A pathological image must-check item filtering method and device, electronic equipment and medium
Patent Information
- Application Number
- CN202611147014.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-07-30
- Publication Date
- 2026-10-09
AI Technical Summary
[0005]本发明的目的在于克服上述现有技术的不足，提供一种基于标本物理加工属性的病理图像必查项过滤方法、装置、电子设备及计算机可读存储介质，以解决将标本天然不可评估项误当作遗漏、以及标本天然不可评估与图像质量不可评估不区分导致误判的技术问题
（1）避免将标本天然不可评估项误当作遗漏：通过依标本类型从必查项清单中过滤天然不可评估项，使穿刺、刮宫、细胞学等标本本就无法评估的切缘、包膜等项不进入待核查清单，避免系统因“未评估这些项”而被误判为漏检；
Smart Images
Figure CN122889281A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of digital pathological image processing technology, and in particular to a method, device, electronic device, and medium for filtering mandatory items in pathological images. Background Technology
[0002] In pathological image-assisted processing, to prevent the missed detection of high-risk lesions, a checklist of essential items is usually maintained for each site. Before issuing results, each item on the checklist is checked to ensure it is excluded, present, or unassessable. Different specimens have objective differences in their evaluable content due to different physical processing methods. For example, core needle puncture specimens naturally do not contain intact cutting edges and capsules.
[0003] In the prior art, one existing solution sets a unified list of items to be excluded for each part and checks them one by one; another existing solution uses optical means to extract the tissue features of the cutting edge for risk classification; and another existing solution detects image quality (such as blurring or folding) and indicates that it is not assessable.
[0004] However, the aforementioned prior art has the following technical defects: (1) A uniform list of mandatory items is used for all specimen types without taking into account the physical processing properties of the specimens. This results in items such as the cut margin and membrane, which are not assessable in specimens such as puncture, curettage, and cytology, still being included in the list of items to be checked. (2) When the above-mentioned naturally unevaluable items are not checked, the system counts them as "not evaluated", which may be mistakenly judged as a missed detection or mistakenly regarded as excluded, resulting in a false negative. (3) The two situations of “the physical properties of the specimen are inherently unassessable” and “the image quality is unassessable” were not distinguished. The two situations should be handled differently (the former is an inherent limitation and should not trigger rejection, while the latter should prompt for supplementary materials). Summary of the Invention
[0005] The purpose of this invention is to overcome the shortcomings of the prior art and provide a method, device, electronic device and computer-readable storage medium for filtering mandatory items in pathological images based on the physical processing properties of specimens, so as to solve the technical problems of mistaking naturally unassessable items of specimens as omissions and misjudgment caused by the inability to distinguish between naturally unassessable items of specimens and unassessable items of image quality.
[0006] To achieve the above objectives, the first aspect of this application provides a method for filtering mandatory search items in pathological images, comprising: Obtain the location and specimen type corresponding to the pathological image, wherein the specimen type represents the physical processing method of the specimen and is taken from a preset set of specimen types; Based on the location, obtain the corresponding list of mandatory items to be checked, which includes items to be excluded that need to be checked in the location; For each item in the list of mandatory items, determine whether each item is a naturally unevaluable item of the specimen type, wherein a naturally unevaluable item refers to an item that cannot be evaluated due to the physical processing properties of the specimen type; Items deemed naturally unevaluable are filtered out from the list of mandatory items to obtain a list of mandatory items applicable to the specimen type, and the filtered items are marked as specimen-related limitations. Output the list of mandatory items applicable to the specimen type and the specimen-related limitations.
[0007] Optionally, determining whether each item belongs to the naturally unevaluable items of the specimen type includes: When the description of each item matches a preset range-dependent keyword, and the specimen type belongs to a preset range-restricted specimen type set, it is determined to be a naturally unevaluable item of the specimen type.
[0008] Optionally, the range-dependent keywords include at least one of cutting edge, depth of infiltration, capsule, vascular, range, and stage; The limited set of specimen types includes at least one of core needle biopsy, curettage, and cytology.
