Region of interest positioning method, computer device and storage medium

Abstract

The application relates to a positioning method of a region of interest, a computer device and a storage medium. The method comprises the following steps: determining the position of a region of interest and the position of each tissue in a medical image; determining the measured distance between the region of interest and each tissue according to the position of the region of interest and the position of each tissue; positioning a target tissue corresponding to the region of interest from each tissue according to each measured distance and a reference distance in a preset template library. The preset template library comprises a group of reference distances corresponding to each tissue, and the group of reference distances comprises the reference distances between each tissue and itself and other tissues. The method can ensure the accuracy of positioning a lesion region.

Description

Technical Field

[0001] This application relates to the field of image processing technology, and in particular to a method for locating a region of interest, a computer device, and a storage medium. Background Technology

[0002] With the continuous development of neural network technology, many problems in the field of image processing are now mostly solved using neural networks in the hope of improving the efficiency and accuracy of image processing. When using neural networks for image processing, locating the region of interest in an image is a common problem.

[0003] In related technologies, taking the localization of lesions on the ribs as an example, the general approach is to first segment the ribs and lesion areas in the image using a neural network, and then locate the specific position of the lesion area on the ribs using the segmentation results.

[0004] However, the above-mentioned technologies have the problem of not being able to guarantee the accuracy of locating the lesion area. Summary of the Invention

[0005] Therefore, it is necessary to provide a method, computer device, and storage medium for locating the region of interest that can ensure the accuracy of locating the lesion area, in response to the above-mentioned technical problems.

[0006] Firstly, this application provides a method for locating a region of interest, the method comprising:

[0007] The location of the region of interest and the location of each tissue are determined based on the acquired medical images;

[0008] The measurement distance between the region of interest and each tissue is determined based on the location of the region of interest and the location of each tissue.

[0009] Based on the measured distances and the reference distances in the preset template library, the target organization corresponding to the region of interest is located from each organization; the preset template library includes a set of reference distances corresponding to each organization, and the set of reference distances includes the reference distances between each organization and itself and other organizations.

[0010] In one embodiment, the process of locating the target tissue corresponding to the region of interest from each tissue based on each measured distance and a reference distance in a preset template library includes:

[0011] Each measured distance is matched with a set of reference distances corresponding to each tissue to determine the set of reference distances that are successfully matched.

[0012] The organizations corresponding to a set of successfully matched reference distances are identified as the target organizations.

[0013] In one embodiment, the above-mentioned matching of each measured distance with a set of reference distances corresponding to each tissue to determine a set of successfully matched reference distances includes:

[0014] Calculate the difference between each measured distance and a set of reference distances corresponding to each tissue to obtain the difference value for each tissue;

[0015] Determine the minimum difference value among all the difference values, and determine the set of reference distances corresponding to the minimum difference value as the set of reference distances for successful matching.

[0016] In one embodiment, before locating the target tissue corresponding to the region of interest from each tissue based on each measured distance and a reference distance in a preset template library, the method further includes:

[0017] At least one measurement distance is selected from all the measurement distances as the comparison distance; wherein the number of comparison distances is less than the number of measurement distances.

[0018] Accordingly, based on the measured distances and reference distances in the preset template library, the target tissue corresponding to the region of interest is located from each tissue, including:

[0019] Based on the comparison distances and the reference distances in the preset template library, the target organization corresponding to the region of interest is located from each organization.

[0020] In one embodiment, the process of locating the target tissue corresponding to the region of interest from each tissue based on each comparison distance and a reference distance in a preset template library includes:

[0021] Normalize each alignment distance to determine the normalized alignment distance;

[0022] Based on the normalized comparison distances and the reference distances in the preset template library, the target organization corresponding to the region of interest is located from each organization.

[0023] In one embodiment, the above-mentioned normalization processing of each alignment distance to determine each normalized alignment distance includes:

[0024] The location of each vertebra was determined based on the acquired medical images;

[0025] Calculate the distance between each pair of adjacent vertebrae based on the position of each vertebra, and determine the average distance based on the calculated distances;

[0026] Divide each alignment distance by the average distance to determine the normalized alignment distance.

[0027] In one embodiment, the process of locating the target tissue corresponding to the region of interest from each tissue based on the normalized comparison distances and the reference distances in a preset template library includes:

[0028] In the preset template library, select the reference distances corresponding to the normalized comparison distance positions from a set of reference distances for each tissue;

[0029] Based on the normalized comparison distances and the reference distances selected from each organization, the target organization corresponding to the region of interest is located from each organization.

[0030] In one embodiment, the aforementioned tissues include various ribs, and the aforementioned region of interest includes lesion areas on the various ribs.

[0031] In one embodiment, each tissue includes a left-side tissue and a right-side tissue. The determination of the measurement distance between the region of interest and each tissue, based on the location of the region of interest and the location of each tissue, includes:

[0032] Based on the location of the region of interest and the location of each organization, determine the organization to be calculated corresponding to the region of interest within each organization; the organization to be calculated is either the left-side organization or the right-side organization.

[0033] The measurement distance between the region of interest and each tissue to be calculated is determined based on the location of the region of interest and the location of the tissues to be calculated.

[0034] In one embodiment, determining the measurement distance between the region of interest and each tissue to be calculated based on the location of the region of interest and the location of the tissue to be calculated includes:

[0035] Determine the centerline corresponding to each organization to be calculated based on its location.

[0036] Based on the location of the region of interest, the distance between the region of interest and each centerline is calculated, thus obtaining the measured distance between the region of interest and each tissue to be calculated.

