Method and system for automatically propagating segmentations in medical images

Abstract

Disclosed herein is a method and system for automatically propagating segmentation in medical images. In one embodiment, the method uses a segmented reference region of interest (RoI) in a reference image to determine segmentation parameters and a plurality of reference points. The method further generates a plurality of shifted points on the current image by shifting the plurality of reference points onto the current image where a target RoI must be segmented. Subsequently, associated seeds are automatically selected from among the shifted points based on the segmentation parameters. Finally, a multi-seed segmentation of the selected associated seeds is performed to estimate and segment a target RoI in the current image, such that the target RoI is a propagated segmentation of the segmented RoI in the reference image.

Description

[Technical Field]

[0001] The present subject matter relates generally to the field of image processing techniques, and more particularly, but not exclusively, to methods and systems for automatically propagating segmentations in medical images. [Background technology]

[0002] Medical imaging has emerged as a major tool for the diagnosis of several diseases. Medical imaging is the process of creating a visual representation of the interior of the body for clinical analysis and medical intervention, as well as for functional analysis of the body's internal organs or tissues. In general, the visual representation may be in the form of images and videos. One of the first and most important processes used to analyze the visual representation is segmentation. Segmentation is the process of partitioning an image into distinct and meaningful segments that correspond to different tissue classes, organs, pathologies, or other biologically relevant structures. Summary of the Invention [Problem to be solved by the invention]

[0003] Generally, the human anatomy is composed of many types of tissues. As a result, in certain diagnostic processes, especially in tumor diagnosis, a single tissue may appear differently in different modalities. Therefore, using a common segmentation process, such as computer-based automated tumor segmentation, remains a continuing challenge in tumor diagnosis.

[0004] Furthermore, most of the existing segmentation procedures are tailored to segment tumors in images of a specific modality and with respect to a specific anatomical structure. Therefore, there is a need today for a segmentation technique that is general across modalities and anatomical structures.

[0005] The information disclosed in the Background section of this disclosure is intended solely to enhance understanding of the general background of the present invention and should not be taken as an admission or in any way as a suggestion that this information forms prior art already known to those skilled in the art. [Means for solving the problem]

[0006] Disclosed herein is a method for automatically propagating a segmentation in a medical image. The method includes determining one or more segmentation parameters based on an analysis of a segmented reference region of interest (reference RoI) in a reference image. The method further includes determining a plurality of reference points corresponding to the segmented reference RoI based on one or more morphological characteristics of the segmented reference RoI. Once the plurality of reference points are determined, the method further includes generating a plurality of translated points on a current image in which a target region of interest (target RoI) is to be segmented by translating each of the plurality of reference points onto the current image. The method further includes automatically selecting associated seeds in the current image from the plurality of translated points based on the one or more segmentation parameters. Finally, the method includes performing multi-seed segmentation of the selected associated seeds to estimate and segment a target RoI in the current image, the target RoI being a propagated segmentation of the segmented RoI in the reference image, and the pixel intensities of the current image being quantitatively equivalent to or can be normalized to the pixel intensities of the reference image.

[0007] The present disclosure further relates to an image segmentation system for automatically propagating segmentation in medical images. The image segmentation system includes a processor and a memory communicatively coupled to the processor. The memory stores processor-executable instructions that, when executed, cause the processor to determine one or more segmentation parameters based on an analysis of a segmented reference region of interest in a reference image. The instructions further cause the processor to determine a plurality of reference points corresponding to a segmented reference region of interest based on one or more morphological characteristics of the segmented reference region of interest. Thereafter, the instructions cause the processor to generate a plurality of shifted points on a current image from which a target region of interest must be segmented by shifting each of the plurality of reference points onto the current image. The instructions further cause the processor to automatically select associated seeds in the current image from the plurality of shifted points based on the one or more segmentation parameters. Finally, the instructions cause the processor to perform a multi-seed segmentation of the selected associated seeds to estimate and segment a target RoI in the current image, the target RoI being a propagated segmentation of the segmented RoI in the reference image, and the pixel intensities of the current image being quantitatively equivalent to or can be normalized to the pixel intensities of the reference image.

[0008] The foregoing summary of the invention is illustrative only and is not intended to be limiting in any way. In addition to the exemplary aspects, embodiments, and features described above, other aspects, embodiments, and features will become apparent by reference to the drawings and the following detailed description.

[0009] The accompanying drawings, which are incorporated in and constitute a part of this disclosure, illustrate exemplary embodiments and, together with the description, explain the disclosed principles. In the drawings, the leftmost digit(s) of a reference number identifies the figure in which the reference number first appears. Like numbers are used throughout the figures to refer to like features and components. Some embodiments of systems and / or methods according to embodiments of the present subject matter will now be described, by way of example only, with reference to the accompanying drawings, in which: [Brief explanation of the drawings]

[0010] [Figure 1] FIG. 1 illustrates an exemplary configuration example for automatically propagating segmentation in medical images, according to some embodiments of the present disclosure. [Figure 2] FIG. 1 is a detailed block diagram illustrating an image segmentation system for automatically propagating segmentation, according to some embodiments of the present disclosure. [Figure 3A] 10A-10C illustrate a process of propagating segmentation for a reference image and a current image according to an exemplary embodiment of the present disclosure. [Figure 3B] 10A-10C illustrate a process of propagating segmentation for a reference image and a current image according to an exemplary embodiment of the present disclosure. [Figure 3C] 10A-10C illustrate a process of propagating segmentation for a reference image and a current image according to an exemplary embodiment of the present disclosure. [Figure 3D] 10A-10C illustrate a process of propagating segmentation for a reference image and a current image according to an exemplary embodiment of the present disclosure. [Figure 3E] 10A-10C illustrate a process of propagating segmentation for a reference image and a current image according to an exemplary embodiment of the present disclosure. [Figure 3F] 10A-10C illustrate a process of propagating segmentation for a reference image and a current image according to an exemplary embodiment of the present disclosure. [Figure 3G] 10A-10C illustrate a process of propagating segmentation for a reference image and a current image according to an exemplary embodiment of the present disclosure. [Figure 3H] 10A-10C illustrate a process of propagating segmentation for a reference image and a current image according to an exemplary embodiment of the present disclosure. [Figure 4] 1 is a flowchart illustrating a method for automatically propagating segmentation in a medical image, according to some embodiments of the present disclosure. [Figure 5] FIG. 1 is a block diagram of an exemplary computer system for implementing embodiments consistent with the present disclosure. DETAILED DESCRIPTION OF THE INVENTION