[0009] Optionally, the preset set of specimen types includes at least one of core needle biopsy, excisional biopsy, surgical biopsy, radical resection, curettage, cytology, frozen section, and unknown.
[0010] Optionally, for the same site, when the specimen type belongs to a range-limited specimen type set, the list of mandatory items applicable to the specimen type is obtained, and the number of items in the list of mandatory items is no more than the number of items in the list of mandatory items when the specimen type is radical resection.
[0011] Optionally, the specimen-related limitations are presented as limitation declarations in subsequent outputs and are distinguished from identifiers indicating that the image quality is unassessable in the data structure.
[0012] Optionally, the list of mandatory items is obtained by filtering according to the parts after deduplicating and aggregating the union of multiple heterogeneous ontologies through concept clusters.
[0013] A second aspect of this application provides a pathological image mandatory search item filtering device, comprising: The attribute acquisition module is used to acquire the location and specimen type corresponding to the pathological image, wherein the specimen type represents the physical processing method of the specimen and is taken from a preset specimen type set; The list reading module is used to obtain the corresponding list of mandatory items based on the location, and the list of mandatory items includes items to be excluded that need to be checked in the location; The unevaluable determination module is used to determine whether each item in the list of mandatory items belongs to the naturally unevaluable items of the specimen type, wherein the naturally unevaluable items refer to items that cannot be evaluated due to the physical processing properties of the specimen type; The filtering and labeling module is used to filter out items that are determined to be naturally unevaluable from the list of mandatory items, obtain a list of mandatory items applicable to the specimen type, and label the filtered items as specimen-related limitations. The output module is used to output the list of mandatory items applicable to this specimen type and the specimen-related limitations.
[0014] A third aspect of this application provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the aforementioned method for filtering mandatory pathological images.
[0015] A fourth aspect of this application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the above-described method for filtering mandatory pathological images.
[0016] The beneficial effects that this application can achieve are as follows: (1) Avoid mistaking naturally unevaluable items of specimens as omissions: By filtering naturally unevaluable items from the list of mandatory items according to specimen type, items such as the cutting edge and capsule of specimens such as puncture, curettage, and cytology that cannot be evaluated are not included in the list of items to be checked, thus avoiding the system being misjudged as a missed detection due to "these items not being evaluated"; (2) Distinguish between two types of non-evaluable items and prevent false negatives: By distinguishing between specimen-related limitations and image quality non-evaluable items in the data structure, the naturally non-evaluable items of the specimen are marked as limitation statements and not included in the rejection count, rather than being mistakenly regarded as excluded. (3) The list of mandatory items is fully covered: the list of mandatory items is derived from the union of multiple heterogeneous ontologies, deduplicated and aggregated, and then filtered by part, so that the high-risk items to be excluded are more fully covered. Attached Figure Description
[0017] Figure 1 A flowchart illustrating the mandatory item filtering method for pathological images provided in this application embodiment; Figure 2 A schematic diagram illustrating the logic for determining naturally unevaluable items provided in the embodiments of this application; Figure 3 A comparison diagram of the data structures for two types of identifiers—specimen-related limitations and image quality that cannot be assessed—provided for embodiments of this application; Figure 4 This is a schematic diagram of the structure of a pathological image mandatory search item filtering device provided in an embodiment of this application; Figure 5 This is a schematic diagram of the structure of a computer device provided in an embodiment of this application. Detailed Implementation
[0018] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numerals in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present invention. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.
[0019] The technical solution of this application and how the technical solution of this application solves the above-mentioned technical problems are described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of this application will now be described with reference to the accompanying drawings.
[0020] Figure 1 This is a flowchart illustrating a method for filtering mandatory search items in pathological images, provided in an embodiment of this application. Figure 1 As shown, this method for filtering mandatory items in pathological images may include the following steps: S1. Obtain the location and specimen type corresponding to the pathological image, where the specimen type represents the physical processing method of the specimen and is taken from a preset specimen type set.