[0037] In one embodiment, the method for establishing the aforementioned preset template library includes:

[0038] Acquire multiple sample images; the sample images include tissues with complete structures;

[0039] Calculate a set of sample distances corresponding to each tissue in each sample image to obtain multiple sets of sample distances for each sample image;

[0040] In the multiple sets of sample distances corresponding to each sample image, the average of the multiple sets of sample distances for the same tissue is processed to obtain a set of reference distances for each tissue.

[0041] Establish a correspondence between each organization and a corresponding set of reference distances to obtain a preset template library.

[0042] In one embodiment, the aforementioned structurally intact tissues include structurally intact thoracic vertebrae and structurally intact ribs.

[0043] In one embodiment, after calculating a set of sample distances corresponding to each tissue in each sample image to obtain multiple sets of sample distances corresponding to each sample image, the method further includes:

[0044] Calculate the distance between every two adjacent vertebrae in each sample image, and determine the average distance corresponding to each sample image based on the calculated multiple distances;

[0045] Divide the distances of multiple sets of samples corresponding to each sample image by their respective average distances to determine the normalized distances of multiple sets of samples corresponding to each sample image.

[0046] Secondly, this application also provides a positioning device for a region of interest, the device comprising:

[0047] The location determination module is used to determine the location of the region of interest and the location of each tissue in the acquired medical image.

[0048] The distance determination module is used to determine the distance between the region of interest and each tissue based on the location of the region of interest and the location of each tissue.

[0049] The positioning module is used to locate the target organization corresponding to the region of interest from each organization based on the measured distances and the reference distances in the preset template library. The preset template library includes a set of reference distances corresponding to each organization, which includes the reference distances between each organization and itself and other organizations.

[0050] Thirdly, this application also provides a computer device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to perform the following steps:

[0051] The location of the region of interest and the location of each tissue are determined based on the acquired medical images;

[0052] The measurement distance between the region of interest and each tissue is determined based on the location of the region of interest and the location of each tissue.

[0053] Based on the measured distances and the reference distances in the preset template library, the target organization corresponding to the region of interest is located from each organization; the preset template library includes a set of reference distances corresponding to each organization, and the set of reference distances includes the reference distances between each organization and itself and other organizations.

[0054] Fourthly, this application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, performs the following steps:

[0055] The location of the region of interest and the location of each tissue are determined based on the acquired medical images;

[0056] The measurement distance between the region of interest and each tissue is determined based on the location of the region of interest and the location of each tissue.

[0057] Based on the measured distances and the reference distances in the preset template library, the target organization corresponding to the region of interest is located from each organization; the preset template library includes a set of reference distances corresponding to each organization, and the set of reference distances includes the reference distances between each organization and itself and other organizations.

[0058] Fifthly, this application also provides a computer program product, comprising a computer program that, when executed by a processor, performs the following steps:

[0059] The location of the region of interest and the location of each tissue are determined based on the acquired medical images;

[0060] The measurement distance between the region of interest and each tissue is determined based on the location of the region of interest and the location of each tissue.

[0061] Based on the measured distances and the reference distances in the preset template library, the target organization corresponding to the region of interest is located from each organization; the preset template library includes a set of reference distances corresponding to each organization, and the set of reference distances includes the reference distances between each organization and itself and other organizations.

[0062] The aforementioned method, computer equipment, and storage medium for locating regions of interest (ROIs) determine the location of the ROI and other tissues within acquired medical images. Based on the locations, it calculates the measured distances between the ROI and each tissue, and then locates the target tissue corresponding to the ROI from among the tissues using these measured distances and reference distances from a pre-set template library. The template library includes a set of reference distances for each tissue, including reference distances between each tissue and itself, as well as distances between other tissues. This method improves the accuracy of ROI location by comparing the distances between the ROI and other tissues with the reference distances in the template library, thus avoiding excessive reliance on ROI segmentation results. Furthermore, the location process incorporates reference distances from multiple tissues, preventing over-reliance on any single tissue and allowing for more data to be used, further enhancing the accuracy of ROI location. Attached Figure Description

[0063] Figure 1 This is an internal structural diagram of a computer device in one embodiment;

[0064] Figure 2 This is a flowchart illustrating a method for locating a region of interest in one embodiment;

[0065] Figure 3 This is a flowchart illustrating the location of the region of interest in another embodiment;

[0066] Figure 4 This is a flowchart illustrating a method for locating a region of interest in another embodiment;

[0067] Figure 5 This is a flowchart illustrating a method for locating a region of interest in another embodiment;

[0068] Figure 6 This is a flowchart illustrating a method for locating a region of interest in another embodiment;

[0069] Figure 7 This is a structural block diagram of a region of interest positioning device in one embodiment. Detailed Implementation

[0070] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0071] The region of interest localization method provided in this application can be applied to computer devices, which can be terminals or servers. Taking a terminal as an example, its internal structure diagram can be as follows: Figure 1 As shown, the computer device includes a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, NFC (Near Field Communication), or other technologies. When executed by the processor, the computer program implements a method for locating a region of interest. The display screen can be an LCD screen or an e-ink display. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad mounted on the computer device casing, or an external keyboard, touchpad, or mouse.

[0072] Those skilled in the art will understand that Figure 1 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0073] In one embodiment, such as Figure 2 As shown, a method for locating a region of interest is provided, which can be applied to... Figure 1 Taking a computer device as an example, the following steps may be included:

[0074] S102, determine the location of the region of interest and the location of each tissue in the acquired medical image.

