Image report verification method, computer device and readable storage medium

Abstract

The present application relates to an image report verification method, a computer device, and a readable storage medium. The image report verification method extracts features from an image to be verified in an acquired image report to be verified, obtaining a first feature map corresponding to the image to be verified; searches a preset feature map library to obtain a second feature map that matches the first feature map; searches a preset image visible feature library to obtain a first visible feature set that matches the second feature map; and verifies the image report to be verified based on the first visible feature set and a second visible feature set corresponding to the text to be verified in the image report to be verified. The image report verification method provided by the present application does not require manual verification and can improve the efficiency of verifying the image report to be verified.

Description

Technical Field

[0001] The present application relates to the field of imaging report technology, and in particular to an imaging report verification method, a computer device, and a readable storage medium. Background Art

[0002] With the development of medical imaging equipment, the role of medical imaging equipment in clinical diagnosis is becoming increasingly important. Usually, the doctor who reads the medical images will analyze the medical images and write an imaging report, and upload the imaging report to the superior doctor so that the superior doctor can review and verify the imaging report.

[0003] In traditional technology, senior doctors directly manually verify the imaging report. For example, they manually check the pictures and text in the imaging report one by one, rely on experience to analyze the lesion characteristics in the pictures, and compare and verify the analyzed lesion characteristics with the text one by one to determine whether the text description is correct. The verification results are recorded in the imaging report to complete the verification of the imaging report.

[0004] However, the above-mentioned method of manually verifying image reports has the problem of low efficiency. Summary of the Invention

[0005] Based on this, it is necessary to provide an image report verification method, computer equipment and readable storage medium to address the above technical problems.

[0006] In a first aspect, an embodiment of the present application provides an image report verification method, comprising:

[0007] Obtain the image report to be verified, which includes the image to be verified and the text to be verified;

[0008] Perform feature extraction on the image to be verified to obtain a first feature map corresponding to the image to be verified;

[0009] Searching for a second feature map that matches the first feature map in a preset feature map library;

[0010] Searching a preset image visible element library to obtain a first visible element set that matches the second feature map;

[0011] The image report to be verified is verified based on the first seen feature set and the second seen feature set corresponding to the text to be verified.

[0012] In one embodiment, the imaging report verification method further includes:

[0013] Determining a feature map index corresponding to the first feature map according to the features of the first feature map;

[0014] Searching for a second feature map that matches the first feature map in a preset feature map library includes:

[0015] A feature map corresponding to the feature map index is searched in a preset feature map library, and the feature map matching the feature map index is determined as the second feature map.

[0016] In one embodiment, if the feature map corresponding to the feature map index includes multiple feature maps, the image report verification method further includes:

[0017] Determine the similarity between the first feature map and each feature map corresponding to the feature map index, and determine the feature map with the greatest similarity as the second feature map.

[0018] In one embodiment, the imaging report verification method further includes:

[0019] Entity extraction is performed on the document to be verified to obtain the second seen feature set corresponding to the document to be verified.

[0020] In one embodiment, verifying the image report to be verified based on the first set of observed elements and the second set of observed elements corresponding to the text to be verified includes:

[0021] Each seen element in the second seen element set is matched with each seen element in the first seen element set, and the second seen elements that are successfully matched and / or unmatched are marked in the image report to be verified.

[0022] In one embodiment, the imaging report verification method further includes:

[0023] If the first seen element set contains redundant seen elements, a mark corresponding to the redundant seen elements will be displayed in the image report to be verified; the redundant seen elements are seen elements included in the first seen element set and not included in the second seen element set.

[0024] In one embodiment, the imaging report verification method further includes:

[0025] Standardizing each observed element in the second observed element set to obtain a standard second observed element set;

[0026] Matching each seen feature in the second seen feature set with each seen feature in the first seen feature set, including:

[0027] Each seen feature in the standard second seen feature set is matched with each seen feature in the first seen feature set.

[0028] In one embodiment, each element in the second element set is standardized to obtain a standard second element set, including:

[0029] Matching the entity to be matched with each entity included in the preset term base; the entity to be matched is the entity corresponding to each element in the second element set;

[0030] If the preset term base contains an entity that matches the entity to be matched, then determining a standard second-seen element set based on the entity that matches the entity to be matched;

[0031] If the preset term library does not contain an entity that matches the entity to be matched, the similarity between the entity to be matched and each entity contained in the preset term library is calculated, and the entity in the preset term library whose similarity with the entity to be matched meets a preset similarity threshold is determined as the entity that matches the entity to be matched, and the standard second-seen element set is determined based on the entity that matches the entity to be matched.

[0032] In one embodiment, determining a standard second-seen element set based on an entity that matches the entity to be matched includes:

[0033] Determine whether the entity that matches the entity to be matched is a standard entity;

[0034] If the entity that matches the entity to be matched is not a standard entity, the standard entity corresponding to the entity to be matched is found from the preset term library, and the entity to be matched is replaced with the standard entity corresponding to the entity to be matched, and the entity to be matched is determined to be an entity in the standard second-seen element set.

[0035] In one embodiment, the imaging report verification method further includes:

[0036] If the similarity between each entity in the preset term base and the entity to be matched does not meet the preset similarity threshold, the entity to be matched is determined as a matching missing entity.