[0011] Those skilled in the art will appreciate that any block diagrams herein represent conceptual views of illustrative systems embodying the principles of the present subject matter. Similarly, any flow diagrams, flow charts, state transition diagrams, pseudocode, or the like, may be substantially embodied in a computer-readable medium and may represent various processes that may be executed by a computer or processor, whether or not such a computer or processor is explicitly shown.

[0012] The word "exemplary" is used herein to mean "serving as an example, instance, or illustration." Any embodiment or implementation of the present subject matter described herein as "exemplary" is not necessarily to be construed as preferred or advantageous over other embodiments.

[0013] While the present disclosure is susceptible to various modifications and alternative forms, specific embodiments thereof have been shown by way of example in the drawings and are described in detail below. It is to be understood, however, that it is not intended to limit the disclosure to the particular forms disclosed, and the invention is intended to cover all modifications, equivalents, and alternatives falling within the spirit and scope of the invention.

[0014] The term "comprises, comprising, includes" or any other variant thereof is intended to cover a non-exclusive inclusion, so that a setup, apparatus, or method having listed components or steps does not include only those components or steps, but may also include other components or steps inherent in such setup or apparatus or method or not expressly listed. In other words, one or more components in a system or apparatus followed by the phrase "comprises ... a" does not, without more constraints, exclude the presence of other or additional components in the system or method.

[0015] Embodiments of the present disclosure can be used to automatically propagate lesion segmentations through one or more follow-up sessions for oncology during routine scans of lesions and / or other diseased areas.

[0016] Therefore, according to one embodiment, the present disclosure discloses a method and an image segmentation system for automatically propagating segmentation in medical images. According to one embodiment, the disclosed method includes picking corresponding fiducial points of segmented regions of interest (RoIs) of a reference image and / or a group of reference images, and moving the fiducial points to a current image of the lesion captured in a follow-up session to perform multi-seed segmentation. The multi-seed segmentation of the selected relevant fiducial points on the current image helps to estimate and segment a target RoI in the current image. According to one embodiment, the target RoI can be a propagated segmentation of the segmented RoI in the reference image. Thus, the present disclosure helps to automatically propagate segmentation in medical images.

[0017] According to one embodiment, the present disclosure performs lesion segmentation for any anatomical region and any modality to infer necessary parameters from the segmented lesion. By way of example, parameters that may be inferred may include, but are not limited to, size, shape, location, intensity distribution of the lesion, homogeneity / heterogeneity of the lesion, and the appearance of background tissue surrounding the lesion.

[0018] According to one embodiment, the present disclosure addresses the technical problem of iterative segmentation of lesions in follow-up scans by automating the propagation of the segmentation using information from the lesions segmented in the first scan.

[0019] In one embodiment, the methods and image segmentation systems disclosed in this disclosure can be used to propagate segmentations of objects / entities with similar characteristics across two or more studies, such as scans, images, or volumes, when the segmentation of the object / entity is available in at least one of such studies.

[0020] In the detailed description of aspects of the present disclosure, reference is made to the accompanying drawings that form a part hereof, and in which is shown by way of illustration specific embodiments in which the invention may be practiced. These embodiments are described in sufficient detail to enable those skilled in the art to practice the invention, and it should be understood that other embodiments may be utilized and changes may be made without departing from the scope of the present disclosure. Accordingly, the following description is not to be taken in a limiting sense.

[0021] FIG. 1 illustrates an exemplary configuration example for automatically propagating segmentation in medical images, according to some embodiments of the present disclosure.

[0022] The example configuration 100 may include an image segmentation system 101 and a reference database 105 associated with the image segmentation system 101. According to one embodiment, the image segmentation system 101 may be a computing device, such as, but not limited to, a desktop computer, laptop, smartphone, or server, and may be configured to automatically propagate segmentation within medical images in accordance with embodiments of the present disclosure. According to one embodiment, the reference database 105 may be a storage unit used to store patient-related information, history medical images and / or history reference images 103 associated with the patient, and other information necessary for propagating the segmentation. In one implementation, the reference database 105 may be part of or reside within the image segmentation system 101.

[0023] According to one embodiment, the reference image 103 and the current image 107 may be images of entities, such as organs, tissues, bones, air cavities, and muscles, of a subject, acquired at different time periods. Here, the subject may be a human, an animal, etc. Further, by way of example, the reference image 103 may be an image of the entity acquired during an initial study of the entity, from which a reference region of interest (RoI) can be segmented. The current image 107 may be an image of the same entity acquired during a subsequent study of the subject. According to one embodiment, the reference image 103 and the current image 107 may be acquired by an external device, such as a computed tomography (CT) scanner, a magnetic resonance imaging (MRI) device, etc. According to one embodiment, the image segmentation system 101 may receive the reference image 103 and the current image 107 from one of the medical devices or from a reference database using a wired and / or wireless communication interface configured between the image segmentation system 101 and the external device. In one embodiment, the image segmentation system 101 and the external device can be implemented as a single system that performs both scanning and segmentation.

[0024] In one embodiment, once a reference image 103 of an entity is acquired, a technical expert associated with the image segmentation system 101 can manually perform segmentation of the reference image 103 to mark the reference image 103 and / or derive a reference RoI from the reference image. Subsequently, once a current image 107 of the entity is acquired, the current image 107, along with the reference image 103 and the segmented reference RoI, can be provided as input to the image segmentation system 101 for automatically segmenting the current image 107.