[0075] The medical images can be acquired in real-time from the subject under test, or they can be obtained from the cloud or a server. These medical images can be three-dimensional or two-dimensional. The medical images include various tissues and regions of interest (ROIs), where each tissue can also be referred to as a structure. As an optional embodiment, the aforementioned tissues include each rib, and the ROI includes lesion areas on each rib. Typically, normal tissue includes 12 left ribs and 12 right ribs. The subject under test may have 24 ribs, or fewer. The lesion areas can be any rib on any one of the subject's ribs, or lesion areas on multiple ribs; there may be one or more lesion areas.

[0076] Specifically, after obtaining the medical image of the object to be tested, image segmentation methods or segmentation models can be used to segment each tissue and the region of interest in the medical image, obtaining the location of each tissue and the location of the region of interest. This can be done by segmenting each tissue and the region of interest simultaneously, or by segmenting each tissue first and then further segmenting the region of interest based on that segmentation. Other segmentation methods are also possible, and no specific limitations are made here.

[0077] S104. Determine the measurement distance between the region of interest and each tissue based on the location of the region of interest and the location of each tissue.

[0078] In this step, after obtaining the location of each organization and the location of the region of interest, the distance calculation formula can be used to calculate the distance between the location of each organization and the location of the region of interest, and obtain the distance between each organization and the region of interest, which is recorded as the measurement distance. Each organization will obtain a measurement distance, so multiple measurement distances can be obtained.

[0079] In addition, regarding the method of calculating the distance between the location of each organization and the location of the region of interest, taking an organization as an example, the organization can be discretized into multiple points. The location of the region of interest can be represented by the location of a selected point (e.g., the center point). Then, the distance between the points of the region of interest and each point on the organization can be calculated. Then, one of the multiple distances obtained can be determined as the measured distance between the organization and the region of interest. The measured distance of other organizations can also be calculated by referring to this calculation method, which will not be elaborated here.

[0080] S106, Based on the measured distances and the reference distances in the preset template library, locate the target tissue corresponding to the region of interest from each tissue.

[0081] The aforementioned preset template library includes a set of reference distances for each tissue. These reference distances include the distances between each tissue and itself, as well as the distances between other tissues. The tissues included in the preset template library are generally complete tissues. The reference distances refer to the distances calculated using these complete tissues, and can be the shortest distances between tissues. These reference distances can be obtained by statistically analyzing large amounts of data. Taking a rib as an example, the template library includes 24 ribs and a set of reference distances for each rib. For a single rib, the set of reference distances includes the shortest distance between the rib and itself (generally 0), and the shortest distance between the rib and the other 23 ribs, for a total of 24 distances. The distances for other ribs can be calculated in the same way, thus obtaining a set of reference distances for each tissue.

[0082] Specifically, after obtaining multiple measurement distances between the region of interest of the object under test and various tissues, these multiple measurement distances and multiple sets of reference distances from the template library are processed to determine the tissues corresponding to these multiple measurement distances, that is, to obtain the target tissues where the region of interest is located.

[0083] In the aforementioned method for locating regions of interest (ROIs), the location of the ROI and the positions of various tissues are determined from the acquired medical images. The measured distances between the ROI and each tissue are calculated based on their positions. Then, the target tissue corresponding to the ROI is located from each tissue based on these measured distances and reference distances in a pre-set template library. The template library includes a set of reference distances for each tissue, including reference distances between each tissue and itself, as well as distances between other tissues. This method locates the target tissue by comparing the distances between the ROI and each tissue with the reference distances in the template library. This eliminates the need to overly rely on the segmentation results of the ROI, thus improving the accuracy of ROI location. Furthermore, the location process incorporates reference distances from multiple tissues, preventing over-reliance on any single tissue. The larger amount of data involved in the location process further enhances the accuracy of ROI location.

[0084] The above embodiments mentioned that the region of interest can be located by using multiple sets of reference distances in the template library. The following embodiments will describe the process in detail.

[0085] In another embodiment, such as Figure 3 As shown, another method for locating a region of interest is provided. Based on the above embodiments, S106 may include the following steps:

[0086] S202, each measured distance is matched with a set of reference distances corresponding to each tissue to determine a set of successfully matched reference distances.

[0087] In this step, when matching each measured distance with a set of reference distances corresponding to each tissue, as an optional embodiment, it may include: calculating the difference between each measured distance and a set of reference distances corresponding to each tissue to obtain the difference value corresponding to each tissue; determining the minimum difference value among the difference values, and determining the set of reference distances corresponding to the minimum difference value as the successfully matched set of reference distances.

[0088] When calculating the difference between each measured distance and a set of reference distances corresponding to each tissue, one can calculate the Euclidean distance between each measured distance and each set of reference distances, and use the calculated Euclidean distance as the difference value between each set of reference distances and each measured distance, thus obtaining multiple difference values. Then, these difference values ​​can be sorted, either from largest to smallest or smallest to largest; the minimum difference value can be obtained from the sorting results. Finally, the set of reference distances corresponding to the minimum difference value can be obtained, thus obtaining the successfully matched set of reference distances.

[0089] S204, the organization corresponding to the successfully matched set of reference distances is identified as the target organization.

[0090] In this step, after obtaining a set of reference distances that have been successfully matched, the corresponding organization can also be obtained, and the obtained organization is the target organization.