[0037] In one embodiment, the imaging report verification method further includes:

[0038] The matching results between the entity to be matched and each entity in the preset terminology library are displayed in the image report to be verified; the matching results include the entity to be matched, the entity corresponding to the entity to be matched in the preset terminology library, and the corresponding relationship between the entity to be matched and the entity to be matched in the preset terminology library.

[0039] In a second aspect, an embodiment of the present application provides an image report verification device, comprising:

[0040] An acquisition module is used to obtain an image report to be verified, which includes a picture to be verified and a text to be verified;

[0041] An extraction module is used to extract features from the image to be verified and obtain a first feature map corresponding to the image to be verified;

[0042] A search module, configured to search a preset feature map library for a second feature map that matches the first feature map;

[0043] The search module is further used to search for a first seen element set matching the second feature map in a preset image seen element library;

[0044] The verification module is used to verify the image report to be verified based on the first seen element set and the second seen element set corresponding to the text to be verified.

[0045] In a third aspect, an embodiment of the present application provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the steps of the method provided in the first aspect are implemented.

[0046] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium having a computer program stored thereon, which implements the steps of the method provided in the first aspect above when the computer program is executed by a processor.

[0047] In a fifth aspect, an embodiment of the present application provides a computer program product, including a computer program, which implements the steps of the method provided in the first aspect when executed by a processor.

[0048] The embodiment of the present application provides an imaging report verification method, a computer device, and a readable storage medium. The imaging report verification method obtains the image to be verified and the text to be verified included in the imaging report to be verified; performs feature extraction on the image to be verified to obtain a first feature map corresponding to the image to be verified; searches for a second feature map that matches the first feature map in a preset feature map library; queries a preset image visible element library to obtain a first visible element set that matches the second feature map; and verifies the imaging report to be verified based on the first visible element set and the second visible element set corresponding to the text to be verified. The imaging report verification method provided in this embodiment verifies the imaging report to be verified by searching for relevant information in a preset feature map library and a preset image visible element library. In this way, during the verification process, there is no need to rely on the experience of medical staff, and there is no need to rely on manual verification of the imaging report to be verified, thereby improving the efficiency of verifying the imaging report to be verified and avoiding waste of manpower and time. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the conventional technology, the following briefly introduces the drawings required for use in the embodiments or the conventional technology descriptions. Obviously, the drawings described below are only some embodiments of the present application. For different technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0050] Figure 1 A schematic diagram of an application scenario of an image report verification method provided by an embodiment;

[0051] Figure 2 A schematic flow chart of the steps of an image report verification method provided in one embodiment;

[0052] Figure 3 A schematic flow chart of the steps of an image report verification method provided in another embodiment;

[0053] Figure 4 A schematic flow chart of the steps of an image report verification method provided in another embodiment;

[0054] Figure 5 A schematic flow chart of the steps of an image report verification method provided in another embodiment;

[0055] Figure 6 A schematic flow chart of the steps of an image report verification method provided in another embodiment;

[0056] Figure 7 A schematic diagram of the ontology structure of medical terms provided in one embodiment;

[0057] Figure 8 A schematic flow chart of the steps of an image report verification method provided in another embodiment;

[0058] Figure 9 A schematic structural diagram of an image report verification device provided in one embodiment;

[0059] Figure 10 A schematic diagram of the structure of a computer device provided in one embodiment. DETAILED DESCRIPTION

[0060] To make the above-mentioned objects, features, and advantages of the present application more clearly understood, the specific embodiments of the present application are described in detail below with reference to the accompanying drawings. The following description sets forth many specific details to facilitate a full understanding of the present application. However, the present application can be implemented in many other ways than those described herein, and those skilled in the art can make similar improvements without violating the scope of the present application. Therefore, the present application is not limited to the specific embodiments disclosed below.

[0061] The serial numbers assigned to the components in this document, such as "first", "second", etc., are only used to distinguish the objects described and do not have any order or technical meaning.

[0062] The image report verification method provided in the embodiment of the present application can be applied to Figure 1In the application environment shown, the application environment includes a terminal 100 and a medical scanning device 200. The terminal 100 can communicate with the medical scanning device 200 via a network. The medical scanning device 200 is used to scan a test subject, obtain multiple images corresponding to the test subject, and send the multiple images to the terminal. The terminal 100 can obtain an image finding file containing image findings and diagnosis results issued by medical personnel based on the images, that is, the terminal can obtain an imaging report (image and image finding file) for each test subject. The terminal 100 can be, but is not limited to, various personal computers, laptops, and tablet computers. The medical scanning device 200 can be, but is not limited to, a CT (Computed Tomography) device, a PET (Positron Emission Computed Tomography) device, or an MR (Magnetic Resonance) device. This embodiment does not limit the type of medical scanning device 200.

[0063] The following specific embodiments describe in detail the technical solution of this application and how the technical solution of this application solves the technical problem. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The embodiments of this application will be described below in conjunction with the accompanying drawings.

[0064] In one embodiment, Figure 2 As shown, a method for verifying an image report is provided. In this embodiment, the method is applied to Figure 1 The method includes the following steps:

[0065] Step 200: Obtain an image report to be verified, where the image report to be verified includes a picture to be verified and a text to be verified.

[0066] The image report to be verified may include an image of the subject being tested, as well as an image file and image image of the imaging findings and diagnostic results issued by the medical staff based on the acquired image of the subject being tested. The image report to be verified includes the image to be verified (the image), and the file to be verified (a text consisting of the imaging findings file and the diagnostic results). The image report to be verified is stored in the terminal's memory, and the terminal can directly access it from the memory when needed.

