Rib count information processing method, device and equipment

Abstract

The application discloses a rib count information processing method, device and equipment, relates to the technical field of medical images, and can output a rib count result with a larger use scale, so that the rib count information is related to more image applications, and the image range to which the rib count information is applicable is expanded. The method comprises the following steps: acquiring rib positioning information determined by training a positioning film; marking the rib positioning information in the positioning film to form a rib positioning graph with a cover mark, and recording rib mark information for rib counting starting from a preset position in the rib positioning graph; and fusing the rib mark information into different types of associated images by using the rib positioning graph, and outputting a rib count result in the associated images.

Description

Technical Field

[0001] This application relates to the field of medical imaging technology, and in particular to a method, apparatus and device for processing rib counting information. Background Technology

[0002] Medical imaging is widely used in medical, industrial, and other fields. It can be used to detect rib trauma and rib metastasis. When making diagnoses based on medical images, rib counting is a frequent requirement in the field of image diagnosis. The human body has 12 ribs; ribs 1 to 10 are connected anteriorly and posteriorly by the soft palate, while ribs 11 to 12 are floating. Accurate rib counting information greatly assists in clinical diagnosis.

[0003] Traditional rib counting methods are mainly manual, including two-dimensional and three-dimensional methods. The conventional two-dimensional method requires the user to visually track the ribs while observing medical images. Taking tomographic images as an example, multiple slices can be manually tracked to count from top to bottom or bottom to top. However, due to the increasing number of tomographic images, the location of a lesion requires the doctor to flip through the images from the starting layer to the lesion layer to pinpoint it. If there are multiple lesions, this process needs to be repeated multiple times. This process is not only slow but also prone to errors, affecting the doctor's efficiency in interpreting images. The conventional three-dimensional method requires rotating the tomographic image during diagnosis. However, medical imaging information systems may not have a three-dimensional module during diagnosis. Even if a three-dimensional module is available, rotation and adjustment of the count are still necessary. Furthermore, for obscured areas, rotation or correlation between the two-dimensional and three-dimensional images is required for better localization. Both two-dimensional and three-dimensional methods are quite time-consuming in rib counting on tomographic images, easily leading to missed diagnoses and increasing the difficulty of disease diagnosis.

[0004] Currently, various software programs offer automatic rib counting methods. With the advent of deep learning technology, rib counting uses medical image processing software to segment each rib for accurate counting, and some software even provides a full display of the ribs. However, automatic rib counting relies on scanning technology, which scans from one end of the rib's overall structure. For medical images where the ribs are not scanned from both ends, it is difficult to obtain accurate rib counting information, thus limiting the range of images to which the rib counting information is applicable. Summary of the Invention

[0005] In view of this, this application provides a method, apparatus and device for processing rib counting information. The main purpose is to solve the problem that the existing rib counting method relies on scanning technology. For CT images where the ribs are not scanned from both ends, it is difficult to obtain accurate rib counting information, which limits the image range to which the rib counting information is applicable.

[0006] According to a first aspect of this application, a method for processing rib count information is provided, the method comprising:

[0007] Obtain rib positioning information determined by training positioning patches;

[0008] The rib positioning information is marked in the positioning piece to form a rib positioning map with a covering mark. The rib positioning map records rib marking information that starts counting ribs from a preset position.

[0009] The rib location map is used to fuse the rib marking information into different types of associated images, and the rib counting results are output in the associated images.

[0010] Furthermore, the acquisition of rib positioning information determined by the training positioning patch specifically includes:

[0011] The rib counting model is constructed by inputting a sample positioning patch carrying a rib positioning tag into a neural network model for training. The rib counting model is used to identify the location information and counting information corresponding to the central vertebral point of each rib within the applicable range of the sample positioning patch.

[0012] The rib counting model is used to identify the positioning piece and determine the rib positioning information.

[0013] Furthermore, the rib positioning information includes the position information and count information corresponding to the central vertebral point of each rib in the positioning piece. The step of marking the rib positioning information in the positioning piece to form a rib positioning map with a covering mark specifically includes:

[0014] Based on the position and count information corresponding to the central vertebral point of each rib in the positioning piece, a position sequence of the human coordinate system under a preset standard protocol is generated;

[0015] According to the position sequence of the human coordinate system under the preset standard protocol, mark graphics are added to the preset area of ​​the positioning piece, and rib counting is performed starting from the preset position to form a rib positioning map with covering marks.