[0021] Pathological images refer to digital images obtained by digitizing tissue sections using digital pathology scanning equipment. These are typically high-resolution RGB color images containing tissue morphology information and are fundamental data for pathological diagnosis. Location refers to the anatomical location or tissue type corresponding to the pathological image, such as lung, breast, prostate, stomach, or intestine; this is a key index field for the mandatory search list. Specimen type refers to the method of specimen collection and preparation, characterizing the specimen's physical processing attributes, such as core needle biopsy, excisional biopsy, radical resection, curettage, cytology, and frozen section. Different specimen types have objectively different assessable content ranges due to their different physical processing methods. The preset specimen type set refers to a predefined closed set containing all possible specimen types. In practical applications, this method is generally executed by a processor. When acquiring specimen type information, the processor verifies whether it belongs to this set; if not, it is marked as "unknown," ensuring data consistency and controllability.
[0022] Based on this, the processor obtains the de-identified pathological image, the corresponding body part, and the corresponding specimen type. The specimen type represents the physical processing method of the specimen and is taken from a pre-agreed closed set of specimen types, thus providing basic data for the subsequent filtering of the mandatory item list.
[0023] S2. Based on the location information, obtain the list of mandatory items for that location. The list of mandatory items includes the items to be excluded that need to be checked for that location.
[0024] The mandatory checklist refers to a list of items to be excluded that would be costly to miss in a particular site. It is a core data structure maintained in pathological image-assisted processing to prevent the omission of high-risk lesions. Each site corresponds to an independent mandatory checklist, and each item in the list represents a pathological feature or diagnostic indicator that needs to be verified, such as tumor size, depth of invasion, resection margin status, vascular invasion, capsule invasion, lymph node metastasis, and stage. Items to be excluded refer to pathological features or diagnostic conclusions that need to be clearly excluded in pathological diagnosis, such as "exclude invasion," "exclude vascular invasion," and "exclude capsule invasion." Whether or not these items are excluded directly affects the final diagnostic conclusion, therefore, each item needs to be verified. The mandatory input configuration refers to the configuration file or database table storing the mandatory checklist, indexed by site. The processor retrieves the corresponding mandatory checklist from this database based on the site information.
[0025] Based on this, the processor reads the list of mandatory items for a given location from the mandatory input configuration. The list of mandatory items includes items that have a high cost of missing detection for that location, thus providing input data for the subsequent determination of naturally unevaluable items.
[0026] S3. For each item in the mandatory checklist, determine whether it belongs to the naturally unevaluable items of the specimen type. Naturally unevaluable items refer to items that cannot be evaluated due to the physical processing attributes of the specimen type.
[0027] Among them, "naturally unevaluable items" refer to items that cannot be objectively evaluated due to the physical processing properties of the specimen type, rather than those unevaluable due to image quality issues. For example, due to limitations in the collection method, core needle puncture specimens usually cannot obtain complete cutting margin tissue; therefore, the "cutting margin status" item is a naturally unevaluable item for this specimen type. Similarly, due to limitations in the collection method, the depth of myometrial invasion cannot be evaluated in curettage specimens; therefore, the "depth of invasion" item is a naturally unevaluable item for this specimen type. The judgment logic is based on a combination of two conditions: first, whether the description of the item matches the preset scope-dependent keywords; and second, whether the specimen type belongs to the preset scope-restricted specimen type set. Only when both conditions are met simultaneously is it determined to be a naturally unevaluable item.
[0028] Figure 2This is a schematic diagram illustrating the logic for determining naturally unevaluable items provided in embodiments of this application. Figure 2 As shown, the determination of naturally unevaluable terms is based on the cross-determination of range-dependent keywords and range-limited specimen type sets.
[0029] Among them, range-dependent keywords refer to pathological feature keywords related to the physical extent of the specimen, including resection margin, depth of invasion, capsule, vessels, extent, and stage. The pathological features described by these keywords depend on the physical integrity or spatial extent of the specimen. For some specimen types, due to the limitations of their physical processing methods, these features cannot be objectively assessed. The range-limited specimen type set refers to the set of specimen types that cannot fully preserve the spatial extent of tissue due to limitations in physical processing methods, including core needle biopsy, curettage, and cytology. These specimen types usually only obtain local tissue or cells and cannot preserve complete tissue boundaries and spatial relationships, therefore, the pathological features corresponding to range-dependent keywords cannot be assessed.