[0067] Step 210: Perform feature extraction on the image to be verified to obtain a first feature map corresponding to the image to be verified.

[0068] After receiving the image report to be verified, the terminal performs feature extraction on the image to be verified in the report to obtain a first feature map corresponding to the image to be verified. This first feature map best reflects the essence of the image to be verified or distinguishes the image to be verified. This embodiment does not limit the specific method for extracting features from the image to be verified; as long as the method can achieve the desired function, it will be sufficient.

[0069] In an optional embodiment, a nonlinear feature extraction method may be used to extract features of the verification image to obtain a first feature map; or a key area (lesion area) of the verification image may be identified and segmented to obtain the first feature map.

[0070] In another optional embodiment, a feature extraction model obtained by pre-training a deep neural network using image samples can be used to extract features from the image to be verified.

[0071] Step 220: Search a preset feature map library to obtain a second feature map that matches the first feature map.

[0072] A preset feature map library includes feature maps that correspond one-to-one to multiple images and is stored in the terminal's memory. After obtaining a first feature map corresponding to the image to be verified, the terminal traverses all feature maps in the preset feature map library and matches them with the first feature map to obtain a second feature map that matches the first feature map.

[0073] Step 230: Search the preset image visible element library to obtain a first visible element set that matches the second feature map.

[0074] The preset image feature library includes each feature in the preset feature library Figure 1 A corresponding set of image elements is stored in the terminal's memory. After obtaining a second feature map that matches the first feature map, the terminal searches for the second feature map in the preset image element library. After finding the second feature map, the terminal obtains the first feature set corresponding to the second feature map. The first feature set is also the feature set corresponding to the first feature map.

[0075] Step 240: Verify the image report to be verified based on the first found element set and the second found element set corresponding to the text to be verified.

[0076] The second set of observed elements corresponding to the text to be verified refers to the collection of all elements in the document to be verified. After obtaining the first set of observed elements corresponding to the first feature map of the image to be verified from the preset set of observed elements in the image, the terminal compares this first set of observed elements with the second set of observed elements corresponding to the text to be verified to verify the image report to be verified. In other words, the terminal performs a one-to-one match between the elements in the first set of observed elements and the elements in the second set of observed elements corresponding to the text to be verified to verify the image report to be verified.

[0077] The imaging report verification method provided in the embodiment of the present application obtains the image to be verified and the text to be verified included in the imaging report to be verified; performs feature extraction on the image to be verified to obtain a first feature map corresponding to the image to be verified; searches in a preset feature map library to obtain a second feature map that matches the first feature map; queries in a preset image visible element library to obtain a first visible element set that matches the second feature map; and verifies the imaging report to be verified based on the first visible element set and the second visible element set corresponding to the text to be verified. The imaging report verification method provided in the present embodiment verifies the imaging report to be verified by searching for relevant information in a preset feature map library and a preset image visible element library. In this way, during the verification process, there is no need to rely on the experience of medical staff, and there is no need to rely on manual verification of the imaging report to be verified, thereby improving the efficiency of verifying the imaging report to be verified and avoiding waste of manpower and time.

[0078] In one embodiment, Figure 3 As shown, each feature map in the preset feature map library has a corresponding feature map index, that is, the feature map corresponding to it can be obtained through the feature map index. The feature map index is determined based on the specific features of each feature map. In this regard, the steps of the image report verification method also include:

[0079] Step 300: Determine a feature map index corresponding to the first feature map based on the features of the first feature map.

[0080] After obtaining the first feature map corresponding to the image to be verified, the terminal determines the feature map index corresponding to the first feature map based on the features of the first feature map. The feature map index can be a specific feature that best represents the first feature map. In other words, a feature that best represents the first feature map is selected from the multiple features included in the first feature map as the feature map index of the first feature map. For example, if the first feature map corresponding to the image to be verified is a grayscale feature map of the image to be verified, the feature map index corresponding to the first feature map can be a grayscale value or grayscale value range that best represents the grayscale feature map.

[0081] When each feature map in the preset feature map library has a corresponding feature map index, a possible implementation method of searching the preset feature map library for a second feature map that matches the first feature map includes:

[0082] Step 310: Search for a feature map corresponding to the feature map index in a preset feature map library, and determine the feature map matching the feature map index as the second feature map.

[0083] After obtaining the feature map index corresponding to the first feature map, the terminal compares the feature map index corresponding to the first feature map with all the preset feature map indices in the preset feature map library. After finding the preset feature map index that matches the feature map index corresponding to the first feature map, the terminal obtains the feature map corresponding to the preset feature map index in the preset feature map library. This feature map is the second feature map that matches the first feature map.

[0084] In this embodiment, the second feature map is searched in a preset feature map library using the feature map index corresponding to the first feature map. Since the feature map index carries less information than the first feature map, searching the preset feature map library for the second feature map is more efficient, thereby improving the efficiency and practicality of the imaging report verification method.

[0085] In one embodiment, when the feature map index corresponding to the second feature map is used to search for a feature map corresponding to the feature map index in the preset feature map library, multiple feature maps corresponding to the feature map index are found in the preset feature map library. In other words, the preset feature map library includes multiple feature map clusters, each feature map cluster is a collection of feature maps with a high degree of similarity, and each feature map cluster has a feature map index. In this case, the image report verification method further includes:

[0086] Determine the similarity between the first feature map and each feature map corresponding to the feature map index, and determine the feature map with the greatest similarity as the second feature map.