[0091] In this embodiment, each measured distance is matched with a set of reference distances corresponding to each tissue in the template library, and the tissue corresponding to the successfully matched set of reference distances is determined as the target tissue, thus locating the target tissue containing the region of interest. This reference distance matching speeds up the localization process, thereby improving the efficiency of region of interest localization. Furthermore, by calculating the difference value and determining the minimum difference value to determine the successfully matched set of reference distances, the efficiency and accuracy of localization can be further improved.

[0092] The above embodiments mention the process of determining the target organization by matching multiple measured distances with multiple sets of reference distances in the template library. Before that, in order to save computation, a portion of the measured distances can be selected to participate in the calculation. The following embodiments will describe this process in detail.

[0093] In another embodiment, such as Figure 4 As shown, another method for locating the region of interest is provided. Based on the above embodiment, before step S106, the method may further include the following steps:

[0094] S302, select at least one measurement distance from each measurement distance as a comparison distance; wherein the number of comparison distances is less than the number of measurement distances.

[0095] In this step, after calculating multiple measurement distances, these distances can be sorted to obtain a sorting result. From this sorting result, the N smallest measurement distances are selected as comparison distances, where N is less than the number of calculated measurement distances. Alternatively, a distance threshold can be pre-set, and then the N smallest distances smaller than the threshold can be directly selected as comparison distances. The distance threshold can be set according to the actual situation.

[0096] For example, suppose there are 6 calculated measurement distances: 15, 32, 64, 49, 21, and 8, in mm. The result of sorting from largest to smallest is: 64, 49, 32, 21, 15, and 8. Selecting the 4 smaller values, we get 32, 21, 15, and 8. These four selected values ​​are the comparison distances.

[0097] After selecting the comparison distance as described above, it can be matched with the reference distance in the template library to locate the region of interest. Accordingly, S106 above may include:

[0098] S304. Based on the comparison distances and the reference distances in the preset template library, locate the target organization corresponding to the region of interest from each organization.

[0099] In this step, the selected alignment distance is usually greater than one, i.e., multiple alignment distances. After selecting the alignment distances, reference distances corresponding to each alignment distance position can be selected from a set of reference distances corresponding to each tissue. Then, the selected alignment distances are directly matched with the reference distances corresponding to each position in each set of reference distances to obtain the matched reference distances. Alternatively, the selected alignment distances can be normalized before matching. As an optional embodiment, this process may include the following steps:

[0100] A1. Normalize each alignment distance to determine the normalized alignment distance.

[0101] Taking the ribs in the aforementioned medical image as examples, the medical image also includes the vertebrae of the test object, such as the thoracic vertebrae. When performing normalization, the position of each vertebra (e.g., the thoracic vertebrae) can be determined based on the acquired medical image. The distance between each two adjacent vertebrae (e.g., the thoracic vertebrae) can be calculated based on the position of each vertebra (e.g., the thoracic vertebrae). The average distance can be determined based on the calculated multiple distances. Each comparison distance is divided by the average distance to determine the normalized comparison distance.

[0102] Specifically, this can be achieved by segmenting the vertebrae in a medical image to obtain the position of each vertebra, and then calculating the center point of each vertebra based on its position. Next, the distance between each pair of adjacent vertebrae is calculated using a distance calculation formula and the center point of each vertebra, resulting in multiple distances. These multiple distances are then summed and averaged to obtain the average distance. Finally, each selected comparison distance is divided by this average distance, and the resulting value is the normalized value for each comparison distance.

[0103] A2, based on the normalized comparison distances and the reference distances in the preset template library, locates the target organization corresponding to the region of interest from each organization.

[0104] After obtaining the normalized alignment distances, to further reduce computational load and improve accuracy, as an optional implementation, reference distances corresponding to the positions of the normalized alignment distances can be selected from a set of reference distances corresponding to each tissue in a preset template library. Based on the normalized alignment distances and the reference distances selected from each tissue, the target tissue corresponding to the region of interest can be located from each tissue. Here, matching and locating with the reference distances at corresponding positions in the template library reduces computational load, improves matching efficiency, and enhances matching accuracy.

[0105] For example, taking ribs as an example, assume that the calculated measurement distances include 12 measurement distances between the region of interest and the left T1-T12 ribs of the object under test. The smaller comparison distances selected by the above distance threshold are for the T3-T9 ribs. The comparison distances corresponding to the T3-T9 ribs are normalized. The template library includes 12 sets of reference distances corresponding to the left T1-T12 ribs and 12 sets of reference distances corresponding to the right T1-T12 ribs. Then, during localization, the 12 sets of reference distances corresponding to the left T1-T12 ribs can be selected from the template library first. Then, based on the T3-T9 ribs corresponding to each comparison distance, the reference distances corresponding to the T3-T9 ribs are selected from each of the 12 sets of reference distances on the left. In this way, the 12 sets of reference distances on the left all include the reference distances corresponding to the T3-T9 ribs. Next, the difference between the T3-T9 comparison distance and the reference distance at the T3-T9 rib position in the 12 sets of reference distances can be calculated to obtain the difference value corresponding to each set of reference distances. The 12 sets of difference values ​​are obtained, and the minimum difference value is obtained. The tissue corresponding to the minimum difference value is assumed to be T5. Then, T5 is the target tissue where the region of interest is located.

[0106] In this embodiment, by selecting a relatively small number of measurement distances that are less than a measurement distance threshold from multiple measurement distances as the comparison distances, the amount of data required for subsequent matching with reference distances in the template library can be reduced, thereby improving matching efficiency and ultimately improving the localization efficiency of the region of interest. Furthermore, selecting smaller measurement distances for subsequent matching can filter out some noise and is more consistent with actual conditions, thus improving the accuracy of the matching results and localization. Further, matching can be performed after normalizing each measurement distance, which can further improve the accuracy of the matching results.