[0030] Based on this, the processor determines whether each item in the mandatory item list is a naturally unevaluable item of the specimen type: when the description of the item matches the preset scope-dependent keywords and the specimen type belongs to the preset scope-restricted specimen type set, it is determined to be a naturally unevaluable item of the specimen type, thereby providing a basis for subsequent filtering operations.
[0031] S4. Filter out items that are determined to be naturally unevaluable from the list of mandatory items to obtain a list of mandatory items applicable to this specimen type, and mark the filtered items as specimen-related limitations.
[0032] The filtering operation involves removing items deemed inherently unevaluable from the original mandatory item list, resulting in a streamlined mandatory item list suitable for the current specimen type. This list only includes items that can be objectively evaluated for that specimen type, avoiding misjudgments caused by including inherently unevaluable items in the checklist. Specimen-related limitations refer to the filtered inherently unevaluable items, which are declared as limitations in subsequent outputs. These limitations indicate that these items cannot be evaluated due to the physical properties of the specimen and are inherent, thus they should not trigger rejection or be misjudged as missed detections. The annotation operation marks the filtered items as specimen-related limitations, distinguishing them from those marked as unevaluable due to image quality in the data structure. This facilitates subsequent processing logic in adopting different processing strategies based on different unevaluable types.
[0033] Figure 3 A comparison diagram of the data structures for two types of identifiers—specimen-related limitations and image quality that cannot be assessed—provided for embodiments of the present invention. For example... Figure 3 As shown, the two types of non-evaluable identifiers are listed as different fields in the data structure.
[0034] Based on this, the processor filters out items that are determined to be naturally unevaluable from the list of mandatory items, and obtains a list of mandatory items applicable to this type of specimen. The filtered items are marked as specimen-related limitations rather than omissions, and are not included in the rejection conditions caused by image quality or insufficient evidence, thereby achieving the correct identification and processing of naturally unevaluable items of the specimen.
[0035] S5. Output a list of mandatory items applicable to this specimen type and specimen-related limitations.
[0036] The output operation involves transmitting a list of mandatory items applicable to the current specimen type and specimen-related limitations to the downstream processing module for subsequent item-by-item verification. The output data structure consists of two parts: a simplified list of mandatory items, containing only items that can be objectively assessed for this specimen type; and a list of specimen-related limitations, including filtered-out naturally unassessable items and their reasons. The downstream processing module performs item-by-item verification based on the output data, determining the exclusion, presence, or unassessability of each item in the simplified list of mandatory items. Items in the specimen-related limitations list are not subject to verification and are directly output as limitation statements.
[0037] Based on this, the processor outputs the list of mandatory items applicable to this specimen type and the specimen-related limitations. The list of mandatory items applicable to this specimen type is used for subsequent item-by-item verification, thereby completing the mandatory item list filtering process based on the physical processing attributes of the specimen.
[0038] Based on the feasible implementation of S3 described above, this application further provides range-dependent keywords including at least one of cutting edge, depth of invasion, capsule, vascular, range, and stage; and a range-limited specimen type set including at least one of core needle biopsy, curettage, and cytology.
[0039] Among these, the resection margin refers to the edge tissue of the surgically removed specimen, used to assess whether the tumor has been completely removed, and is an important indicator for tumor pathological diagnosis. However, for core needle biopsy specimens, since only local tissue is obtained, the completeness of the resection margin cannot be assessed. Depth of invasion refers to the depth to which the tumor infiltrates surrounding tissues, and is an important basis for tumor staging. However, for curettage specimens, since only endometrial tissue is obtained, the depth of myometrial invasion cannot be assessed. The capsule refers to the fibrous capsule surrounding certain organs or tumors. Capsule invasion is an important characteristic of malignant tumors, but for cytological specimens, since only exfoliated cells are obtained, the capsule status cannot be assessed. Vascular involvement refers to blood vessels and lymphatic vessels. Vascular invasion is an important pathway for tumor metastasis, but for core needle biopsy specimens, due to the limited sampling range, vascular structures may not be observable. The extent refers to the distribution of the tumor within the tissue, while staging refers to the tumor stage determined based on factors such as tumor size, depth of invasion, lymph node metastasis, and distant metastasis. Both of these indicators depend on the integrity and spatial extent of the specimen, and are difficult to accurately assess for specimens with limited extent.