[0087] The terminal uses the feature map index corresponding to the first feature map to search for a feature map index corresponding to the first feature map in a preset feature map library. The feature map index corresponds to a feature map cluster, and the feature map cluster has multiple feature maps. The terminal compares the first feature map with the multiple feature maps in the feature map cluster, calculates the similarity between the first feature map and each feature map in the feature map cluster, compares the multiple similarities, and determines the feature map with the greatest similarity as the second feature map corresponding to the first feature map.

[0088] In this embodiment, the feature map index corresponding to the feature map index of the first feature map is first searched in the preset feature map library through the feature map index, and then the second feature map is searched in multiple feature maps in the feature map cluster corresponding to the feature map index. Figure 1 Determining the second characteristic map by comparing it with the first characteristic map can improve the efficiency of finding the second characteristic map, thereby improving the efficiency and practicality of the image report verification method.

[0089] In one embodiment, the image verification method further comprises:

[0090] Entity extraction is performed on the document to be verified to obtain the second seen feature set corresponding to the document to be verified.

[0091] The text to be verified includes the image documentation provided by medical personnel based on the image to be verified. Each entity in the document to be verified is an element of the image documentation corresponding to the image to be verified. After receiving the text to be verified, the terminal extracts the entities in the text to be verified, obtaining a second set of visual elements consisting of multiple visual elements corresponding to the text to be verified. Specifically, the entities in the text to be verified include specific body parts, specific disease names, and specific disease symptoms.

[0092] In an optional embodiment, if Figure 4 As shown in the figure, the steps of extracting entities from the text to be verified and obtaining the second feature set include:

[0093] Step 400: Segment and deduplicate the text to be verified according to the preset delimiters to obtain a set of text segments.

[0094] The text to be verified includes multiple sentences to be verified. The terminal segments the multiple sentences to be verified according to preset delimiters, and deduplicates the segmented text to obtain multiple text segments for each sentence to be verified, forming a text segment set corresponding to the text to be verified.

[0095] Step 410: Perform entity recognition and relationship extraction on each text segment in the text segment set to obtain a second feature set.

[0096] After obtaining the set of text segments corresponding to the text to be verified, the terminal identifies the entities in each segment and extracts the relationships between each entity to obtain a second set of observed features. For example, the entities in each segment can be represented by a five-tuple: <organ, symptom, lesion location, lesion shape, lesion size range>, with each five-tuple being a feature point. Feature points corresponding to multiple text segments can form the second set of observed features corresponding to the text to be verified.

[0097] In this embodiment, by extracting entities from the text to be verified, a second set of elements corresponding to the text to be verified can be obtained. This method of obtaining the second set of elements is simple in logic and easy to implement.

[0098] In one embodiment, a possible implementation method for verifying an image report to be verified based on a first set of observed elements and a second set of observed elements corresponding to the text to be verified includes the following steps:

[0099] Each seen element in the second seen element set is matched with each seen element in the first seen element set, and the second seen elements that are successfully matched and / or unmatched are marked in the image report to be verified.

[0100] The preset image visible element library includes the image visible element set corresponding to each feature map in all preset feature map libraries. After obtaining the first visible element set matched by the second feature map, the terminal matches each visible element in the second visible element set corresponding to the text to be verified with all visible elements in the first visible element set to determine whether there are visible elements in the first visible element set that match all visible elements in the second visible element set. After the matching is completed, if the terminal is successful, that is, if there are visible elements in the first visible element set that match all visible elements in the second visible element set, it will be marked in the image report to be verified; if the matching fails, that is, if there are visible elements in the first visible element set that do not match the visible elements in the second visible element set, it will also be marked in the image report to be verified. This embodiment does not limit the specific marking method.

[0101] In a specific embodiment, the terminal may use different colors to mark the elements that are successfully matched and those that fail to match, or may use marking boxes of different shapes to mark the elements that are successfully matched and those that fail to match.

[0102] In this embodiment, by matching each element in the first element set with each element in the second element set, a more accurate verification of the text to be verified in the image report to be verified can be achieved. Furthermore, the matching results are marked in the image report to be verified, which makes it easier for medical staff to clearly obtain the verification results of the image report to be verified.

[0103] In one embodiment, the image report verification method further comprises:

[0104] If the first seen element set contains redundant seen elements, a mark corresponding to the redundant seen elements will be displayed in the image report to be verified; the redundant seen elements are seen elements included in the first seen element set and not included in the second seen element set.

[0105] After matching each set of visual elements in the second set of visual elements with all the visual elements in the first set of visual elements, the terminal determines that the first set of visual elements contains redundant visual elements, and then displays a mark corresponding to the redundant visual element in the image report to be verified. In other words, if the terminal determines through matching that a visual element is included in the first set of visual elements but not in the second set of visual elements, that is, the visual element is missing in the document to be verified, the terminal will mark the location of the redundant visual element in the image report to be verified to remind medical staff that the visual element is missing at that location.

[0106] In a specific embodiment, the terminal can directly display the redundant elements at the corresponding position of the image report to be verified. The terminal can also display a marking index at the position of the redundant elements in the image report to be verified. When the medical staff places the mouse on the marking index, the specific content of the redundant elements can be displayed, and a selection control (button) can be displayed for the medical staff to choose whether to add the redundant elements to the image report to be verified, so that the medical staff can make a selection according to actual needs.

