involves determining one or more parameters of the viewing window that allow subtle differences between two regions of interest to be discernible that would not be discernible if the same regions were displayed using a standard viewing window, such as a window with pre-set parameters typically used in the field of displaying CT images. This will be understood in more detail below.

[0054]

[0058] The viewing window for displaying a medical image is determined by the range of HU values ​​used to display the image. Standard viewing windows are based on a discrete number of preselected ranges of HU values ​​intended to enhance the display of certain types of tissue, bone, or other physiological features in the area scanned by the CT imager. For example, a standard window known as the "brain window" has a width of 80 HU values ​​and is centered at a HU value of 40. When such a viewing window is used, each voxel with a HU value of 0 is displayed as black, and each voxel with a HU value greater than 80 is displayed as white. HU values ​​within the window width are linearly mapped to grayscale values.

[0055]

[0059] The use of such windows limits (or hides) various types of physiological information that may be important for diagnosis and treatment. For example, the brain window described above allows certain types of brain tissue to be displayed, but hides other types of tissue, such as skull (bone) tissue. Thus, using the brain window where skull tissue is of more interest excludes information that may be of significant value for providing medical care.

[0056]

[0060] Furthermore, the use of standard (default) viewing windows is not optimally tailored, for example, when small differences are compared between regions of interest in the images (such as those contained in two bounding boxes). For example, in an ASPECTS application where the lenticule region in the left hemisphere is compared to the lenticule region in the right hemisphere, none of the standard viewing windows have parameters that allow for detailed visual evaluation between the two regions. Depending on how significantly these and other ASPECTS regions vary from patient to patient, this problem becomes more or less severe. Standard viewing windows do not resolve this difference between patients and are therefore ineffective. As a result, the preset HU ranges of the standard viewing windows hide certain important health conditions of patients that have clinical significance. Furthermore, the selection of which standard window to use is currently a manual decision based on the radiologist's judgment. Such an approach is time-consuming and prone to errors.

[0057]

[0061] To solve these and other problems, according to one or more embodiments, a customized range of HU values ​​is automatically determined for displaying the first and second regions of interest based on a difference histogram generated between those regions. Because the difference histogram is based on values ​​generated for each specific patient, the viewing window is customized to fit the specific condition of each patient, which improves or optimizes the use for comparison of ASPECTS regions that cannot be performed using, for example, standard viewing windows.

[0058]

[0062] Thus, for example, for each particular pair of regions of interest that do not fit into any of the standard viewing windows, a different viewing window is determined, which allows features that would otherwise be hidden by the standard viewing windows to be displayed in a precise and recognizable manner, which in turn allows health care practitioners to diagnose diseases or other health conditions and / or develop more effective treatment plans.

[0059]

[0063] In determining the viewing window, one custom parameter that is determined is the window width, e.g., a particular range of HU values ​​to be displayed. The window width determines the image contrast, e.g., the wider the width, the less contrast. Another custom parameter that is determined is the window level. The window level determines the brightness of the image, e.g., the higher the level, the more brightness. Setting the window level determines a median or midpoint grayscale value for the window width. Setting a custom window level allows different types of tissue in the first and second regions to become apparent, e.g., in cases where the different types of tissue would be hidden if a standard viewing window were used without this example.

[0060]

[0064] In one embodiment, the customized parameters of the viewing window are determined based on one or more predefined rules. The rules are generated, for example, to reveal features in the first and second regions that would be hidden if a standard viewing window were used to display the image without the present example. The rules thus function to customize the display of the image in a manner that improves (or even optimizes) the display and recognition of important clinical information contained in the intensity values ​​of the image.

[0061]

[0065] When the parameters include width and level, in one embodiment, the rules set forth in Equations 1 and 2 are applied to the values ​​in the difference histogram to calculate the width and level of the customized viewing window. Width=max{x|abs(DIFF_A_B(x))>0}-min{x|abs(DIFF_A_B(x))>0} (1) level=min{x|abs(DIFF_A_B(x))>0}+1 / 2width (2)

[0062]

[0066] In Equation 1 and Equation 2, the letter A corresponds to the histogram generated for the first region of interest, and the letter B corresponds to the histogram generated for the second region of interest. After these values ​​are obtained, a difference histogram DIFF_A_B is calculated by subtracting the histogram values ​​of the first region of interest from the corresponding histogram values ​​of the second region of interest. The width calculated by Equation 1 is based on deriving the maximum x value for which the difference value is greater than zero and the minimum x value for which the difference value is greater than zero, and taking the difference between these values. The midpoint value of the window is then set according to Equation 2.