[0025] According to one embodiment, upon receiving the reference image 103, the current image 107, and the reference RoI, the image segmentation system 101 may determine one or more segmentation parameters associated with the reference image 103 based on an analysis of the segmented reference RoI in the reference image 103. According to one embodiment, the one or more segmentation parameters determined from the reference image 103 may include, but are not limited to, one or more morphological characteristics of the segmented reference RoI, the position of the segmented reference RoI within the reference image 103, the intensity distribution of the segmented reference RoI, and the background intensity distribution of the background of the segmented reference RoI.

[0026] According to one embodiment, the one or more morphological characteristics may include, but are not limited to, a minor axis and a longest diameter of the segmented RoI of reference, and a size and shape of the segmented RoI of reference. According to one embodiment, the size and shape of the segmented RoI of reference can be determined based on the length and coordinate of the minor axis and the longest diameter of the segmented RoI of reference.

[0027] According to one embodiment, the intensity distribution of the segmented RoI of reference can be the variation of pixel intensities in the region of the segmented RoI of reference. Similarly, the background intensity distribution of the background of the segmented RoI of reference can be the variation of pixel intensities in the region surrounding the segmented RoI of reference. In one implementation, the background intensity distribution of the segmented RoI of reference can be determined by determining multiple intensity distributions of multiple background pixels / points corresponding to quadrants of the segmented RoI of reference with respect to the center of the segmented RoI of reference. According to one embodiment, the range of the intensity distributions can be determined based on the amount of overlap between the pixel intensities of the background and the pixel intensities of the segmented RoI of reference in each of the multiple background quadrants.

[0028] According to one embodiment, after determining one or more segmentation parameters from the segmented RoI of reference, the image segmentation system 101 can determine a plurality of reference points corresponding to the segmented RoI of reference based on one or more morphological characteristics of the segmented RoI of reference. According to one embodiment, the plurality of reference points can be points located along the minor axis and the longest diameter of the segmented RoI of reference.

[0029] According to one embodiment, after determining the plurality of reference points, the image segmentation system 101 can generate a plurality of displaced points on the current image 107 of the entity from which the target RoI 109 has to be segmented. According to one embodiment, the plurality of displaced points can be obtained by displacing each of the plurality of reference points onto the current image 107.

[0030] In one embodiment, once multiple displaced points are generated, the image segmentation system 101 can automatically select an associated seed from among the multiple displaced points in the current image 107 based on one or more segmentation parameters.

[0031] According to one embodiment, once relevant seeds are selected from the multiple moved points, the image segmentation system 101 can perform multi-seed segmentation of the selected relevant seeds to estimate and segment a target RoI 109 in the current image 107. According to one embodiment, the target RoI 109 may correspond to a propagated segmentation of the segmented RoI in the reference image 103. According to one embodiment, a prerequisite for performing segmentation of the target RoI 109 may be that the pixel intensities of the current image 107 must be quantitatively equivalent to or normalized to the pixel intensities of the reference image 103.

[0032] According to one embodiment, after the target RoI 109 is segmented from the current image 107, the target RoI 109 can be compared with the segmented reference RoI to determine changes in one or more segmentation parameters of each of the segmented reference RoI and the target RoI 109. According to one embodiment, during subsequent studies of the entity, the current image 107 and the segmented target RoI 109 of the entity together can be considered as the reference image 103 with the segmented reference RoI.

[0033] FIG. 2 shows a detailed block diagram illustrating an image segmentation system 101 for automatically propagating segmentation, according to some embodiments of the present disclosure.

[0034] In one implementation, the image segmentation system 101 may include an I / O interface 201, a processor 203, and a memory 205. The I / O interface 201 may be configured to communicate with one or more sources and / or external devices to receive the reference image 103 and the current image 107. Furthermore, the I / O interface 201 may be used to connect the image segmentation system 101 to a display interface for displaying the reference image 103, the current image 107, and segmented regions of the image to a user. According to one embodiment, the memory 205 may be communicatively coupled to the processor 203. The processor 203 may be configured to perform one or more functions of the image segmentation system 101.

[0035] In some implementations, the image segmentation system 101 may have modules 209 and data 207 for performing various processes according to embodiments of the present disclosure. According to one embodiment, the data 207 may be stored in memory 205 and may include, but is not limited to, one or more segmentation parameters 211, a plurality of reference points 213, a plurality of moved points 215, a target region of interest 217, and other data 219.

[0036] In some embodiments, the data 207 can be stored in the memory 205 in the form of various data structures. Furthermore, the data 207 can be organized using a data model such as a relational data model or a hierarchical data model. The other data 219 can include temporary data and temporary files generated by the module 209 while performing various functions of the image segmentation system 101. By way of example, the other data 219 can include, but is not limited to, one or more history or reference images 103 of the entity, morphological characteristics of the segmented RoI, etc.

[0037] According to one embodiment, data 207 may be processed by one or more modules 209 of image segmentation system 101. As used herein, the term module refers to an application specific integrated circuit (ASIC), electronic circuitry, processor (shared, dedicated, or group), and memory executing one or more software or firmware programs, combinatorial logic circuits, and / or other suitable components that provide the described functionality. According to one embodiment, other modules 231 may be used to perform various and diverse functions of image segmentation system 101. It should be understood that such modules 209 may be represented as a single module or a combination of different modules.

[0038] In one embodiment, the one or more modules 209 may be stored as instructions executable by the processor 203. In another embodiment, each of the one or more modules 209 may be a separate hardware unit communicatively coupled to the processor 203 to perform one or more functions of the image segmentation system 101. The one or more modules 209 may include, but are not limited to, a parameter determination module 221, a fiducial point determination module 223, a seed generation module 225, a seed selection module 227, a segmentation module 229, and other modules 231.