[0016] Further, the step of adding marked graphics to a preset area in the positioning patch according to the position sequence of the human coordinate system under the preset standard protocol, and counting ribs starting from a preset position to form a rib positioning map with covering markings, specifically includes:

[0017] According to the position sequence of the human coordinate system under the preset standard protocol, multiple coverage layers are generated in the positioning map. Each standard human coordinate system position coordinate corresponds to a coverage layer, and different coverage layers have different pixel values.

[0018] Based on the pixel values ​​corresponding to the coverage layer, a preset area mapped by the coverage layer in the positioning patch is determined;

[0019] A marked graphic is added to the preset area in the positioning piece, and ribs are counted starting from a preset position to form a rib positioning map with a covering mark.

[0020] Furthermore, after marking the rib positioning information in the positioning patch to form a rib positioning map with a covering mark, the method further includes:

[0021] Using the rib location map as the basis for rib counting, the rib counting information in the tomographic image after rib segmentation and center positioning is corrected to obtain the updated rib counting information in the tomographic image.

[0022] Furthermore, the step of using the rib location map as the basis for rib counting, and correcting the rib count information in the tomographic image after rib segmentation and center localization to obtain updated rib count information in the tomographic image, specifically includes:

[0023] The rib location map is matched with the tomographic image to obtain the rib marking information of the matching target rib in the rib location map;

[0024] The rib count information in the tomographic image is corrected based on the rib marking information of the target rib in the rib location map to obtain the updated rib count information of the tomographic image.

[0025] Furthermore, the step of fusing the rib marking information into different types of associated images using the rib location map, and outputting the rib counting result in the associated images, specifically includes:

[0026] The rib location map is used to store the rib marking information as location marking images of different association types;

[0027] The location marker images of different association types are embedded into the corresponding association images, and the rib count results are output in the association images.

[0028] According to a second aspect of this application, a processing apparatus for rib counting information is provided, the apparatus comprising:

[0029] The acquisition unit is used to acquire rib positioning information determined by the training positioning patch;

[0030] A marking unit is used to mark the rib positioning information in the positioning piece to form a rib positioning map with a covering mark. The rib positioning map records rib marking information for counting ribs starting from a preset position.

[0031] The fusion unit is used to fuse the rib marking information into different types of associated images using the rib location map, and output the rib counting result in the associated images.

[0032] Furthermore, the acquisition unit includes:

[0033] A construction module is used to train a neural network model based on a sample positioning patch carrying a rib positioning tag. The rib counting model is used to identify the location and counting information of the central vertebral point of each rib within the applicable range of the sample positioning patch.

[0034] The determination module is used to identify the positioning piece using the rib counting model and determine the rib positioning information.

[0035] Furthermore, the rib positioning information includes the position information and count information corresponding to the central vertebral point of each rib in the positioning piece, and the marking unit includes:

[0036] The generation module is used to generate a position sequence of the human coordinate system under a preset standard protocol based on the position information and counting information corresponding to the central vertebral point of each rib in the positioning piece.

[0037] The module is used to add marker graphics to a preset area in the positioning piece according to the position sequence of the human coordinate system under the preset standard protocol, and to count ribs starting from a preset position to form a rib positioning map with covering markers.

[0038] Furthermore, the added modules include:

[0039] The generation submodule is used to generate multiple coverage layers in the positioning map according to the position sequence of the human coordinate system under the preset standard protocol. Each position coordinate of the standard human coordinate system corresponds to a coverage layer, and different coverage layers have different pixel values.

[0040] The determination submodule is used to determine the preset area mapped by the coverage layer in the positioning piece based on the pixel value corresponding to the coverage layer;

[0041] A submodule is added to add a marked graphic to the preset area in the positioning piece and count the ribs starting from a preset position to form a rib positioning map with a covering mark.

[0042] Furthermore, the device also includes:

[0043] The correction unit is used to mark the rib positioning information in the positioning piece to form a rib positioning map with a covering mark, and then use the rib positioning map as the basis for rib counting to correct the rib counting information in the tomographic image after rib segmentation and center positioning, so as to obtain the updated rib counting information in the tomographic image.

[0044] Furthermore, the correction unit includes:

[0045] The acquisition module is used to match the rib location map with the tomographic image to obtain the rib marking information of the matching target rib in the rib location map.

[0046] The correction module is used to correct the rib count information in the tomographic image based on the rib mark information of the target rib in the rib location map, so as to obtain the updated rib count information of the tomographic image.