[0063]

[0067] In another embodiment, different rules are applied to determine the width and level of the custom viewing window, for example, the width is set based on a predefined quantile of values ​​in the difference histogram shown in Equation 3, and the level is calculated based on Equation 4. width=scaler(DIFF_A_B,PERCENTAGE) (3) level=min{x|abs(DIFF_A_B(x))>0}+1 / 2width (4) where scaler is a function that ensures that a particular PERCENTAGE of the range of differences is included. In one embodiment, the PERCENTAGE is set to include 95% of the range (i.e., 0.95). This value is either pre-set by the system or allowed to be changed by the user.

[0064]

[0068] In other embodiments, different rules than those described above are used to automatically generate one or more parameters for a custom viewing window. In one embodiment, one or more look-up tables are stored in the system memory for use by the rules engine to compute the parameters of the viewing window. In this case, a predetermined set of values ​​for the corresponding difference histograms are pre-calculated and stored. In such a case, after knowing the difference histogram values ​​for the first and second regions of the particular image being analyzed, the rules engine performs a look-up of the look-up table to obtain the custom width and level of the viewing window.

[0065]

[0069] At 270, the region of interest (or the entire image including the region of interest) is displayed based on the custom viewing window automatically determined at step 260. Because the viewing window is customized to the intensity values ​​generated for a particular patient scan, features that would otherwise be hidden are displayed in a perceptible form. For example, differences in grayscale values ​​in a first region and a second region of the image become perceptible. These differences allow a physician, radiologist, or other health care practitioner to localize problems, abnormalities, or other effects that may result in more effective treatment procedures for a particular symptom experienced by the patient. This customized approach represents a significant improvement over when all brain scans are processed with the same subset of standard viewing windows.

[0066]

[0070] Figure 5A shows an image of a brain scan displayed using a standard viewing window, and Figure 5B shows an image of the same brain scan displayed using a custom viewing window generated in accordance with one or more embodiments. The brain scan is of a patient who has suffered a stroke, and the image has been segmented into ASPECTS regions corresponding to the superior image slices.

[0067]

[0071] In Fig. 5A, a difference histogram 510 is shown, which plots the intensity difference between regions including the putamen in the left and right hemispheres of the brain. These regions (e.g., corresponding to the left and right lenticular nucleus ASPECTS regions) are labeled with arrows 511 and 512 and are displayed using a standard viewing window. As shown in Fig. 5A, the width of the standard viewing window 515 spreads over a wide range of HU values, and most of the HU values ​​do not correspond to any of the difference values ​​in the difference histogram, e.g., the width of the standard viewing window is much wider than the range of HU values ​​in the difference histogram. As a result, the grayscale values ​​(and thus clinical information) in regions 511 and 512 are virtually indistinguishable in the left and right putamen regions.

[0068]

[0072] However, one of the putamen regions 512 contains a lesion resulting from a stroke. The presence of this lesion is not visible in image 520 due to the indistinguishable grayscale values ​​in regions 511 and 512 produced by the standard viewing window. Thus, the use of the standard viewing window hides clinical information that may be important in diagnosing and treating the patient.

[0069]

[0073] In FIG. 5B, a histogram 550 is shown depicting the intensity difference between the same putamen regions in the left and right hemispheres of the brain labeled 511 and 512. However, unlike in FIG. 5A, the brain scan image 660 is displayed using a custom viewing window generated according to an embodiment of the system and method described herein. Unlike the standard viewing window 515 applied in FIG. 5A, the custom viewing window 575 is adjusted to the values ​​in the difference histogram 550, e.g., in FIG. 5A, the standard viewing window 515 is centered on a HU value of 35 and has a width that extends from a HU value of 5 to a HU value of 65. In contrast, the custom viewing window 575 in FIG. 5B is centered on a HU value of 33 and has a width that extends from a HU value of 25 to a HU value of 41 HU, thus matching the values ​​in the difference histogram 550 generated from the image.