[0039] According to one embodiment, the parameter determination module 221 can be used to determine one or more segmentation parameters 211 based on an analysis of the segmented reference RoI in the reference image 103. According to one embodiment, the first step towards propagating the segmentation of the reference image 103 can be the determination of one or more segmentation parameters 211, such as one or more morphological characteristics of the segmented reference RoI, the position of the segmented reference RoI in the reference image 103, the intensity distribution of the segmented reference RoI, and the intensity distribution of the background of the segmented reference RoI.

[0040] According to one embodiment, the parameter determination module 221 can determine one or more morphological characteristics of the segmented RoI of reference by determining the minor axis and the longest diameter of the segmented RoI of reference, and then determining the size and shape of the segmented RoI of reference based on the lengths and coordinates of the minor axis and the longest diameter of the segmented RoI of reference. Furthermore, the position of the segmented RoI of reference in the reference image 103 can be determined based on the coordinates of the point on the minor axis and the longest diameter.

[0041] According to one embodiment, the parameter determination module 221 can determine the intensity distribution of the segmented RoI of reference by calculating a histogram of the intensities of the pixels forming the segmented RoI of reference. Furthermore, the parameter determination module 221 can derive the heterogeneity of the segmented RoI of reference by thresholding the histogram at a predetermined frequency. For example, pixel intensity values ​​having a frequency value greater than 60% of the highest frequency value in the histogram can be considered. The thresholding can result in multiple pixel intensity ranges, referred to as lesion ranges (Lr). In one embodiment, if the segmented RoI of reference is homogeneous, there can be only a single pixel intensity range. On the other hand, if the segmented RoI of reference is heterogeneous, the thresholding of the histogram can result in multiple pixel intensity ranges.

[0042] According to one embodiment, the parameter determination module 221 can determine the intensity distribution of the background and / or surrounding regions of the segmented reference RoI by a process similar to calculating a histogram. That is, the intensity distribution can be determined by calculating a histogram and then thresholding the histogram at a predetermined frequency. However, this can also be done by grouping pixels into four quadrants based on their coordinates, with the center of the segmented reference RoI treated as the origin for the quadrant. As a result, there may be four histograms corresponding to each of the four quadrants and four groups of pixel intensity ranges resulting from thresholding the histograms. These pixel intensity ranges can be referred to as background ranges (BGr).

[0043] According to one embodiment, after obtaining Lr and four BGr, the parameter determination module 221 can calculate a neighbor threshold (NT) and a background histogram percent (BbHp). In one embodiment, NT can represent the minimum number of neighboring pixels of a point that must satisfy the Lr and BGr criteria for the point to be narrowed down and / or selected as a reference point. According to one embodiment, BgHp can represent the actual background histogram threshold percentage that can be considered for intensity comparison to narrow down the associated reference seed.

[0044] According to one embodiment, the parameter determination module 221 can determine background parameters using the following method. First, for each quadrant, the background histogram can be thresholded at multiple frequency ranges, such as 10%, 30%, 50%, and 70%. Thus, at the end of thresholding, BGr can be obtained for every frequency threshold for each quadrant of background pixels. According to one embodiment, the overlap between Lr and each of BGr can be calculated, and the values ​​of NT and BgHp can be iteratively derived, as illustrated in Tables A and B below.

[0045] According to one embodiment, the criteria for deriving parameters may vary for different types and / or sources of images. For example, criteria for deriving parameters for images obtained from CT (Computed Tomography) may be shown in Table A. Similarly, criteria for deriving parameters for images obtained from Magnetic Resonance Imaging (MRI) may be shown in Table B. TIFF0007800423000001.tif79161 TIFF0007800423000002.tif93162

[0046] According to one embodiment, the overlap percentage (shown in Tables A and B) may be the percentage of Lr that overlaps with BGr. Additionally, the overlap count (shown in Table B) may be the number of quadrants of BGr that overlap with Lr.

[0047] According to one embodiment, the reference point determination module 223 can be used to determine a plurality of reference points 213 corresponding to the segmented RoI of reference based on one or more morphological characteristics of the segmented RoI of reference. In one embodiment, the longest diameter and shortest axis of the segmented RoI of reference, which are determined based on a reference study of the reference image 103, lie on a cross-section of the RoI, so the longest diameter and shortest axis can be considered to cover most of the distinct intensity region within the segmented RoI. Therefore, in one embodiment, the plurality of reference points 213 on the segmented RoI of reference can be determined as coordinates of distinct points identified on the longest diameter and shortest axis.

[0048] According to one embodiment, the process of obtaining a reference RoI and determining a plurality of reference points 213 on the reference RoI can be illustrated using the exemplary representations in FIGS. 3A-3C.

[0049] Figure 3A shows a reference image 103 including a reference RoI 303. Figure 3B shows the longest diameter 303A and the short axis 303B of the segmented reference RoI 303. Here, multiple reference points 213 corresponding to the segmented reference RoI 303 can be determined by identifying distinct sets of points on the longest diameter 303A and the short axis 303B of the segmented reference RoI 303, as shown in Figure 3C.

[0050] According to one embodiment, the seed generation module 225 can be used to generate a plurality of displaced points 215 on the current image 107 by displacing each of a plurality of reference points 213 onto the current image 107 from which the target RoI 109 has to be segmented. According to one embodiment, once all of the reference points 213 have been determined and extracted from the reference image 103, the seed generation module 225 can transform each of the plurality of reference points 213 onto the current image 107 to obtain a plurality of displaced points 215 on the current image 107. That is, the reference points 213 determined from the reference image 103 can be transformed into the current image 107 as displaced points 215, as shown in FIG. 3D .