[0107] The above embodiments mention that the distance between the region of interest and each tissue can be calculated. The tissues mentioned above can include left tissues and right tissues. In order to further save computation, it is possible to first distinguish whether the region of interest belongs to the left tissue or the right tissue before performing the calculation. The following embodiments will describe this process in detail.

[0108] In another embodiment, such as Figure 5 As shown, another method for locating a region of interest is provided. Based on the above embodiments, S104 may include the following steps:

[0109] S402, based on the location of the region of interest and the location of each organization, determine the organization to be calculated corresponding to the region of interest in each organization; the organization to be calculated is either the left organization or the right organization.

[0110] In this step, taking the ribs in the aforementioned medical image as examples, the medical image also includes the vertebrae of the subject. This can be achieved by segmenting the vertebrae in the medical image to obtain the position of each vertebra, and calculating the position of the center point of each vertebra based on its position. The position of the center point of the region of interest can also be determined by the position of the region of interest. Then, the uppermost vertebra (i.e., closest to the head) can be selected from the center points of the vertebrae, and the position of the center point of the selected vertebra can be obtained. Finally, the position of the region of interest (ROI) or the position of the center point of the selected vertebra can be used to determine whether the ROI is located to the left or right of the selected vertebra.

[0111] Assuming the region of interest is located to the left of the selected vertebra, the ribs to the left of the vertebra are considered the tissue to be calculated, denoted as the left-side tissue, and their location is obtained. If the region of interest is located to the right of the selected vertebra, the ribs to the right of the vertebra are considered the tissue to be calculated, denoted as the right-side tissue, and their location is obtained. Furthermore, the left-side and right-side tissues can be symmetrical or asymmetrical.

[0112] S404, determine the measurement distance between the region of interest and each tissue to be calculated based on the location of the region of interest and the location of the tissues to be calculated.

[0113] In this step, after selecting the organization to be calculated from the various organizations, the measurement distance between the region of interest and each organization to be calculated can be directly calculated. Specifically, when calculating the measurement distance, as an optional embodiment, the centerline corresponding to each organization to be calculated can be determined based on the position of each organization to be calculated; based on the position of the region of interest, the distance between the region of interest and each centerline can be calculated to obtain the measurement distance between the region of interest and each organization to be calculated.

[0114] Specifically, centerlines can be extracted from the locations of the tissues to be calculated, yielding centerlines for each tissue. These centerlines can be composed of discrete points. Then, the distances between the region of interest (ROI) and each point on each centerline can be calculated. Multiple distances can be obtained for each centerline, and the smallest distance is selected as the distance between that centerline and the ROI. This process yields the distance for each centerline, representing the measured distance between each tissue and the ROI.

[0115] In this embodiment, the location of the region of interest determines whether it belongs to the left or right side of the tissue, and distance calculations are performed on the corresponding side of the tissue. This reduces the computational workload by half, significantly improving distance calculation efficiency and thus localization efficiency. Furthermore, this method can be used for both initial and subsequent fine-tuning, effectively improving localization accuracy. Further, distance calculations are performed along the centerline of the region of interest and the tissue, ensuring the effectiveness and accuracy of the distance calculations.

[0116] The above embodiments mentioned that the measured distance can be matched with the reference distance in the template library, and also briefly mentioned the process of establishing the template library. The following embodiments will describe the process of establishing the template library in detail.

[0117] In another embodiment, such as Figure 6 As shown, another method for locating the region of interest is provided. Based on the above embodiments, the establishment of the aforementioned preset template library may include the following steps:

[0118] S502, acquire multiple sample images; the sample images include tissues with complete structures.

[0119] As an optional embodiment, taking the rib as an example, each structurally intact tissue in each sample image may include a structurally intact thoracic vertebra, a structurally intact rib, or other structurally intact tissues. Specifically, medical images of multiple objects may be obtained, and the medical images with structurally intact tissues may be used as sample images.

[0120] After obtaining the sample images, the positions of the vertebrae and ribs in each sample image can be obtained through segmentation algorithms or segmentation models. The center point of each vertebra can be calculated based on its position, and the center line of each rib can be extracted based on its position to obtain the center line of each rib.

[0121] S504, calculate a set of sample distances corresponding to each tissue in each sample image, and obtain multiple sets of sample distances corresponding to each sample image.

[0122] In this step, using a sample image as an example, the distances from one rib to the other ribs can be calculated using the center lines of each rib. Let's take calculating a set of sample distances corresponding to the first rib as an example: The shortest distances from the center line of the first rib to the center lines of all other ribs can be calculated and arranged sequentially to form the 24-dimensional feature vector of the first rib, thus obtaining a set of reference distances corresponding to the first rib. The shortest distance from the first rib to itself is 0.

[0123] Similarly, the sample distances corresponding to each rib in a sample image can be calculated in this way, and arranged sequentially in rows to obtain multiple sets of sample distances for a single sample image, which can be 24*24 feature vectors. Likewise, multiple sets of sample distances corresponding to each of all sample images can be calculated in this way, with each sample image having a corresponding 24*24 feature vector.

[0124] Furthermore, each of the above sample images includes vertebrae, and the distance between every two adjacent vertebrae in each sample image can be calculated, and the average distance corresponding to each sample image can be determined based on the calculated multiple distances; the multiple sets of sample distances corresponding to each sample image are divided by their respective average distances to determine the normalized multiple sets of sample distances corresponding to each sample image.