[0070]

[0074] As is evident from a comparison of Figures 5A and 5B, the image result displayed in Figure 5B using the custom viewing window 575 is very different from the image result displayed using the standard viewing window 515 in Figure 5A. When using the custom viewing window, the grayscale values ​​in the putamen region 512 (containing the stroke lesion) are substantially darker than the grayscale values ​​in the putamen region 511. This is a direct result of the customized viewing window and is clearly recognized by the healthcare professional. This comparison shows that important clinical findings are hidden by the use of the standard viewing window, but these clinical findings are easily visible when the custom viewing window is used to display the brain scan.

[0071]

[0075] In one embodiment, the first region of interest and the second region of interest are in different images, for example, the first region of interest is included in a first brain scan image acquired at a first time point, and the second region of interest is included in a second brain scan image from the same patient for the same corresponding area at a second subsequent time point. As shown so far, the region of interest is manually selected (e.g., via a bounding box) or is automatically generated, for example, by a segmentation algorithm. The difference between the first time point and the second time point is any length of time. Such an embodiment is useful, for example, to enable determination of whether any changes have occurred over time.

[0072]

[0076] For example, the first brain scan corresponds to FIG. 5B, where a lesion has formed in the putamen region as a result of a stroke. The second brain scan is acquired during a follow-up examination weeks, months, or years later, or during the occurrence of another complication. In such cases, a difference histogram is generated based on the HU values ​​extracted for the same putamen region of the patient. The difference histogram is then input to a rules engine to generate one or more parameters of a custom viewing window to identify the patient's condition. Before the difference histogram is generated, the images (or regions of interest) are overlaid to allow more accurate results to be generated.

[0073]

[0077] In some embodiments, more than one region of interest is compared. In such cases, three or more regions of interest (for the same or different images) are grouped into at least two meta regions of interest, each of which corresponds to a union of at least two of the three or more regions of interest. Take for example the case where three regions of interest are specified. The first meta region corresponds to the first region of interest and the second region of interest, the second meta region corresponds to the second region of interest and the third region of interest. In some examples, the third meta region corresponds to the first region and the third region.

[0074]

[0078] In such a case, a difference histogram is calculated for each of the meta regions based on the differences in the histograms (or voxel values) generated for the first region of interest, the second region of interest, and the third region of interest. Different viewing windows are then generated by applying one or more rules to the values ​​in the difference histograms, and the custom viewing windows are then used to display the image as described above. As with all embodiments described herein, instead of histograms, other types of normalized intensity distributions may be generated to generate the parameters of the custom viewing windows.

[0075]

[0079] Although CT scans are described as an example in some embodiments, the systems and methods described herein apply in other embodiments to other types of images, including, but not limited to, magnetic resonance imaging (MRI) scans.

[0076]

[0080] FIG. 6 illustrates an embodiment of a medical image analyzer 600 that performs embodiments of the methods described herein. With reference to FIG. 6, the medical image analyzer 600 includes a controller 610 and a memory 620. The controller executes instructions stored in the memory to perform the operations and methods described herein. In this embodiment, the instructions stored in the memory 620 include a first set of instructions 621 for performing a segmentation algorithm, a second set of instructions 622 for implementing a histogram generator, a third set of instructions 623 for implementing a difference histogram generator, and a fourth set of instructions for implementing a rules engine 624. These sets of instructions respectively perform the operations of the features in the medical image analyzer of FIG. 1, for example, and the operations of the method embodiments described herein. In this embodiment, the segmenter and labeler are included in the medical image analyzer. However, the segmenter and labeler may be combined with the medical image analyzer in some embodiments, for example, as shown in FIG. 1.

[0077]

[0081] In accordance with one or more of the foregoing embodiments, the methods, steps, and / or operations described herein are implemented by code or instructions executed by a computer, processor, controller, or other signal processing device that is in addition to or in addition to those described herein. Because the algorithms underlying the methods (or the operation of the computer, processor, controller, or other signal processing device) are described in detail, the code or instructions for implementing the operations of the method embodiments transform the computer, processor, controller, or other signal processing device into a special purpose processor for implementing the methods described herein.

[0078]

[0082] Further, another embodiment includes a computer readable medium, e.g., a non-transitory computer readable medium, for storing the above-mentioned code or instructions, such as a volatile or non-volatile memory or other storage device that is removably or permanently coupled to a computer, processor, controller, or other signal processing device for executing code or instructions to perform operations of the system and method embodiments described herein.