[0047] Furthermore, the fusion unit includes:

[0048] The storage module is used to store the rib marking information as positioning mark images of different association types using the rib positioning map;

[0049] An embedding module is used to embed the positioning marker images of different association types into the corresponding association images, and output the rib counting results in the association images.

[0050] According to a third aspect of this application, a storage medium is provided that stores a computer program thereon, which, when executed by a processor, implements the above-described method for processing rib counting information.

[0051] According to a fourth aspect of this application, a rib counting information processing device is provided, including a storage medium, a processor, and a computer program stored on the storage medium and executable on the processor, wherein the processor executes the program to implement the above-described rib counting information processing method.

[0052] By utilizing the above technical solutions, the method, apparatus, and device for processing rib counting information provided in this application, compared with the existing methods that rely on scanning technology for rib counting, can identify a wider range of rib counts by acquiring rib positioning information determined through training positioning films. The rib positioning information is then marked to form a rib positioning map with covering markings. This rib positioning map records rib marking information for rib counting starting from a preset position. Accurate rib counting information can be obtained without scanning CT images from both ends. The rib positioning map is used to fuse the rib marking information into different types of associated images, outputting rib counting results with a wider application scale. The rib counting information is associated with more image applications, expanding the image range applicable to rib counting information and improving the clinical image reading efficiency of rib counting.

[0053] The above description is only an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this application more obvious and understandable, the following are specific embodiments of this application. Attached Figure Description

[0054] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings:

[0055] Figure 1 A flowchart illustrating a method for processing rib counting information provided in an embodiment of this application is shown.

[0056] Figure 2 A flowchart illustrating another method for processing rib count information provided in an embodiment of this application is shown.

[0057] Figure 3 This illustration shows a schematic diagram of a rib counting information processing device provided in an embodiment of this application;

[0058] Figure 4 A schematic diagram of another rib counting information processing device provided in an embodiment of this application is shown. Detailed Implementation

[0059] The present application will be described in detail below with reference to the accompanying drawings and embodiments. It should be noted that, unless otherwise specified, the embodiments and features described in the embodiments of the present application can be combined with each other.

[0060] In related technologies, diagnosing based on processed images is the most common task for radiologists. Diagnosing, describing, and reporting fractures (lesions) in CT images is a crucial part of radiologist image interpretation. During diagnosis, it's necessary to determine the anatomical location, especially the number of ribs, and then describe the lesion for follow-up analysis or reference by other departments. However, with the widespread use of thin-slice CT images, the volumetric data of CT images is increasing. While doctors can detect minute fractures or lesions, confirming the location of lesions due to the increased number of slices has become challenging, especially describing ribs. Humans typically have 12 pairs of ribs, each with a unique number, numbered from top to bottom as rib 1, ..., rib 12. Without a consistent and quick reference point and other auxiliary methods, doctors need to scroll through the images from the initial slice to the lesion slice to determine the location of a lesion. This can be done by scrolling up or down to the final slice to count, which is time-consuming for a single lesion. If there are multiple lesions, this process needs to be repeated multiple times. This process is not only slow, but also prone to errors, which seriously affects the efficiency of doctors in reading images. Automated rib counting methods are crucial for doctors' efficiency and improving the quality of diagnosis and treatment. However, automated counting devices are generally based on post-processing software and have certain requirements for scanning, requiring scanning to start from both ends, which limits their applicability.

[0061] To address this problem, this embodiment provides a method for processing rib count information, such as... Figure 1 As shown, this method can be directly applied to the server side of a medical platform, and includes the following steps:

[0062] 101. Obtain rib positioning information determined by the training positioning patch.

[0063] The localization image is an image obtained by scanning the object to be tested using a device. The object to be tested is the scanned object of computed tomography (CT) imaging, typically the human chest in the medical field. Here, a neural network model can be used to train the localization image to obtain the center information of the ribs in the image, that is, the position information of the central vertebral point of the rib, and thus determine the rib localization information.

[0064] Understandably, rib location information can be used to locate ribs and vertebrae in a positioning image, obtaining positional information. It can also be used to count ribs and vertebrae in the positioning image, obtaining count information. Since the positioning image has a large scanning range, it can obtain positional and count information for all vertebrae and ribs in the positioning image. This makes the rib location information contain more comprehensive positional and count information for ribs and vertebrae. Even if the tomographic image does not contain all ribs and vertebrae, accurate rib positional and count information can still be obtained using rib location information, reducing the requirements of the rib counting information on the scanning process.