[0125] Specifically, the above process calculates the center point of each vertebra in each sample image. Taking a single sample image as an example, the distance between the center points of every two adjacent vertebrae in the image is calculated. Multiple distances are summed and averaged to obtain the average distance. Then, each value in the corresponding 24*24 feature vector of the sample image is divided by this average distance to obtain the normalized 24*24 feature vector. This achieves the normalization of multiple sets of sample distances in the sample image. Other sample images can also be normalized in this way to obtain corresponding normalized results. Normalizing the sample distances here ensures they correspond to the measured distances, thus making the matching results obtained during subsequent distance matching more accurate.

[0126] S506, In the multiple sets of sample distances corresponding to each sample image, the average of the multiple sets of sample distances for the same tissue is processed to obtain a set of reference distances corresponding to each tissue.

[0127] In this step, after obtaining multiple sets of sample distances after normalization of each sample image, the sample distances corresponding to the same tissue in each sample image can be summed and averaged. For example, the sample distances corresponding to the T2 rib in each sample can be summed and averaged to obtain a set of average distances corresponding to each tissue, that is, a set of reference distances corresponding to each tissue.

[0128] S508 establishes a correspondence between each organization and a corresponding set of reference distances to obtain a preset template library.

[0129] In this step, a set of reference distances corresponding to each tissue can be bound to the corresponding tissue. For example, a set of reference distances corresponding to the T2 rib can be bound to the T2 rib. Other tissues can also be bound in the same way, thus obtaining a preset template library.

[0130] In this embodiment, by calculating multiple sets of sample distances corresponding to complete sample images of each tissue structure, and taking the average of multiple sets of sample distances of the same tissue to obtain a set of reference distances for each tissue, a template library is established. By calculating the reference distances through complete structures and taking the average, the calculated reference distances of the tissues can cover the distances of all tissues and are relatively accurate, thereby improving the accuracy of the final template library.

[0131] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0132] Based on the same inventive concept, this application also provides a region-of-interest (ROI) localization device for implementing the above-described method for locating the RIO. The solution provided by this device is similar to the implementation described in the above-described method; therefore, the specific limitations in the one or more RIO localization device embodiments provided below can be found in the limitations of the RIO localization method described above, and will not be repeated here.

[0133] In one embodiment, such as Figure 7 As shown, a positioning device for a region of interest is provided, comprising: a position determination module 11, a distance measurement determination module 12, and a positioning module 13, wherein:

[0134] The location determination module 11 is used to determine the location of the region of interest and the location of each tissue in the acquired medical image.

[0135] The distance determination module 12 is used to determine the distance between the region of interest and each tissue based on the location of the region of interest and the location of each tissue.

[0136] The positioning module 13 is used to locate the target organization corresponding to the region of interest from each organization based on each measured distance and the reference distance in the preset template library; the preset template library includes a set of reference distances corresponding to each organization, and the set of reference distances includes the reference distances between each organization and itself and other organizations.

[0137] In another embodiment, a different location device for the region of interest is provided. Based on the above embodiments, the location module 13 may include:

[0138] The matching unit is used to match each measured distance with a set of reference distances corresponding to each tissue, and to determine a set of reference distances that are successfully matched.

[0139] The determination unit is used to identify the organization corresponding to a set of successfully matched reference distances as the target organization.

[0140] Optionally, the matching unit described above is specifically used to calculate the difference between each measured distance and a set of reference distances corresponding to each tissue, to obtain the difference value corresponding to each tissue; to determine the minimum difference value among the difference values, and to determine the set of reference distances corresponding to the minimum difference value as a successfully matched set of reference distances.

[0141] In another embodiment, a different region of interest localization device is provided. Based on the above embodiments, before the localization module 13 locates the target tissue corresponding to the region of interest from each tissue according to each measured distance and a preset template library, the device may further include:

[0142] A selection module is used to select at least one measurement distance from each measurement distance as a comparison distance; wherein the number of comparison distances is less than the number of measurement distances.

[0143] Accordingly, the aforementioned positioning module 13 is specifically used to locate the target organization corresponding to the region of interest from each organization based on each comparison distance and the reference distance in the preset template library.

[0144] Optionally, the positioning module 13 may include:

[0145] The normalization unit is used to normalize each alignment distance and determine the normalized alignment distance.

[0146] The localization unit is used to locate the target organization corresponding to the region of interest from each organization based on the normalized comparison distance and the reference distance in the preset template library.

[0147] In another embodiment, the aforementioned tissues include individual ribs, and the aforementioned region of interest includes lesion areas on the individual ribs.

[0148] In another embodiment, a different location device for the region of interest is provided. Based on the above embodiments, the distance measurement and determination module 12 may include:

[0149] The judgment unit is used to determine the organization to be calculated corresponding to the region of interest in each organization based on the location of the region of interest and the location of each organization; the organization to be calculated is either the left organization or the right organization.

[0150] The distance determination unit is used to determine the measurement distance between the region of interest and each organization to be calculated based on the location of the region of interest and the location of the organization to be calculated.

[0151] Optionally, the aforementioned distance determination unit is specifically used to determine the centerline corresponding to each organization to be calculated based on the location of each organization to be calculated; and to calculate the distance between the region of interest and each centerline based on the location of the region of interest, thereby obtaining the measured distance between the region of interest and each organization to be calculated.