[0107] In this embodiment, if the first set of visual elements in the preset image visual element library contains redundant visual elements that are not included in the second set of visual elements corresponding to the document to be verified, a marker indicating the redundant visual element set can be displayed in the image report to be verified, so that medical personnel can clearly obtain the verification results of the image report to be verified. Furthermore, the image report to be verified can be supplemented based on the redundant visual elements, resulting in a more accurate, comprehensive, and standardized image report, thereby improving the practicality and reliability of the image report verification method.

[0108] In one embodiment, see Figure 5 , the steps of the image report verification method also include:

[0109] Step 500: Standardize each seen element in the second seen element set to obtain a standard second seen element set.

[0110] The terminal obtains a second seen element set corresponding to the text to be verified, and before matching the second seen element set with a first seen element set in a preset image seen element library, standardizes each seen element in the second seen element set to obtain a standard second seen element set.

[0111] After obtaining the standard second seen element set, the step of matching each seen element in the second seen element set with each seen element in the first seen element set includes:

[0112] Step 510: Match each seen element in the standard second seen element set with each seen element in the first seen element set.

[0113] Each visual element in the first visual element set is a standardized visual element. After obtaining each standard visual element in the standard second visual element set, the terminal matches each standard visual element with each visual element in the first visual element set to verify the image report to be verified.

[0114] In this embodiment, each element in the second set of elements is standardized before being matched against all elements in the first set of elements. This can prevent different descriptions of image findings in text produced by different medical personnel, improve the efficiency of image matching, and thus improve the efficiency of verifying image reports.

[0115] In one embodiment, Figure 6 As shown, a possible implementation method of standardizing each visible element in the second visible element set to obtain a standard second visible element set includes the following steps:

[0116] Step 600: Match the entity to be matched with each entity included in the preset term library; the entity to be matched is the entity corresponding to each seen element in the second seen element set.

[0117] A term represents a single, specialized concept within a professional context, abstracting and generalizing the essential characteristics of a subject. In this example, medical terminology is used, which refers to words and phrases representing scientific concepts within the medical field. The preset terminology library contains various entity types. For example, the preset terminology library includes entities for diseases, symptoms, diagnostic and treatment methods, body structures, and drug names.

[0118] Specifically, the preset terminology library includes each standard entity, each standard entity's alias, and the relationship between multiple standard entities. In this embodiment, first, the ontology structure of medical terms is constructed according to actual business needs, such as Figure 7As shown, the ontology structure of medical terms includes standard entities: specific body parts, specific diseases and specific symptoms, as well as the relationships between diseases and body parts and symptoms. For example, the symptoms of a disease, the body part where the disease is located, etc. Each standard entity has one or more aliases. For example, the standard entity is the temporomandibular joint, and the aliases of this standard entity are mandibular joint and temporomandibular joint. Secondly, the entities defined in the ontology structure of medical terms and the relationships between entities are identified and extracted from a large number of structured and unstructured data sources such as medical literature, books, reports, and cases. Then, each entity and the relationship between each entity are represented in a preset shape and stored in a preset terminology library. For example, each entity and the relationship between each entity can be represented in the form of a triple, <entity, attribute, attribute value>, <entity, relationship, entity>.

[0119] Each element in the second set of elements has a corresponding entity, namely, an entity to be matched. Each element in the second set of elements may have multiple entities to be matched. The terminal matches the entities to be matched against all entities in the preset terminology library, specifically searching the preset terminology library for an entity that matches each entity in each element in the second set of elements.

[0120] Step 610: If the preset term library contains an entity that matches the entity to be matched, a standard second-found element set is determined based on the entity that matches the entity to be matched.

[0121] After matching the entity to be matched against all entities in the preset terminology library, the terminal determines that the preset terminology library contains an entity that matches the entity to be matched. The terminal then determines the entity in the preset terminology library that matches the entity to be matched as an entity in the standard second-viewed element set corresponding to the second-viewed element. In other words, the terminal determines the entity in the standard second-viewed element set that is the result of normalizing the entities in the second-viewed element set based on the entity in the preset terminology library that matches the entity to be matched.

[0122] Step 620: If the preset term library does not contain an entity that matches the entity to be matched, the similarity between the entity to be matched and each entity contained in the preset term library is calculated, and the entity in the preset term library whose similarity with the entity to be matched meets the preset similarity threshold is determined as the entity that matches the entity to be matched, and the standard second seen element set is determined based on the entity that matches the entity to be matched.

[0123] After matching the entity to be matched with all entities in the preset terminology library, if the terminal determines that the preset terminology library does not contain an entity that matches the entity to be matched, the terminal calculates the similarity between the entity to be matched and each entity in the preset terminology library, compares the multiple similarities calculated with the preset similarity threshold, and selects the entity corresponding to the similarity that meets the preset similarity threshold as the entity that matches the entity to be matched. Specifically, the method for determining whether the similarity meets the preset similarity threshold includes: the terminal compares the multiple similarities with the preset similarity threshold; if the similarity is greater than or equal to the preset similarity threshold, it indicates that the similarity meets the preset similarity threshold; if the similarity is less than the preset similarity threshold, it indicates that the similarity does not meet the preset similarity threshold.