[0065] In the specific process of training the positioning patch using a neural network model, the vertebrae and ribs of the positioning patch can be marked first to form a positioning patch carrying location labels and count labels. This sample data is then input into the neural network model. The multi-layer structure of the neural network model is then used to train the mapping relationship between the ribs and vertebrae in the positioning patch and the location information based on the location labels, and to train the mapping relationship between the ribs and vertebrae in the positioning patch and the count information based on the count labels. This allows the position information of the vertebrae and ribs in the positioning patch to be predicted, and the count information of the ribs and vertebrae to be determined based on the position information of the ribs and vertebrae, thus outputting the rib positioning information.

[0066] The execution subject of this invention can be a rib counting information processing device, specifically configured on the server side of a medical platform, to obtain rib positioning information determined by training positioning films. Compared with rib and vertebral information that cannot be counted from the beginning or end, this rib positioning information has rib position information and rib counting information with a larger scanning range, which can provide more accurate rib position information and rib counting information, and can associate and apply the rib counting information to a wider range of medical images, thereby improving the clinical value of rib counting.

[0067] 102. Mark the rib positioning information in the positioning piece to form a rib positioning map with a covering mark.

[0068] Since rib location information and rib count information are not directly reflected in the positioning film, and considering the need for tagged storage of rib location and count information in the positioning film, it is necessary to mark the rib location information in the positioning film to form a rib location map. This rib location map records rib marking information starting from a preset position for rib counting. Here, the preset position corresponds to one end of the rib's overall structure, which can be either the upper or lower end. Specifically, in the process of marking the rib location information in the positioning film, the rib location information can first be used to generate a position sequence of human coordinates under a preset standard protocol. This preset standard protocol can be DICOM (Digital Imaging and Communications in Medicine), as the transmission and storage of medical images in the system follow the DICOM standard. Then, using the position sequence of human coordinates as the marking basis, different identifiers are used at the corresponding coordinate positions to form labels. These identifiers can be graphics, text, lines, strings, etc. Finally, the marked labels are embedded into the positioning map to form a rib location map with overlay markings.

[0069] Furthermore, to broaden the applicability of rib location maps, the rib location maps can be rotated to generate batch DICOM images with rib three-dimensional position information. These DICOM images contain orientation and position information, as well as multi-layered volumetric structures. Since the rib location maps store labels formed by rib count markers, these DICOM images with rib three-dimensional position information are equivalent to three-dimensional rib images viewed from different angles with count and position markers, eliminating the need for manual verification of rib count information.

[0070] It is understandable that by marking the rib location information in the positioning image, a rib location image with rib count labels can be formed. This rib location information has rib count labels with a larger scanning range, which can provide more accurate rib count information and assist other related images in rib counting, thereby expanding the application scenarios of rib counting.

[0071] 103. Using the rib location map, the rib marking information is fused into different types of associated images, and the rib counting results are output in the associated images.

[0072] Considering that there are many associated images that require rib counting in medical scenarios, such as CT images and MR images, the rib marking information in the rib localization map can be fused into different types of associated images, thereby expanding the scope of rib counting and improving its practical value.

[0073] Specifically, in the process of fusing rib marker information into different types of associated images, different fusion methods can be used for different types of associated images. For CT tomographic images, the number of ribs can be marked directly on top of each tomographic image, or it can be marked by embedding tags in each tomographic image; there is no limitation here. For MRI images, rigid registration can be used to mark the number of ribs, and the required number of ribs in the MRI image can be generated based on rigid registration. It should be noted that different types of associated images can be selected according to the actual application scenario. Here, the image information stored in the medical platform system can be referred to for association selection.

[0074] The rib counting information processing method provided in this application embodiment, compared with the existing method that relies on scanning technology for rib counting, obtains rib positioning information determined by training positioning films, which can identify a wider range of rib counts, and marks the rib positioning information to form a rib positioning map with a covering mark. This rib positioning map records rib marking information for rib counting starting from a preset position. Accurate rib counting information can be obtained without scanning CT images from both ends. The rib positioning map is used to fuse the rib marking information into different types of associated images, outputting rib counting results with a wider range of applications. The rib counting information is associated with more image applications, expanding the image range applicable to rib counting information and improving the clinical image reading efficiency of rib counting.