[0040] Based on the feasible implementation of S1 described above, this application further provides a preset set of specimen types including at least one of core needle biopsy, excisional biopsy, surgical biopsy, radical resection, curettage, cytology, frozen section, and unknown.
[0041] Among these methods, core needle biopsy is a biopsy method that uses a core needle to obtain tissue samples. It is commonly used for diagnosing tumors in organs such as the breast and prostate. It involves a small sample size and limited spatial area. Excisional biopsy involves removing a small piece of tissue from the lesion site for pathological examination. It involves a moderate sample size and preserves some tissue boundary information. Resective biopsy involves removing the entire lesion for pathological examination. It involves a larger sample size and preserves complete tissue boundaries and spatial relationships. Radical resection is a surgical procedure that removes not only the lesion but also surrounding normal tissue and lymph nodes. It yields the most complete specimen and allows for the assessment of all pathological features. Curettage involves scraping endometrial tissue with a curette. It only obtains endometrial tissue and cannot assess myometrial invasion. Cytology involves collecting exfoliated cells for smear examination. It only obtains cells and cannot assess tissue structure. Frozen section is a method of rapidly freezing tissue sections during surgery for pathological examination. The special specimen processing method may affect the assessment of certain pathological features. "Unknown" refers to a situation where the index type cannot be determined or is not in the preset set. The system marks it as "unknown" and performs the most conservative processing method.
[0042] Based on the feasible implementation of S4 described above, this application further provides a list of mandatory items applicable to the same specimen type when the specimen type belongs to a range-limited specimen type set. The number of items in the list is no more than the number of mandatory items in the list of the same specimen type when the specimen type is radical resection.
[0043] This feature highlights the rationale behind the invention: radical resection specimens are the most complete and can assess all pathological features, therefore their mandatory checklist includes all items to be excluded; while limited-range specimen types, due to limitations in physical processing methods, cannot assess certain pathological features, thus their mandatory checklist is necessarily shorter than that of radical resection specimens. This feature can also be used for system self-checks: if the number of items in the mandatory checklist for a limited-range specimen type is greater than that for a radical resection specimen, it indicates a problem with the filtering logic, requiring manual review.
[0044] Based on the feasible implementation of S4 described above, this application further provides that specimen-related limitations are declared as limitation items in subsequent outputs, and are distinguished from the identifiers due to unassessable image quality in the data structure.
[0045] The limitation declaration item refers to pathological features declared in the final output report that cannot be assessed due to limitations in the physical properties of the specimen. Its purpose is to inform the user that the inability to assess these items is due to objective reasons, rather than system omissions or image quality issues. The data structure distinguishes between specimen-related limitations and image quality unassessable features using different fields or identifiers. For example, specimen-related limitations use the "specimen_blind" field, while image quality unassessable features use the "quality_unassessable" field. Downstream processing logic executes different processing strategies based on these fields: for specimen-related limitations, they are directly output as limitation declarations and are not included in the rejection criteria; for image quality unassessable features, the user is prompted to supplement materials or rescan, and the feature is included in the rejection criteria.
[0046] Based on the feasible implementation of S2 described above, this application further provides that the source of the mandatory item list is obtained by filtering according to the location after deduplicating and aggregating the union of multiple heterogeneous ontologies through concept clusters.
[0047] Heterogeneous ontologies refer to mandatory data from different data sources or knowledge systems, such as clinical guidelines, expert consensus, literature reports, and historical cases. These data sources may differ in data structure, terminology, and coverage. Concept cluster deduplication and aggregation involves clustering mandatory items from heterogeneous ontologies into concept clusters based on semantic similarity. Each concept cluster represents a pathological feature, and the most representative item is selected from each cluster as the aggregation result, thus eliminating duplicates and similar items. Site-based filtering involves selecting mandatory items for specific sites from the aggregation results, forming a list of mandatory items for that site. This method creates a more comprehensive list of mandatory items, avoiding the problem of incomplete coverage from a single data source.