[0051] According to one embodiment, the seed selection module 227 can be used to automatically select relevant seeds 311 in the current image 107 from the plurality of shifted points 215 based on one or more segmentation parameters 211. Figure 3E shows relevant seeds 311 narrowed down and selected from the shifted points 215. According to one embodiment, the pixel intensity ranges corresponding to the segmented reference RoI 303 and for each of the four quadrants of background pixels may be the most important segmentation parameters 211 used to select relevant seeds 311 in the current image 107.

[0052] According to one embodiment, one of the multiple moved points 215 “P” can be narrowed down as a relevant seed only if the seed point “P” satisfies one or more conditions defined below: 1. The pixel intensity of point P is within the range defined by Lr. 2. The pixel intensity of point P is outside the background intensity range BGr, where the background intensity range may be the set of intensity ranges obtained by thresholding the histogram by the BgHp of the background pixels of the quadrant to which point P belongs. 3. The number of pixels directly adjacent to point P that satisfy conditions 1 and 2 above is greater than or equal to NT. 4. The intensity Ip at point P is: Ip < Lr < BGr or Ip > Lr > BGr

[0053] According to one embodiment, point P can be narrowed down if point P satisfies all three conditions 1, 2 and 3 above, or if point P satisfies only condition 4 above.

[0054] In one embodiment, if point P is not refined (selected), the neighboring points of point P are checked and can be refined if they satisfy conditions 1, 2 and 3 above.

[0055] According to one embodiment, if none of the multiple moved points 215 are narrowed down after repeating the above analysis for all of the multiple moved points 215, a moved point 215 that only satisfies condition 1 can also be narrowed down as the associated seed 311.

[0056] According to one embodiment, for CT images, the bilaterally filtered pixel intensity of point P can be used for all of the above conditions 1 to 4. On the other hand, for MR images, no filtering is applied and pixel intensities can be directly compared. Also, in CT images, intensity represents quantitative information, so two CT scans can be directly compared. However, in MR images, pixel intensity only represents qualitative information, so two scans of the same application may first need to be normalized before being compared.

[0057] In one embodiment, the segmentation module 229 can be used to perform multi-seed segmentation of the selected associated seeds 311 to estimate and segment the target RoI 109 in the current image 107. In one implementation, multi-seed segmentation can be considered as an extension of the single-seed segmentation technique performed for all selected associated seed points obtained in the above process to address the heterogeneity of the reference RoI. According to one embodiment, the multi-seed segmentation of the selected associated seeds 311 can be performed using one of the existing multi-seed segmentation techniques, such as a parametric method, a level set method, or a clustering method. Alternatively, the multi-seed segmentation can be performed using a region-growing technique, as exemplified in this disclosure.

[0058] According to one embodiment, the segmentation module 229 can receive user input indicating whether the target RoI 109 determined in the current image 107 is larger than, smaller than, or the same size as that of the reference RoI determined in the reference image 103. This user input can be used to limit the area searched and / or scanned while performing multi-seed segmentation.

[0059] In one embodiment, if the target RoI 109 in the current image 107 is larger than that of the reference RoI, the search region radius can be set equal to the longest diameter of the reference RoI.

[0060] In one embodiment, if the target RoI 109 is smaller than that of the reference RoI, the search region radius can be set to 50% of the longest diameter of the reference lesion.

[0061] In one embodiment, if the target RoI 109 is the same size as the reference RoI, the search region radius can be set to 75% of the longest diameter of the reference lesion.

[0062] That is, the difference between the size of the target RoI 109 and the size of the reference RoI can be correlated and adjusted using one of the adjustments described above.

[0063] 3F shows the regions obtained by performing multi-seed segmentation of the associated seeds 311 on the current image 107. According to one embodiment, the multi-seed segmentation may exclude some regions of the target RoI 109 that have different intensities compared to the overall intensity of the RoI from the generation of the target RoI 109. However, this can be addressed by using an intensity-based k-means clustering region-growing technique within the region defined by the multi-seed segmentation output.

[0064] In one embodiment, the regions obtained by performing multi-seed segmentation of the associated seeds 311 can be used to determine an overlap region 313 on the current image 107. The overlap region 313 can then be extracted from the current image 107 and treated as the target RoI 109 corresponding to the current image 107, as shown in Figure 3G. That is, Figure 3G shows the target RoI 109 extracted by performing the above steps on the current image 107.

[0065] According to one embodiment, as shown in FIG. 3H, a comparison of the reference RoI and the target RoI 109 helps determine the change in RoI between the reference image 103 and the current image 107 of the same entity.

[0066] FIG. 4 shows a flowchart illustrating a method for automatically propagating segmented medical images according to some embodiments of the present disclosure.

[0067] As illustrated in Figure 4, method 400 includes one or more blocks illustrating a method for automatically propagating a segmentation in a medical image using the image segmentation system 101 illustrated in Figure 1. Method 400 may be described in the general context of computer-executable instructions. Generally, computer-executable instructions may include routines, programs, objects, components, data structures, procedures, modules, and functions that perform particular functions or implement particular abstract data types.

[0068] The order in which method 400 is described is not intended to be construed as a limitation, and any number of the described method blocks can be combined to implement the method. Additionally, individual blocks can be deleted from the method without departing from the spirit and scope of the subject matter described herein. Furthermore, the method can be implemented in any suitable hardware, software, firmware, or combination thereof.

[0069] In block 401, the method 400 includes determining one or more segmentation parameters 211 based on an analysis of a segmented reference region of interest (RoI) in the reference image 103. According to an embodiment, determining the one or more segmentation parameters 211 may include, but is not limited to, determining one or more morphological characteristics of the segmented reference RoI 303, the location of the segmented reference RoI 303 in the reference image 103, the intensity distribution of the segmented reference RoI 303, and the background intensity distribution of the background of the segmented reference RoI 303.