[0075] Furthermore, as a refinement and extension of the specific implementation of the above embodiments, and to fully illustrate the specific implementation process of this embodiment, this embodiment provides another method for processing rib counting information, such as... Figure 2 As shown, the method includes:

[0076] 201. Based on the sample positioning pieces carrying rib positioning tags, input them into the neural network model for training to construct a rib counting model.

[0077] The rib location tags can include position tags and count tags formed for vertebrae and ribs in the sample location patch. By training the sample location patch with a neural network model, a rib counting model can be constructed. This rib counting model is used to identify the position and count information corresponding to the central vertebral point of each rib within the applicable range of the sample location patch. The central position of the rib can be obtained, which is the position information corresponding to the central vertebral point of the rib. After counting, the count information of the central vertebral point of each rib and the count information of the corresponding rib can be obtained, thus obtaining the rib location information.

[0078] 202. Use the rib counting model to identify the positioning piece and determine the rib positioning information.

[0079] Understandably, the rib counting model is built using sample localization patches for training. During the training process, there are location matching and counting matching tasks designed for the rib localization labels. Here, a loss function can be set for each task, and the two tasks can be optimized simultaneously. That is, the weighted average of the loss functions of the two tasks can be used to obtain the final loss function. The parameters of the neural network model can be optimized in reverse through the final loss function to improve the training effect of the rib counting model.

[0080] 203. Based on the position information and count information corresponding to the central vertebral point of each rib in the positioning piece, generate a position sequence of the human body coordinate system under a preset standard protocol.

[0081] In this step, a position mark can be formed based on the position and count information of the central vertebral point of each rib in the positioning image to obtain the horizontal line information corresponding to the central vertebral point of each rib. The position sequence of the human coordinate system under the preset standard protocol is formed through the horizontal line information. The position sequence of the human coordinate system is different from the position sequence of the image coordinate system. It is usually a position sequence formed for the orientation of the human body. The orientation of the human body can be selected from the head to the legs or from the legs to the head.

[0082] 204. According to the position sequence of the human coordinate system under the preset standard protocol, add a marked graphic to the preset area in the positioning piece, and start counting the ribs from the preset position to form a rib positioning map with a covering mark.

[0083] The preset area is the region of interest for the location coordinates. The marked graphics can be in the form of references or annotations, usually in bitmap format, images or text, such as lines used to describe the rules for setting up equipment and parts.

[0084] Specifically, multiple coverage layers can be generated in the positioning map according to the position sequence of the human coordinate system under the preset standard protocol. Each standard human coordinate system position coordinate corresponds to a coverage layer. Different coverage layers have different pixel values. Based on the pixel values ​​corresponding to the coverage layers, the preset area mapped by the coverage layer in the positioning map is determined. Then, a marker graphic is added to the preset area in the positioning map, and rib counting is started from the preset position to form a rib positioning map with covering markers.

[0085] 205. Using the rib location map as the basis for rib counting, the rib counting information in the tomographic image after rib segmentation and center positioning is corrected to obtain the updated rib counting information in the tomographic image.

[0086] It is understandable that the rib count information obtained from rib segmentation and center localization in the tomographic image can be obtained by training a neural network on the tomographic image until the value output by the loss function meets the preset conditions. Since the tomographic image does not count ribs and vertebrae from the beginning or end, this rib count information needs to be further re-determined by professionals. Here, the rib count information in the rib localization map is used to assist in correcting the rib count information in the tomographic image in order to provide more accurate rib count results.

[0087] Specifically, the rib location map and the tomographic image can be matched to obtain the rib marker information of the matching target rib in the rib location map. Then, the rib count information in the tomographic image is corrected based on the rib marker information of the target rib in the rib location map to obtain the updated rib count information in the tomographic image. For example, the rib location map contains position and count markers for 1-12 ribs, and the tomographic image contains position and count markers for 1-5 ribs. However, the tomographic image is not actually scanned from the top or bottom. Through comparison, it is found that the rib count marker of 1 here is not the first rib, but the fifth rib in the human body. That is to say, the rib count information in the tomographic image is equivalent to the position and count markers of the fifth to ninth ribs in the human body, and the rib count information in the tomographic image is further updated.

[0088] 206. Using the rib location map, store the rib marking information as location mark images of different association types.