[0124] After obtaining an entity that matches the entity to be matched, the terminal determines a standard second-visible element set based on the entity. After obtaining an entity that matches the entity to be matched, the terminal may directly determine the entity as an entity in the standard second-visible element set, or may determine the entity as an entity in the second-visible element set after processing the entity. This embodiment does not limit the specific method for determining the standard second-visible element based on the entity that matches the entity to be matched, as long as the function is achieved.

[0125] In one embodiment, Figure 8 As shown, a possible implementation method for determining a standard second-seen element set according to an entity that matches the entity to be matched includes the following steps:

[0126] Step 800: Determine whether the entity that matches the entity to be matched is a standard entity.

[0127] The preset terminology database contains a standard entity corresponding to an entity and an alias of the standard entity. After obtaining an entity that matches the entity to be matched, the terminal determines whether the entity that matches the entity to be matched is a standard entity.

[0128] Step 810: If the entity that matches the entity to be matched is not a standard entity, the standard entity corresponding to the entity to be matched is found from the preset term library, and the entity to be matched is replaced with the standard entity corresponding to the entity to be matched, and the replaced entity to be matched is determined as an entity in the standard second seen element set.

[0129] The terminal determines whether the entity that matches the entity to be matched is a standard entity. If it is determined that the entity that matches the entity to be matched is a standard entity, the entity that matches the entity to be matched is directly used as the entity in the standard second seen element set. In other words, the entity that matches the entity to be matched is the entity in the standard second seen element set after the entities in the second seen element set are standardized. If it is determined that the entity that matches the entity to be matched is not a standard entity (the entity that matches the entity to be matched is an alias of the standard entity), the standard entity corresponding to the entity that matches the entity to be matched (the entity to be matched) is searched in the preset terminology library, and the entity to be matched is replaced with the standard entity. The entity to be matched after replacement is an entity in the standard second seen element set.

[0130] In this embodiment, by matching the entity to be matched with the entity in the preset term library, the entity to be matched is replaced with the standard entity according to the entity in the preset term library, thereby achieving standardization of the entity to be matched. The method for standardization of the entity to be matched provided by this embodiment is simple in logic and easy to implement.

[0131] In one embodiment, the image report verification method further includes:

[0132] If the similarity between each entity in the preset term base and the entity to be matched does not meet the preset similarity threshold, the entity to be matched is determined as a matching missing entity.

[0133] After comparing the multiple similarities with a preset similarity threshold, the terminal determines that the similarities between the entity in the preset terminology library and the entity to be matched do not meet the preset similarity threshold, indicating that the preset terminology library does not store an entity that matches the entity to be matched. The terminal then determines the entity to be matched as a missing match entity. In other words, if no entity matching the entity to be matched is found in the preset terminology library, the entity to be matched is determined to be a missing match entity.

[0134] In one embodiment, the image report verification method further includes:

[0135] The image report to be verified displays the matching results between the entity to be matched and the entity in the preset terminology library; the matching results include the entity to be matched, the entity corresponding to the entity to be matched in the preset terminology library, and the corresponding relationship between the entity to be matched and the entity to be matched in the preset terminology library.

[0136] After matching the entity to be matched with the entity contained in the preset terminology library, the terminal displays the matching result in the image report to be verified.

[0137] Specifically, the entity that matches the entity to be matched in the preset terminology library is recorded as the first entity. The first entity is a standard entity, that is, the entity to be matched is the same as the first entity. Then, the matching result displayed in the image report to be verified is <entity to be matched, first entity, 0>, where 0 indicates that the preset terminology library contains an entity that matches the entity to be matched (the first entity), and the first entity is a standard entity. If the first entity is not a standard entity, the matching result displayed in the image report to be verified is <entity to be matched, first entity, 1>, where 1 indicates that the preset terminology library contains an entity that matches the entity to be matched (the first entity), and the first entity is not a standard entity.

[0138] An entity in the preset terminology library whose similarity with the entity to be matched meets a preset similarity threshold is recorded as the second entity. If the second entity is a standard entity, the matching result displayed in the image report to be verified is <entity to be matched, second entity, 2>, where 2 indicates that the preset terminology library contains an entity (second entity) whose similarity with the entity to be matched meets the preset similarity threshold, and the second entity is a standard entity. If the second entity is not a standard entity, the matching result displayed in the image report to be verified is <entity to be matched, second entity, 3>, where 3 indicates that the preset terminology library contains an entity (second entity) whose similarity with the entity to be matched meets the preset similarity threshold, and the second entity is not a standard entity.

[0139] If the similarity between each entity in the preset terminology library and the entity to be matched does not meet the preset similarity threshold, the matching structure displayed in the image to be verified report is <entity to be matched, empty, -1>, where -1 indicates that the similarity between each entity in the preset terminology library and the entity to be matched does not meet the preset similarity threshold.

[0140] In this embodiment, by displaying the specific matching results in the image report to be verified, the matching results can be clearly shown to medical staff, so that the medical staff can clearly obtain the verification results of the image report to be verified.

[0141] In an optional embodiment, if the entity that matches the entity to be matched in the preset term library is not a standard entity, a mark index can be displayed at the position of the entity to be matched in the imaging report. When the mouse is at the mark index, a selection control can be displayed for medical staff to choose whether to replace the entity to be matched with the standard entity, so that medical staff can choose whether to replace the entity to be matched according to actual needs.