[0070] According to one embodiment, the background intensity distribution of the background can be determined using the following method: 1. By determining a plurality of intensity distributions of a plurality of background pixels / points corresponding to quadrants with respect to the center of the segmented reference RoI 303. According to one embodiment, the plurality of intensity distributions can be determined based on the amount of overlap between the pixel intensities of the background of each of the plurality of background quadrants and the pixel intensities of the segmented reference RoI 303. 2. By grouping background pixels into a plurality of background quadrants with the center of the segmented reference RoI 303 as the origin. Further, a background threshold corresponding to each of the plurality of background quadrants is determined based on the amount of overlap between the pixel intensities of the respective background quadrants and the pixel intensities of the segmented reference RoI 303. Finally, a background intensity distribution corresponding to each of the background quadrants can be determined based on the corresponding background threshold.

[0071] In one embodiment, the intensity distribution of the segmented reference RoI 303 can be determined by generating a histogram of pixels in the segmented reference RoI 303 and determining one or more intensity ranges of the segmented reference RoI 303 based on the histogram and a predetermined threshold. According to one embodiment, the one or more intensity ranges can represent the intensity distribution of the segmented reference RoI 303.

[0072] In block 403, the method 400 includes determining a plurality of reference points 213 corresponding to the segmented RoI of reference 303 based on one or more morphological characteristics of the segmented RoI of reference 303. According to one embodiment, determining the one or more morphological characteristics of the segmented RoI of reference 303 may include, but is not limited to, determining a minor axis 303B and a longest diameter 303A of the segmented RoI of reference 303. Furthermore, determining the one or more morphological characteristics may include determining a size and shape of the segmented RoI of reference 303 based on the lengths and coordinates of the minor axis 303B and the longest diameter 303A of the segmented RoI of reference 303.

[0073] In block 405, the method 400 includes generating a plurality of moved points 215 on the current image 107 by moving each of the plurality of reference points 213 onto the current image 107 in which the target RoI 109 has to be segmented. According to one embodiment, the reference image 103 and the current image 107 may be images of a single entity taken at different time periods.

[0074] In block 407, the method 400 includes automatically selecting relevant seeds 311 in the current image 107 from the plurality of displaced points 215 based on the one or more segmentation parameters 211. According to one embodiment, the relevant seeds 311 in the current image 107 can be automatically selected by selecting one or more displaced points 215 from the plurality of displaced points 215 if the intensity of the one or more displaced points 215 falls within the intensity distribution of the segmented reference RoI 303. Furthermore, the relevant seeds 311 in the current image can also be automatically selected based on the following criteria: 1. If the intensity of one or more displaced points 215 does not fall into the background intensity distribution of the segmented reference RoI 303. 2. If the intensity of at least a predetermined number of neighboring pixels of one or more displaced points 215 falls within the intensity distribution of the segmented reference RoI 303. 3. If the intensities of at least a predetermined number of neighboring pixels of one or more shifted points 215 do not fall within the background intensity distribution range of the background. 4. If the intensity of one or more displaced points 215 is smaller than the intensity distribution range of the segmented reference RoI 303 and the background intensity distribution range of the background. 5. If the intensity of one or more displaced points 215 is greater than the highest intensity value of the intensity distribution of the segmented reference RoI 303 and the highest intensity value of the background intensity distribution range of the background. 6. In block 409, the method 400 includes performing a multi-seed segmentation of the selected associated seeds 311 to estimate and segment the target RoI 109 in the current image 107. According to one embodiment, the target RoI 109 is a propagated segmentation of the segmented RoI in the reference image 103. According to one embodiment, a prerequisite for performing the segmentation of the target RoI 109 may be that the pixel intensities of the current image 107 must be quantitatively equivalent to or normalized to the pixel intensities of the reference image 103.

[0075] Computer Systems 5 illustrates a block diagram of an exemplary computer system 500 for implementing embodiments according to the present disclosure. According to one embodiment, the computer system 500 can be the image segmentation system 101, which is used to automatically propagate segmentations in medical images. The computer system 500 can have a central processing unit 502. The processor 502 can have at least one data processor that executes program components for executing user- or system-generated business processes. A user can include a person, a patient, medical personnel and / or technicians, someone using the image segmentation system 101, etc. The processor 502 can include specialized processing units, such as an integrated system (bus) controller, a memory management control unit, a floating-point unit, a graphics processing unit, a digital signal processing unit, etc.

[0076] The processor 502 can be arranged to communicate with one or more input / output devices (511 and 512) via an I / O interface 501. The I / O interface 501 may use communication protocols / methods such as, but not limited to, audio, analog, digital, stereo, IEEE-1394, serial bus, universal serial bus (USB), infrared, PS / 2, BNC, coaxial, component, composite, digital visual interface (DVI), high-definition multimedia interface (HDMI), radio frequency (RF), S-video, video graphics array (VGA), IEEE 802.n / b / g / n / x, Bluetooth, cellular (e.g., code division multiple access (CDMA), high-speed packet access (HSPA+), global system for mobile communications (GSM), long-term evolution (LTE), etc.). Using the I / O interface 501, the computer system 500 can communicate with one or more I / O devices 511 and 512.

[0077] In one embodiment, the processor 502 can be arranged to communicate with a communication network 509 via a network interface 503. The network interface 503 can communicate with the communication network 509. The network interface 503 can employ connection protocols including, but not limited to, direct connection, Ethernet (e.g., twisted pair 10 / 100 / 1000 Base T), Transmission Control Protocol / Internet Protocol (TCP / IP), Token Ring, IEEE 802.11a / b / g / n / x, etc. Using the network interface 503 and the communication network 509, the computer system 500 can communicate with a reference database 105 to receive segmented reference regions of interest (RoIs) of the reference images 103. Furthermore, the communication network 509 can be used to receive the reference images 103 and the current image 107 from one or more sources, such as an X-ray scanner.

[0078] The communications network 509 can be implemented as one of several types of networks, such as an intranet or a local area network, and as such within an organization. The communications network 509 can be either a dedicated network or a shared network representing an association of several types of networks that use various protocols, such as Hypertext Transfer Protocol, Transmission Control Protocol / Internet Protocol, Wireless Application Protocol, etc., to communicate with each other. Furthermore, the communications network 509 can include various network devices, including routers, bridges, servers, computing devices, storage devices, etc.