[0089] Considering the image features of different associated images, rib marking information can be stored as labeled images of different association types. Using localized labeled images of different association types can better match and label them with associated images. For example, CT tomographic images can be used as associated images. Considering the application of multiple tomographic images, labeled images can be formed on each layer based on rib marking information and placed on each tomographic image. Then, fused pixels or multiple overlay labels can be used for association matching. Alternatively, a segmented view of the rib can be used as an associated image. Considering the independent use of the segmented view, rib marking information can be stored as a separate labeled image, and then the labeled image can be directly mapped to the segmented view of the rib for association matching.

[0090] 207. Embed the positioning marker images of different association types into the corresponding association images, and output the rib counting results in the association images.

[0091] In practical applications, since the associated images contain rib count results, these results can be directly output and displayed in the system. Furthermore, the rib count results can be separated and combined with the ribs and vertebrae, allowing users to easily determine the location through the software system's positioning.

[0092] Furthermore, by directly saving the separated ribs and skeletal regions as new image sequences, and adding a digital model directly onto the associated images using a 3D model as the rib counting result, the rib counting results of the associated images can also be sent to external systems for use, thus expanding the application scope of the rib counting results.

[0093] For clinical imaging diagnosis, rib counting is a complex and time-consuming activity. Existing rib counting requires the use of dedicated medical systems, which have high computer configuration requirements. Moreover, different medical systems have different designs, making it difficult to integrate into existing physician work systems. This application combines image processing technology with preset standard protocols to fuse positioning markers into different types of related images, which can output rib counting results that are easy to use in clinical practice. This automated process can be integrated into actual user diagnosis, simplifying the daily work of physicians.

[0094] Furthermore, as Figures 1-2 In terms of specific implementation, this application provides a rib counting information processing device, such as... Figure 3 As shown, the device includes: an acquisition unit 31, a marking unit 32, and a fusion unit 33.

[0095] The acquisition unit 31 can be used to acquire rib positioning information determined by the training positioning patch;

[0096] The marking unit 32 can be used to mark the rib positioning information in the positioning piece to form a rib positioning map with a covering mark. The rib positioning map records rib marking information that starts counting ribs from a preset position.

[0097] The fusion unit 33 can be used to fuse the rib marking information into different types of associated images using the rib positioning map, and output the rib counting result in the associated image.

[0098] The rib counting information processing device provided in this invention, compared with the existing method that relies on scanning technology for rib counting, obtains rib positioning information determined by training positioning films, enabling the identification of a wider range of rib counts. It also marks the rib positioning information to form a rib positioning map with covering markings. This rib positioning map records rib marking information starting from a preset position for rib counting. Accurate rib counting information can be obtained without scanning CT images from both ends. The rib positioning map is used to fuse the rib marking information into different types of associated images, outputting rib counting results with a wider application scale. This allows the rib counting information to be associated with more image applications, expanding the image range applicable to rib counting information and improving the clinical image interpretation efficiency of rib counting.

[0099] In specific application scenarios, such as Figure 4 As shown, the acquisition unit 31 includes:

[0100] The construction module 311 can be used to train a neural network model based on a sample positioning piece carrying a rib positioning tag, and to construct a rib counting model. The rib counting model is used to identify the position information and counting information corresponding to the central vertebral point of each rib within the applicable range of the sample positioning piece.

[0101] The determination module 312 can be used to identify the positioning piece using the rib counting model and determine the rib positioning information.

[0102] In specific application scenarios, such as Figure 4 As shown, the rib positioning information includes the position information and count information corresponding to the central vertebral point of each rib in the positioning piece, and the marking unit 32 includes:

[0103] The generation module 321 can be used to generate a position sequence of the human body coordinate system under a preset standard protocol based on the position information and counting information corresponding to the central vertebral point of each rib in the positioning piece.

[0104] The addition module 322 can be used to add marked graphics to the preset area of ​​the positioning piece according to the position sequence of the human coordinate system under the preset standard protocol, and count ribs starting from the preset position to form a rib positioning map with covering markings.

[0105] In specific application scenarios, such as Figure 4 As shown, the added module 322 includes:

[0106] The generation submodule 3221 can be used to generate multiple coverage layers in the positioning map according to the position sequence of the human coordinate system under the preset standard protocol. Each standard human coordinate system position coordinate corresponds to a coverage layer, and different coverage layers have different pixel values.

[0107] The determining submodule 3222 can be used to determine the preset area mapped by the covering layer in the positioning piece based on the pixel value corresponding to the covering layer;

[0108] Adding submodule 3223 can be used to add marked graphics to the preset area in the positioning piece and count ribs starting from a preset position to form a rib positioning map with covering marks.