[0152] In another embodiment, a different region of interest localization device is provided. Based on the above embodiments, the device further includes a template library creation module, which may include:

[0153] A sample acquisition unit is used to acquire multiple sample images; the sample images include tissues with complete structures.

[0154] The sample distance calculation unit is used to calculate a set of sample distances corresponding to each tissue in each sample image, and obtain multiple sets of sample distances corresponding to each sample image.

[0155] The mean processing unit is used to perform mean processing on the multiple sets of sample distances corresponding to each sample image for the same tissue, so as to obtain a set of reference distances corresponding to each tissue.

[0156] The template library creation unit is used to establish a correspondence between each organization and a corresponding set of reference distances to obtain a preset template library.

[0157] Each module in the aforementioned region of interest locating device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can invoke and execute the operations corresponding to each module.

[0158] In one embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to perform the following steps:

[0159] The location of the region of interest (ROI) and the location of each tissue are determined based on the acquired medical images. The measurement distance between the ROI and each tissue is determined based on the location of the ROI and the location of each tissue. The target tissue corresponding to the ROI is located from each tissue based on the measurement distance and the reference distance in the preset template library. The preset template library includes a set of reference distances corresponding to each tissue, which includes the reference distances between each tissue and itself and other tissues.

[0160] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0161] Each measured distance is matched with a set of reference distances corresponding to each tissue to determine the set of reference distances that are successfully matched; the tissue corresponding to the set of reference distances that are successfully matched is determined as the target tissue.

[0162] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0163] Calculate the difference between each measured distance and a set of reference distances corresponding to each tissue to obtain the difference value for each tissue; determine the minimum difference value among all difference values, and determine the set of reference distances corresponding to the minimum difference value as the successfully matched set of reference distances.

[0164] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0165] At least one measurement distance is selected from all measurement distances as the comparison distance; based on each comparison distance and the reference distance in the preset template library, the target tissue corresponding to the region of interest is located from each tissue; wherein, the number of the above comparison distances is less than the number of measurement distances.

[0166] In one embodiment, the aforementioned tissues include individual ribs, and the aforementioned region of interest includes lesion areas on the individual ribs.

[0167] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0168] Based on the location of the region of interest and the location of each organization, determine the organization to be calculated corresponding to the region of interest within each organization; the organization to be calculated is either the left-side organization or the right-side organization; based on the location of the region of interest and the location of the organization to be calculated, determine the measurement distance between the region of interest and each organization to be calculated.

[0169] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0170] Determine the centerline corresponding to each organization based on its location; calculate the distance between the region of interest and each centerline based on the location of the region of interest, and obtain the measured distance between the region of interest and each organization.

[0171] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0172] Acquire multiple sample images; the sample images include tissues with complete structures; calculate a set of sample distances corresponding to each tissue in each sample image to obtain multiple sets of sample distances corresponding to each sample image; in the multiple sets of sample distances corresponding to each sample image, perform mean processing on the multiple sets of sample distances for the same tissue to obtain a set of reference distances corresponding to each tissue; establish a correspondence between each tissue and the corresponding set of reference distances to obtain a preset template library.

[0173] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, the computer program performing the following steps when executed by a processor:

[0174] The location of the region of interest (ROI) and the location of each tissue are determined based on the acquired medical images. The measurement distance between the ROI and each tissue is determined based on the location of the ROI and the location of each tissue. The target tissue corresponding to the ROI is located from each tissue based on the measurement distance and the reference distance in the preset template library. The preset template library includes a set of reference distances corresponding to each tissue, which includes the reference distances between each tissue and itself and other tissues.

[0175] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0176] Each measured distance is matched with a set of reference distances corresponding to each tissue to determine the set of reference distances that are successfully matched; the tissue corresponding to the set of reference distances that are successfully matched is determined as the target tissue.

[0177] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0178] Calculate the difference between each measured distance and a set of reference distances corresponding to each tissue to obtain the difference value for each tissue; determine the minimum difference value among all difference values, and determine the set of reference distances corresponding to the minimum difference value as the successfully matched set of reference distances.

[0179] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0180] At least one measurement distance is selected from all measurement distances as the comparison distance; based on each comparison distance and the reference distance in the preset template library, the target tissue corresponding to the region of interest is located from each tissue; wherein, the number of the above comparison distances is less than the number of measurement distances.

[0181] In one embodiment, the aforementioned tissues include individual ribs, and the aforementioned region of interest includes lesion areas on the individual ribs.

[0182] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0183] Based on the location of the region of interest and the location of each organization, determine the organization to be calculated corresponding to the region of interest within each organization; the organization to be calculated is either the left-side organization or the right-side organization; based on the location of the region of interest and the location of the organization to be calculated, determine the measurement distance between the region of interest and each organization to be calculated.

[0184] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0185] Determine the centerline corresponding to each organization based on its location; calculate the distance between the region of interest and each centerline based on the location of the region of interest, and obtain the measured distance between the region of interest and each organization.

[0186] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0187] Acquire multiple sample images; the sample images include tissues with complete structures; calculate a set of sample distances corresponding to each tissue in each sample image to obtain multiple sets of sample distances corresponding to each sample image; in the multiple sets of sample distances corresponding to each sample image, perform mean processing on the multiple sets of sample distances for the same tissue to obtain a set of reference distances corresponding to each tissue; establish a correspondence between each tissue and the corresponding set of reference distances to obtain a preset template library.