[0079] In some embodiments, the processor 502 can be arranged to communicate with memory 505 (e.g., RAM 513 and ROM 514 as shown in FIG. 5) via a storage interface 504. The storage interface 504 can connect to memory 505, including but not limited to a memory drive, removable, etc., which may employ connection protocols such as Serial Advanced Technology Attachment (SATA), Integrated Drive Electronics (IDE), IEEE-1394, Universal Serial Bus (USB), Fibre Channel, Small Computer System Interface (SCSI), etc. The memory drive can further include a drum, magnetic disk drive, magneto-optical drive, optical drive, redundant array of independent disks (RAID), solid state memory device, solid state drive, etc.

[0080] Memory 505 can store a collection of program or database components, including, but not limited to, user / applications 506, an operating system 507, a web browser 508, etc. In some embodiments, computer system 500 can store user / application data 506, such as data, variables, records, etc., as described in this disclosure. Such databases can be implemented as fault-tolerant, relational, scalable, secure databases (e.g., Oracle or Sybase).

[0081] Operating system 507 may facilitate resource management and operation of computer system 500. Examples of operating systems include, but are not limited to, Apple Macintosh OS X, UNIX, Unix-like system distributions (e.g., Berkeley Software Distribution (BSD), FreeBSD, NetBSD, OpenBSD, etc.), Linux distributions (e.g., Red Hat, Ubuntu, K-Ubuntu, etc.), IBM OS / 2, Microsoft Windows (XP, Vista / 7 / 8, etc.), Apple iOS, Google Android, Blackberry Operating System (OS), etc.

[0082] A user interface may facilitate the display, instruction, interaction, operation, or manipulation of program components through textual or graphical facilities. For example, a user interface may provide computer interaction interface elements, such as cursors, icons, check boxes, menus, windows, widgets, etc., on a display system operatively connected to computer system 500. Graphical user interfaces (GUIs) include, but are not limited to, Apple Macintosh operating system Aqua, IBM OS / 2, Microsoft Windows (e.g., Aero, Metro, etc.), Unix-X-Windows, web interface libraries (e.g., ActiveX, Java, JavaScript, AJAX, HTML, Adobe Flash, etc.), etc.

[0083] Furthermore, one or more computer-readable storage media can be utilized in implementing embodiments in accordance with the present invention. A computer-readable storage medium refers to any type of physical memory in which information or data readable by a processor can be stored. Thus, a computer-readable storage medium can store instructions for execution by one or more processors, including instructions for causing the processor to perform steps or stages associated with the embodiments described herein. The term "computer-readable medium" should be understood to include tangible objects and exclude carrier waves and transitory signals, i.e., non-transitory objects. Examples include random access memory (RAM), read-only memory (ROM), volatile memory, non-volatile memory, hard drives, compact disc (CD) ROMs, digital video discs (DVDs), flash drives, disks, and other known physical storage media.

[0084] The terms "embodiment," "one or more embodiments," "some embodiments," and "one embodiment" mean "one or more (but not all) embodiments of the invention," unless otherwise specified.

[0085] "Including," "comprising," "having," and variations thereof, unless otherwise expressly stated, mean "including but not limited to." The listing of listed items does not imply that any or all items are mutually exclusive unless otherwise expressly stated.

[0086] The terms "a," "an," and "the" mean "one or more" unless otherwise specified. A description of an embodiment having several components in communication with each other does not imply that all such components are required. On the contrary, a variety of optional components are described to illustrate the wide variety of possible embodiments of the present invention.

[0087] Where a single device or article is described herein, it is clear that multiple devices / articles (whether or not they cooperate) can be used in place of the single device / article. Similarly, where multiple devices or articles are described herein (whether or not they cooperate), it is clear that the single device / article can be used in place of the multiple devices or articles, or that a different number of devices / articles can be used in place of the number of devices or programs shown. The functionality and / or features of a device can alternatively be embodied by one or more other devices not explicitly described as having such functionality / features. Thus, other embodiments of the present invention need not include the device itself.

[0088] Finally, the language used in the specification has been chosen primarily for purposes of readability and instruction, and not to define or limit the scope of the inventive subject matter. Accordingly, it is intended that the scope of the invention be limited not by this detailed description, but rather by the claims that follow. Accordingly, the embodiments of the present invention illustrate, but do not limit, the scope of the invention, which is set forth in the following claims.

[0089] While various aspects and embodiments are disclosed herein, other aspects and embodiments will be apparent to those skilled in the art. The various aspects and embodiments disclosed herein are for purposes of illustration and not limitation, with the true scope and spirit being indicated by the following claims. [Explanation of symbols]

[0090] 100 Configuration Examples 101 Image Segmentation System 103 Reference Image 105 Reference Database 107 Current Images 109 Target Regions of Interest (RoI) 201 I / O Interface 203 processor 205 memory 207 Data 209 Modules 211 Segmentation Parameters 213 Reference point 215 Moved Points 219 Other Data 221 Parameter Determination Module 223 Reference Point Determination Module 225 Seed Generation Module 227 Seed Selection Module 229 Segmentation Module 231 other modules 301 Example Reference Image 303 Segmented Reference RoI 303A Longest diameter 303B Short shaft 311 Related Seeds 313 Duplicate area 501 I / O Interface of Exemplary Computer System 502 Processor of Exemplary Computer System 503 Network Interface 504 Storage Interface 505 Memory of an Exemplary Computer System 506 users / applications 507 Operating Systems 508 Web Browser 509 Communication Network 511 Input Device 512 Output Device 513 RAM 514 ROM