[0109] In specific application scenarios, such as Figure 4 As shown, the device further includes:

[0110] The correction unit 34 can be used to mark the rib positioning information in the positioning piece to form a rib positioning map with a covering mark, and then use the rib positioning map as the basis for rib counting to correct the rib counting information in the tomographic image after rib segmentation and center positioning, so as to obtain the updated rib counting information in the tomographic image.

[0111] In specific application scenarios, such as Figure 4 As shown, the correction unit 34 includes:

[0112] The acquisition module 341 can be used to match the rib location map with the tomographic image to obtain the rib mark information of the matching target rib in the rib location map.

[0113] The correction module 342 can be used to correct the rib count information in the tomographic image based on the rib mark information of the target rib in the rib location map, so as to obtain the updated rib count information of the tomographic image.

[0114] In specific application scenarios, such as Figure 4 As shown, the fusion unit 33 includes:

[0115] Storage module 331 can be used to store the rib marking information as positioning mark images of different associated types using the rib positioning map;

[0116] The embedding module 332 can be used to embed the positioning marker images of different association types into the corresponding association images, and output the rib counting results in the association images.

[0117] It should be noted that other corresponding descriptions of the functional units involved in the rib counting information processing device applicable to the server side provided in this embodiment can be found in [reference]. Figure 1 and Figure 2 The corresponding descriptions in [the document] will not be repeated here.

[0118] Based on the above, Figures 1-2 Accordingly, this application embodiment also provides a storage medium storing a computer program thereon, which, when executed by a processor, implements the above-described method. Figures 1-2 The method for processing rib count information is shown below;

[0119] Based on this understanding, the technical solution of this application can be embodied in the form of a software product. This software product can be stored in a non-volatile storage medium (such as a CD-ROM, USB flash drive, or portable hard drive), and includes several instructions to cause a computer device (such as a personal computer, server, or network device) to execute the methods described in the various implementation scenarios of this application.

[0120] Based on the above, Figures 1-2 The method shown, and Figures 3-4 To achieve the above objectives, the present application also provides a server-side physical device, specifically a computer, server, or other network device, as shown in the virtual device embodiment. This physical device includes a storage medium and a processor; the storage medium stores a computer program; the processor executes the computer program to implement the above-described... Figures 1-2 The method for processing rib count information is shown.

[0121] Optionally, the aforementioned physical devices may also include a user interface, a network interface, a camera, radio frequency (RF) circuitry, sensors, audio circuitry, a Wi-Fi module, etc. The user interface may include a display screen, input units such as a keyboard, etc., and optional user interfaces may also include USB interfaces, card reader interfaces, etc. The network interface may optionally include standard wired interfaces, wireless interfaces (such as Wi-Fi interfaces), etc.

[0122] Those skilled in the art will understand that the physical device structure for processing rib counting information provided in this embodiment does not constitute a limitation on the physical device, and may include more or fewer components, or combine certain components, or have different component arrangements.

[0123] The storage medium may also include an operating system and a network communication module. The operating system is a program that manages the hardware and software resources of the physical device for processing the aforementioned rib counting information, supporting the operation of the information processing program and other software and / or programs. The network communication module is used to enable communication between the various components within the storage medium, as well as communication with other hardware and software in the information processing physical device.

[0124] Through the above description of the embodiments, those skilled in the art can clearly understand that this application can be implemented using software plus necessary general-purpose hardware platforms, or it can be implemented in hardware. By applying the technical solution of this application, compared with the existing methods, this application can utilize rib location maps to fuse rib marking information into different types of associated images, outputting rib counting results with a larger application scale, associating rib counting information with more image applications, expanding the image range applicable to rib counting information, and improving the clinical image reading efficiency of rib counting.

[0125] Those skilled in the art will understand that the accompanying drawings are merely schematic diagrams of a preferred embodiment, and the modules or processes shown in the drawings are not necessarily essential for implementing this application. Those skilled in the art will understand that the modules in the apparatus of the embodiment can be distributed within the apparatus of the embodiment as described, or can be modified to be located in one or more apparatuses different from this embodiment. The modules of the above-described embodiment can be combined into one module, or further divided into multiple sub-modules.

[0126] The serial numbers in this application are for descriptive purposes only and do not represent the superiority or inferiority of any particular implementation scenario. The above disclosures are merely a few specific implementation scenarios of this application; however, this application is not limited thereto, and any variations conceived by those skilled in the art should fall within the protection scope of this application.