[0079]

[0083] The processors, systems, controllers, segmenters, generators, labelers, logic, engines, simulators, models, networks, scalers, and other signal generation and signal processing functions of the embodiments described herein may be implemented, for example, by logic including hardware, software, or both. When implemented at least partially in hardware, the processors, systems, controllers, segmenters, logic, engines, generators, labelers, simulators, models, networks, scalers, and other signal generation and signal processing functions may be any one of a variety of integrated circuits, including, for example, but not limited to, application specific integrated circuits, field programmable gate arrays, combinations of logic gates, systems on a chip, microprocessors, or other types of processing or control circuitry.

[0080]

[0084] When implemented at least in part in software, the processors, systems, controllers, segmenters, logic, engines, generators, labelers, simulators, models, networks, scalers, and other signal generation and signal processing functions may include, for example, memory or other storage devices for storing code or instructions executed by, for example, a computer, processor, microprocessor, controller, or other signal processing device, which may be in addition to those described herein or elements described herein. Because algorithms forming the basis of the methods (or the operation of a computer, processor, microprocessor, controller, or other signal processing device) are described in detail, the code or instructions for implementing the operations of the method embodiments may transform a computer, processor, controller, or other signal processing device into a special purpose processor for performing the methods described herein.

[0081]

[0085] Benefits, advantages, solutions to problems, and any elements which give rise to or make more evident any benefit, advantage, or solution are not to be construed as key, necessary, or essential features or elements of any or all claims. The present invention is defined solely by the appended claims, including any amendments made during the pendency of this application, and all equivalents of the claims as patented.

[0082]

[0086] Although various exemplary embodiments have been described in detail with particular reference to certain exemplary aspects thereof, it should be understood that the invention is capable of other exemplary embodiments and that its details can be modified in various obvious respects. As will be apparent to those skilled in the art, variations and modifications can be made while remaining within the spirit and scope of the invention. The embodiments can be combined to form further embodiments. Thus, the foregoing disclosure, description, and figures are for illustrative purposes only and do not limit the invention in any manner, the invention being defined by the claims. The embodiments can be combined to form further embodiments.

Claims

1. generating a first intensity distribution for a first image region; generating a second intensity distribution for a second image region; calculating a value based on the first intensity distribution and the second intensity distribution; automatically determining a custom viewing window based on the calculated value; displaying the first image region and the second image region using the custom viewing window; and determining the custom viewing window includes determining one or more parameters of the custom viewing window including at least one of a width or a level of the custom viewing window; Determining the one or more parameters comprises calculating the one or more parameters based on a difference value calculated between the first intensity distribution and the second intensity distribution. A computer-implemented method for processing medical image information.

2. determining the one or more parameters comprises applying at least one rule to the difference values ​​to determine the one or more parameters. The method of claim 1.

3. the width is based on a maximum intensity value in the difference value where the difference value is not zero and a minimum intensity value in the difference value where the difference value is not zero; The method of claim 1.

4. the value calculated based on the first intensity distribution and the second intensity distribution comprises a difference value between the first intensity distribution and the second intensity distribution. The method of claim 1.

5. the first image region and the second image region are present in the same image; The method of claim 1.

6. the first image region is a first area in a first brain scan; the second image region is a second area in the first brain scan; The method of claim 1.

7. the first image region is obtained from a first bounding box; the second image region is obtained from a second bounding box; The method of claim 6.

8. automatically identifying the first image region and the second image region based on a segmentation algorithm. The method of claim 6.

9. automatically identifying the second image region based on the designation of the first image region. The method of claim 6.

10. a generator for generating a first intensity distribution for a first image region and a second intensity distribution for a second image region; an arithmetic logic unit that calculates a value based on the first intensity distribution and the second intensity distribution; a rules engine that automatically determines a custom viewing window based on the calculated values ​​and outputs information defining the custom viewing window to a display for displaying the first image region and the second image region; Equipped with The rules engine determines one or more parameters of the custom viewing window, including at least one of a width or a level of the custom viewing window; and determining the one or more parameters comprises calculating the one or more parameters based on difference values ​​calculated between the first intensity distribution and the second intensity distribution. Medical image analyzer.

11. the rules engine applying at least one rule to the difference values ​​to determine the one or more parameters; 11. The medical image analyzer of claim 10.

12. the width is based on a maximum intensity value in the difference values, a minimum intensity value in the difference values, and a scalar; 11. The medical image analyzer of claim 10.

13. the first image region and the second image region are present in the same image; 11. The medical image analyzer of claim 10.

Images