Claims

1. 1. A method for automatically propagating a segmentation in a medical image, the method comprising: determining one or more segmentation parameters based on analysis of a segmented reference region of interest in the reference image, wherein determining the one or more segmentation parameters comprises determining one or more morphological characteristics of the segmented reference region of interest, and determining the one or more morphological characteristics of the segmented reference region of interest comprises determining a minor axis and a longest diameter of the segmented reference region of interest; determining a plurality of reference points on the segmented reference region of interest based on the one or more morphological characteristics of the segmented reference region of interest, the plurality of reference points being positioned along the minor axis and the longest diameter of the segmented reference region of interest; generating a plurality of displaced points on a current image from which a target region of interest has to be segmented by displacing each of the plurality of reference points onto the current image; automatically selecting an associated seed from the plurality of shifted points on the current image based on the one or more segmentation parameters, the selecting including, for each of the plurality of shifted points on the current image, selecting the point as the associated seed if a pixel intensity at the location of the point falls within an intensity distribution of the segmented reference region of interest of the reference image; performing a multi-seed segmentation of the selected associated seeds to estimate and segment a target region of interest in the current image, the target region of interest being a propagated segmentation of the segmented reference region of interest in the reference image, and pixel intensities of the current image being quantitatively equivalent to or normalized to pixel intensities of the reference image; A method having the following.

2. The method of claim 1 , wherein the reference image and the current image are images of an object acquired at different time periods.

3. 2. The method of claim 1, wherein the step of determining the one or more segmentation parameters further comprises determining at least one of a position of the segmented reference region of interest in the reference image, an intensity distribution of the segmented reference region of interest, and a background intensity distribution of a background of the segmented reference region of interest.

4. Determining the one or more morphological characteristics of the segmented reference region of interest includes: determining a size and a shape of the segmented reference region of interest based on the lengths and coordinates of the minor axis and the longest diameter of the segmented reference region of interest; The method of claim 3 further comprising:

5. Determining the background intensity distribution of the background includes: grouping the background pixels into a plurality of background quadrants having an origin at a center of the segmented reference region of interest; determining a plurality of intensity distributions of a plurality of background pixels / points corresponding to the quadrants relative to a center of the segmented reference region of interest based on an amount of overlap between the pixel intensities of the background of each of the plurality of background quadrants and pixel intensities of the segmented reference region of interest; The method of claim 3, comprising:

6. Determining the background intensity distribution of the background comprises: determining a background threshold value corresponding to each of the plurality of background quadrants based on an amount of overlap between pixel intensities of the respective background quadrant and pixel intensities of the segmented reference region of interest; determining a background intensity distribution corresponding to each of the plurality of background quadrants based on the corresponding background threshold; The method of claim 5 , comprising:

7. Determining an intensity distribution of the segmented reference region of interest includes: generating a histogram of pixels within the segmented reference region of interest; determining one or more intensity ranges of the segmented reference region of interest based on the histogram and a predetermined threshold, the one or more intensity ranges representing an intensity distribution of the segmented reference region of interest; The method of claim 3, comprising:

8. 2. The method of claim 1, wherein the step of automatically selecting associated seeds in the current image comprises selecting one or more of the moved points if their intensities do not fall within a background intensity distribution of the segmented reference region of interest.

9. 9. The method of claim 8, wherein the step of automatically selecting the associated seeds in the current image comprises selecting one or more of the moved points if the intensities of at least a predetermined number of neighboring pixels of the one or more moved points fall within the intensity distribution of the segmented reference region of interest.

10. 10. The method of claim 9, wherein the step of automatically selecting the associated seeds in the current image comprises selecting one or more of the moved points if the intensities of at least a predetermined number of neighboring pixels of the one or more moved points do not fall within a background intensity distribution range of a background of the segmented reference region of interest.

11. 2. The method of claim 1, wherein the step of automatically selecting the associated seed in the current image comprises determining whether an intensity of one or more of the moved points is less than an intensity distribution range of the segmented reference region of interest and a background intensity distribution range of a background of the segmented reference region of interest.

12. 2. The method of claim 1, wherein automatically selecting the associated seed within the current image comprises determining whether the intensity of one or more of the moved points is greater than a maximum intensity value of an intensity distribution of the segmented reference region of interest and greater than a maximum intensity value of a background intensity distribution range of a background of the segmented reference region of interest.

13. 1. An image segmentation system for automatically propagating a segmentation in a medical image, the image segmentation system comprising: a processor; a memory communicatively coupled to the processor, the memory storing processor-executable instructions that, when executed, cause the processor to: a process for determining one or more segmentation parameters based on an analysis of a segmented reference region of interest in a reference image, wherein the process for determining the one or more segmentation parameters comprises determining one or more morphological characteristics of the segmented reference region of interest, and determining the one or more morphological characteristics of the segmented reference region of interest comprises determining a minor axis and a longest diameter of the segmented reference region of interest; determining a plurality of reference points on the segmented reference region of interest based on one or more morphological characteristics of the segmented reference region of interest, the plurality of reference points being positioned along the minor axis and the longest diameter of the segmented reference region of interest; generating a plurality of displaced points on a current image from which a target region of interest has to be segmented by displacing each of the plurality of reference points onto the current image; automatically selecting an associated seed from the plurality of shifted points on the current image based on the one or more segmentation parameters, the selecting including, for each of the plurality of shifted points on the current image, selecting the point as the associated seed if a pixel intensity at the location of the point falls within an intensity distribution of the segmented reference region of interest of the reference image; performing multi-seed segmentation of the selected associated seeds to estimate and segment the target region of interest in the current image, the target region of interest being a propagated segmentation of the segmented reference region of interest in the reference image, and pixel intensities of the current image being quantitatively equivalent to or normalized to pixel intensities of the reference image; a memory for executing the An image segmentation system having:

14. The image segmentation system of claim 13 , wherein the reference image and the current image are images of an object acquired at different time periods.

15. A computer program that causes a computer to execute each step of the method according to any one of claims 1 to 12.

Images