[0142] It should be understood that, although the various steps in the flowcharts involved in the various embodiments described above are displayed in sequence according to the instructions of the arrows, these steps are not necessarily executed in sequence in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and these steps can be executed in other orders. Moreover, at least a portion of the steps in the flowcharts involved in the various embodiments described above can include multiple steps or multiple stages, and these steps or stages are not necessarily executed and completed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a portion of steps or stages in other steps.

[0143] Based on the same inventive concept, embodiments of the present application also provide an image report verification device for implementing the aforementioned image report verification method. The solution provided by this device is similar to the solution described in the aforementioned method. Therefore, the specific limitations in one or more medical image processing device embodiments provided below can be found in the limitations of the image report verification method described above and will not be further elaborated here.

[0144] In one embodiment, Figure 9 As shown, an image report verification device 10 is provided, which includes an acquisition module 11, an extraction module 12, a search module 13 and a verification module 14.

[0145] The acquisition module 11 is used to obtain the image report to be verified, which includes the image to be verified and the text to be verified;

[0146] The extraction module 12 is used to extract features from the image to be verified and obtain a first feature map corresponding to the image to be verified;

[0147] The search module 13 is used to search for a second feature map that matches the first feature map in a preset feature map library;

[0148] The search module 13 is further configured to search for a first seen element set matching the second feature map in a preset image seen element library;

[0149] The verification module 14 is configured to verify the image report to be verified based on the first found element set and the second found element set corresponding to the text to be verified.

[0150] In one embodiment, the imaging report verification device 10 further includes a determination module. The determination module is configured to determine a feature map index corresponding to the first feature map based on the features of the first feature map; the search module 13 is further configured to search a preset feature map library for a feature map corresponding to the feature map index, and determine the feature map matching the feature map index as the second feature map.

[0151] In one embodiment, the determination module is further configured to determine a similarity between the first feature map and each feature map corresponding to the feature map index, and determine the feature map with the greatest similarity as the second feature map.

[0152] In one embodiment, the imaging report verification device 10 further includes an entity extraction module configured to extract entities from the document to be verified and obtain a second set of observed elements corresponding to the document to be verified.

[0153] In one embodiment, the verification module 14 is specifically configured to match each visible element in the second visible element set with each visible element in the first visible element set, and mark the second visible elements that are matched successfully and / or failed to match in the image report to be verified.

[0154] In one embodiment, the imaging report verification device 10 further includes a display module. The display module is configured to display a mark corresponding to a redundant visual element in the imaging report to be verified if the first visual element set includes redundant visual elements; the redundant visual elements are visual elements included in the first visual element set but not included in the second visual element set.

[0155] In one embodiment, the imaging report verification device 10 further includes a standardization processing module. The standardization processing module is configured to perform standardization processing on each visual element in the second visual element set to obtain a standard second visual element set; and the verification module 14 is specifically configured to match each visual element in the standard second visual element set with each visual element in the first visual element set.

[0156] In one embodiment, the standardization processing module is specifically used to match the entity to be matched with each entity included in the preset term library; the entity to be matched is the entity corresponding to each element in the second seen element set; if the preset term library contains an entity that matches the entity to be matched, the standard second seen element set is determined based on the entity that matches the entity to be matched; if the preset term library does not contain an entity that matches the entity to be matched, the similarity between the entity to be matched and each entity included in the preset term library is calculated, and the entity in the preset term library whose similarity with the entity to be matched meets the preset similarity threshold is determined as the entity that matches the entity to be matched, and the standard second seen element set is determined based on the entity that matches the entity to be matched.

[0157] In one embodiment, the standardization processing module is also used to determine whether the entity that matches the entity to be matched is a standard entity; if the entity that matches the entity to be matched is not a standard entity, the standard entity corresponding to the entity to be matched is found from the preset term library, and the entity to be matched is replaced with the standard entity corresponding to the entity to be matched, and the replaced entity to be matched is determined as an entity in the standard second seen element set.

[0158] In one embodiment, the determination module is further configured to determine the entity to be matched as a matching missing entity if the similarity between each entity in the preset terminology library and the entity to be matched does not satisfy the preset similarity threshold.

[0159] In one embodiment, the display module is also used to display the matching results between the entity to be matched and the entities in the preset terminology library in the image report to be verified; the matching results include the entity to be matched, the entity corresponding to the entity to be matched in the preset terminology library, and the corresponding relationship between the entity to be matched and the entity to be matched in the preset terminology library.

[0160] Each module in the imaging report verification device can be implemented in whole or in part through software, hardware, or a combination thereof. Each module can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in the computer device in the form of software, so that the processor can call and execute the corresponding operations of each module.

[0161] In one embodiment, a computer device is provided. The computer device may be a terminal, and its internal structure diagram may be as follows: Figure 10 As shown. The computer device includes a processor, a memory, a communication interface, a display screen and an input device connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, a mobile cellular network, NFC (near field communication) or other technologies. When the computer program is executed by the processor, an image report verification method is implemented. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covering the display screen, or a button, trackball or touchpad provided on the computer device housing, or an external keyboard, touchpad or mouse.

[0162] Those skilled in the art will understand that Figure 10The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different component arrangement.