Claims

1. A method for processing rib count information, characterized in that, include: Obtain rib positioning information determined by training positioning patches; The rib positioning information is marked in the positioning piece to form a rib positioning map with a covering mark. The rib positioning map records rib marking information starting from a preset position to count the ribs. The preset position is one end of the rib's upper and lower body. The rib location map is used to fuse the rib marking information into different types of associated images, and the rib counting results are output in the associated images; The rib positioning information includes the position information and count information corresponding to the central vertebral point of each rib in the positioning piece; The step of marking the rib positioning information in the positioning patch to form a rib positioning map with a covering mark specifically includes: Based on the position and count information corresponding to the central vertebral point of each rib in the positioning piece, a position sequence of the human coordinate system under a preset standard protocol is generated; According to the position sequence of the human coordinate system under the preset standard protocol, mark graphics are added to the preset area of ​​the positioning piece, and ribs are counted starting from the preset position to form a rib positioning map with covering marks.

2. The method according to claim 1, characterized in that, The acquisition of rib positioning information determined by the training positioning patch specifically includes: The rib counting model is constructed by inputting a sample positioning patch carrying a rib positioning tag into a neural network model for training. The rib counting model is used to identify the location and counting information corresponding to the central vertebral point of each rib within the applicable range of the sample positioning patch. The rib counting model is used to identify the positioning piece and determine the rib positioning information.

3. The method according to claim 1, characterized in that, The step of adding marker graphics to a preset area in the positioning patch according to the position sequence of the human coordinate system under the preset standard protocol, and counting ribs starting from a preset position to form a rib positioning map with covering markers, specifically includes: According to the position sequence of the human coordinate system under the preset standard protocol, multiple coverage layers are generated in the positioning map. Each standard human coordinate system position coordinate corresponds to a coverage layer, and different coverage layers have different pixel values. Based on the pixel values ​​corresponding to the coverage layer, a preset area mapped by the coverage layer in the positioning patch is determined; A marked graphic is added to the preset area in the positioning piece, and ribs are counted starting from a preset position to form a rib positioning map with a covering mark.

4. The method according to claim 1, characterized in that, After marking the rib positioning information in the positioning patch to form a rib positioning map with a covering mark, the method further includes: Using the rib location map as the basis for rib counting, the rib counting information in the tomographic image after rib segmentation and center positioning is corrected to obtain the updated rib counting information in the tomographic image.

5. The method according to claim 4, characterized in that, The step of using the rib location map as the basis for rib counting, and correcting the rib count information in the tomographic image after rib segmentation and center positioning to obtain updated rib count information in the tomographic image, specifically includes: The rib location map is matched with the tomographic image to obtain the rib marking information of the matching target rib in the rib location map; The rib count information in the tomographic image is corrected based on the rib marking information of the target rib in the rib location map to obtain the updated rib count information of the tomographic image.

6. The method according to any one of claims 1-5, characterized in that, The step of fusing the rib marking information into different types of associated images using the rib location map, and outputting the rib counting result in the associated images, specifically includes: The rib location map is used to store the rib marking information as location marking images of different association types; The location marker images of different association types are embedded into the corresponding association images, and the rib count results are output in the association images.

7. A device for processing rib counting information, characterized in that, include: The acquisition unit is used to acquire rib positioning information determined by the training positioning patch; A marking unit is used to mark the rib positioning information in the positioning piece to form a rib positioning map with a covering mark. The rib positioning map records rib marking information starting from a preset position to count the ribs. The preset position is one end of the rib's upper and lower body. The fusion unit is used to fuse the rib marking information into different types of associated images using the rib location map, and output the rib counting result in the associated images; The rib positioning information includes the position information and count information corresponding to the central vertebral point of each rib in the positioning piece; The marking unit includes: The generation module is used to generate a position sequence of the human coordinate system under a preset standard protocol based on the position information and counting information corresponding to the central vertebral point of each rib in the positioning piece. The module is used to add marker graphics to a preset area in the positioning piece according to the position sequence of the human coordinate system under the preset standard protocol, and to count ribs starting from a preset position to form a rib positioning map with covering markers.

8. A storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the method for processing rib counting information as described in any one of claims 1 to 6.

9. A device for processing rib counting information, comprising a storage medium, a processor, and a computer program stored on the storage medium and executable on the processor, characterized in that, When the processor executes the program, it implements the method for processing rib count information as described in any one of claims 1 to 6.

Images