[0188] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, performs the following steps:

[0189] The location of the region of interest (ROI) and the location of each tissue are determined based on the acquired medical images. The measurement distance between the ROI and each tissue is determined based on the location of the ROI and the location of each tissue. The target tissue corresponding to the ROI is located from each tissue based on the measurement distance and the reference distance in the preset template library. The preset template library includes a set of reference distances corresponding to each tissue, which includes the reference distances between each tissue and itself and other tissues.

[0190] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0191] Each measured distance is matched with a set of reference distances corresponding to each tissue to determine the set of reference distances that are successfully matched; the tissue corresponding to the set of reference distances that are successfully matched is determined as the target tissue.

[0192] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0193] Calculate the difference between each measured distance and a set of reference distances corresponding to each tissue to obtain the difference value for each tissue; determine the minimum difference value among all difference values, and determine the set of reference distances corresponding to the minimum difference value as the successfully matched set of reference distances.

[0194] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0195] At least one measurement distance is selected from all measurement distances as the comparison distance; based on each comparison distance and the reference distance in the preset template library, the target tissue corresponding to the region of interest is located from each tissue; wherein, the number of the above comparison distances is less than the number of measurement distances.

[0196] In one embodiment, the aforementioned tissues include individual ribs, and the aforementioned region of interest includes lesion areas on the individual ribs.

[0197] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0198] Based on the location of the region of interest and the location of each organization, determine the organization to be calculated corresponding to the region of interest within each organization; the organization to be calculated is either the left-side organization or the right-side organization; based on the location of the region of interest and the location of the organization to be calculated, determine the measurement distance between the region of interest and each organization to be calculated.

[0199] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0200] Determine the centerline corresponding to each organization based on its location; calculate the distance between the region of interest and each centerline based on the location of the region of interest, and obtain the measured distance between the region of interest and each organization.

[0201] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0202] Acquire multiple sample images; the sample images include tissues with complete structures; calculate a set of sample distances corresponding to each tissue in each sample image to obtain multiple sets of sample distances corresponding to each sample image; in the multiple sets of sample distances corresponding to each sample image, perform mean processing on the multiple sets of sample distances for the same tissue to obtain a set of reference distances corresponding to each tissue; establish a correspondence between each tissue and the corresponding set of reference distances to obtain a preset template library.

[0203] It should be noted that the data involved in this application (including but not limited to data used for analysis, stored data, and displayed data) are all information and data authorized by the user or fully authorized by all parties.

[0204] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0205] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0206] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A method for locating a region of interest, characterized in that, The method includes: The location of the region of interest and the location of each tissue are determined based on the acquired medical images; the tissues include each rib, the region of interest includes the lesion area on each rib, and the medical images also include each vertebra. The measurement distance between the region of interest and each of the tissues is determined based on the location of the region of interest and the location of each tissue. At least one measurement distance is selected from the measured distances as a comparison distance. The distance between two adjacent vertebrae is determined according to the position of each vertebra. The average distance between the two adjacent vertebrae is determined based on each distance. The ratio of each comparison distance to the average distance is determined as the normalized comparison distance. The number of comparison distances is less than the number of measurement distances. Based on the normalized comparison distances and the reference distances in the preset template library, the target organization corresponding to the region of interest is located from each organization; the preset template library includes a set of reference distances corresponding to each organization, and the set of reference distances includes the reference distances between each organization and itself and other organizations.

2. The method according to claim 1, characterized in that, The step of locating the target organization corresponding to the region of interest from each of the organizations based on the normalized comparison distances and the reference distances in the preset template library includes: The normalized alignment distances of each tissue are matched with a set of reference distances corresponding to each tissue to determine a set of reference distances that are successfully matched. The organization corresponding to the set of successfully matched reference distances is determined as the target organization.

3. The method according to claim 2, characterized in that, The step of matching each of the normalized alignment distances with a set of reference distances corresponding to each of the tissues to determine a set of successfully matched reference distances includes: Calculate the difference between the normalized alignment distance and a set of reference distances for each tissue to obtain the difference value for each tissue. The minimum difference value among the various difference values ​​is determined, and the set of reference distances corresponding to the minimum difference value is determined as the set of reference distances that are successfully matched.

4. The method according to any one of claims 1-3, characterized in that, Each of the tissues includes a left-side tissue and a right-side tissue. Determining the measurement distance between the region of interest and each of the tissues based on the location of the region of interest and the location of each tissue includes: Based on the location of the region of interest and the location of each of the tissues, the tissue to be calculated corresponding to the region of interest in each of the tissues is determined; the tissue to be calculated is either the left tissue or the right tissue. The measurement distance between the region of interest and each of the tissues to be calculated is determined based on the location of the region of interest and the location of the tissues to be calculated.

5. The method according to claim 4, characterized in that, The step of determining the measurement distance between the region of interest and each of the tissues to be calculated based on the location of the region of interest and the location of the tissues to be calculated includes: Determine the centerline corresponding to each of the tissues to be calculated based on their positions; Based on the location of the region of interest, the distance between the region of interest and each of the center lines is calculated to obtain the measured distance between the region of interest and each of the tissues to be calculated.

6. The method according to claim 1, characterized in that, The preset template library includes the following creation methods: Acquire multiple sample images; the sample images include tissues with complete structures; Calculate a set of sample distances corresponding to each tissue in each of the sample images to obtain multiple sets of sample distances corresponding to each of the sample images; In the multiple sets of sample distances corresponding to each sample image, the average of the multiple sets of sample distances for the same tissue is processed to obtain a set of reference distances for each tissue. Establish a correspondence between each organization and a corresponding set of reference distances to obtain the preset template library.

7. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 6.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

Images