[0163] In one embodiment, a computer device is provided, including a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, the following steps are performed:

[0164] Obtain the image report to be verified, which includes the image to be verified and the text to be verified;

[0165] Perform feature extraction on the image to be verified to obtain a first feature map corresponding to the image to be verified;

[0166] Searching for a second feature map that matches the first feature map in a preset feature map library;

[0167] Searching a preset image feature library to obtain a first feature set that matches the second feature map;

[0168] The image report to be verified is verified based on the first seen feature set and the second seen feature set corresponding to the text to be verified.

[0169] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the following steps are implemented:

[0170] Obtain the image report to be verified, which includes the image to be verified and the text to be verified;

[0171] Perform feature extraction on the image to be verified to obtain a first feature map corresponding to the image to be verified;

[0172] Searching for a second feature map that matches the first feature map in a preset feature map library;

[0173] Searching a preset image feature library to obtain a first feature set that matches the second feature map;

[0174] The image report to be verified is verified based on the first seen feature set and the second seen feature set corresponding to the text to be verified.

[0175] In one embodiment, a computer program product is provided, comprising a computer program, which, when executed by a processor, implements the following steps:

[0176] Obtain the image report to be verified, which includes the image to be verified and the text to be verified;

[0177] Perform feature extraction on the image to be verified to obtain a first feature map corresponding to the image to be verified;

[0178] Searching for a second feature map that matches the first feature map in a preset feature map library;

[0179] Searching a preset image feature library to obtain a first feature set that matches the second feature map;

[0180] The image report to be verified is verified based on the first seen feature set and the second seen feature set corresponding to the text to be verified.

[0181] Those skilled in the art will appreciate that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0182] The technical features of the above embodiments can be combined arbitrarily. To make the description concise, 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.

[0183] The above-described embodiments merely represent several implementation methods of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that a person of ordinary skill in the art may make various modifications and improvements without departing from the spirit of the present invention, all of which fall within the scope of protection of the present invention. Therefore, the scope of protection of the present invention shall be determined by the appended claims.

Claims

1. A method for verifying an image report, characterized in that: include: Obtaining an image report to be verified, wherein the image report to be verified includes a picture to be verified and a text to be verified; Performing feature extraction on the image to be verified to obtain a first feature map corresponding to the image to be verified; Searching for a second feature map that matches the first feature map in a preset feature map library; Searching a preset image visible element library to obtain a first visible element set that matches the second feature map; Performing entity extraction on the text to be verified to obtain a second element set corresponding to the text to be verified; Verifying the image report to be verified based on the first seen element set and the second seen element set corresponding to the text to be verified; The step of performing entity extraction on the text to be verified to obtain a second element set corresponding to the text to be verified includes: Segment and remove duplicates of the text to be verified according to the preset delimiters to obtain a set of text segments; Entity recognition and relationship extraction are performed on each text segment in the text segment set to obtain a second feature set.

2. The image report verification method according to claim 1, characterized in that: The method further comprises: Determining a feature map index corresponding to the first feature map according to features of the first feature map; The searching for a second feature map matching the first feature map in a preset feature map library includes: A feature map corresponding to the feature map index is searched in the preset feature map library, and the feature map matching the feature map index is determined as the second feature map.

3. The image report verification method according to claim 2, characterized in that: If the feature map corresponding to the feature map index includes multiple feature maps, the method further includes: Determine the similarity between the first feature map and each feature map corresponding to the feature map index, and determine the feature map with the greatest similarity as the second feature map.

4. The image report verification method according to claim 1, characterized in that: Verifying the image report to be verified based on the first seen element set and the second seen element set corresponding to the text to be verified includes: Each seen element in the second seen element set is matched with each seen element in the first seen element set, and the second seen elements that are successfully matched and / or unmatched are marked in the image report to be verified.

5. The image report verification method according to claim 4, characterized in that: The method further comprises: If the first seen element set includes redundant seen elements, a mark corresponding to the redundant seen elements is displayed in the image report to be verified; the redundant seen elements are seen elements included in the first seen element set and not included in the second seen element set.

6. The image report verification method according to claim 4, characterized in that: The method further comprises: performing standardization processing on each observed element in the second observed element set to obtain a standard second observed element set; The matching each seen element in the second seen element set with each seen element in the first seen element set includes: Each seen element in the standard second seen element set is matched with each seen element in the first seen element set.

7. The image report verification method according to claim 6, characterized in that: The step of performing standardization on each of the elements in the second set of elements to obtain a standard second set of elements includes: Matching the entity to be matched with each entity included in the preset term library; the entity to be matched is the entity corresponding to each element in the second element set; If the preset term library contains an entity that matches the entity to be matched, determining the standard second found element set according to the entity that matches the entity to be matched; If the preset term library does not contain an entity that matches the entity to be matched, the similarity between the entity to be matched and each entity contained in the preset term library is calculated, and the entity in the preset term library whose similarity with the entity to be matched meets a preset similarity threshold is determined as the entity that matches the entity to be matched, and the standard second seen element set is determined based on the entity that matches the entity to be matched.

8. The image report verification method according to claim 7, characterized in that: The determining of the standard second seen element set according to the entity matching the to-be-matched entity includes: Determining whether the entity matching the entity to be matched is a standard entity; If the entity that matches the entity to be matched is not the standard entity, the standard entity corresponding to the entity to be matched is searched from the preset term library, and the entity to be matched is replaced with the standard entity corresponding to the entity to be matched, and the entity to be matched is determined to be an entity in the standard second seen element set.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 8 are implemented.

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

Images