[0048] Figure 4This is a schematic diagram of a pathological image mandatory search item filtering device provided in an embodiment of the present invention. Figure 4 As shown, this mandatory screening filter for pathological images includes: The attribute acquisition module 81 is used to acquire the location and specimen type corresponding to the pathological image, wherein the specimen type represents the physical processing method of the specimen and is taken from a preset specimen type set.
[0049] The list reading module 82 is used to obtain the list of mandatory items corresponding to a location based on the location. The list of mandatory items includes the items to be excluded that need to be checked in that location.
[0050] The Unevaluable Decision Module 83 is used to determine whether each item in the mandatory item list belongs to the naturally unevaluable items of the specimen type. Naturally unevaluable items refer to items that cannot be evaluated due to the physical processing attributes of the specimen type.
[0051] The filtering and labeling module 84 is used to filter out items that are determined to be naturally unevaluable from the list of mandatory items, to obtain a list of mandatory items applicable to this specimen type, and to label the filtered items as specimen-related limitations.
[0052] Output module 85 is used to output a list of mandatory items applicable to this specimen type and specimen-related limitations.
[0053] Figure 5 This is a schematic diagram of the structure of a computer device provided in an embodiment of the present invention. Figure 5 As shown, the device may include a processor with one or more processing cores, a memory with one or more computer-readable storage media, a communication component, and other components. The processor, memory, and communication component are connected via a bus.
[0054] In the specific implementation process, at least one processor executes computer execution instructions stored in memory, causing at least one processor to perform the above-described method for filtering mandatory items in pathological images. The specific implementation process of the processor can be found in the above-described method embodiments, and its implementation principle and technical effects are similar; therefore, it will not be repeated here.
[0055] Furthermore, the processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), etc. A general-purpose processor can be a microprocessor or any conventional processor. The steps of the method disclosed in this invention can be directly manifested as execution by a hardware processor, or execution by a combination of hardware and software modules within the processor.
[0056] The memory may include Random Access Memory (RAM) and non-volatile memory (NVM), such as at least one disk storage device. The bus may be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. Buses can be categorized as address buses, data buses, control buses, etc. For ease of illustration, the buses shown in the accompanying drawings are not limited to a single bus or a single type of bus.
[0057] In some embodiments, a computer program product is also provided, including a computer program or instructions that, when executed by a processor, implement the steps in any of the above-described methods for filtering mandatory pathological images.
[0058] For details on the implementation of each of the above operations, please refer to the previous examples, which will not be repeated here.
[0059] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be performed by instructions, or by instructions controlling related hardware. These instructions can be stored in a computer-readable storage medium and loaded and executed by a processor.
[0060] To this end, embodiments of the present invention provide a computer-readable storage medium storing a plurality of program codes, which can be loaded by a processor to execute the steps in any of the pathological image mandatory search item filtering methods provided in embodiments of the present invention.
[0061] The storage medium may include: read-only memory (ROM), random access memory (RAM), disk or optical disk, etc.
[0062] According to one aspect of the present invention, a computer program product or computer program is provided, the computer program product or computer program including computer instructions stored in a computer-readable storage medium.
[0063] Since the instructions stored in the storage medium can execute the steps in any of the pathological image mandatory search item filtering methods provided in the embodiments of the present invention, the beneficial effects that any of the pathological image mandatory search item filtering methods provided in the embodiments of the present invention can achieve can be realized, as detailed in the preceding embodiments, and will not be repeated here.
[0064] Other embodiments of the invention will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This invention is intended to cover any variations, uses, or adaptations of the invention that follow the general principles of the invention and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope of the invention is indicated by the appended claims.
[0065] It should be understood that the present invention is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope.
[0066] The above are merely preferred embodiments of this application and do not limit the patent scope of this application. Any equivalent structural or procedural transformations made using the content of this application's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of this application.
Claims
1. A method for filtering mandatory search items in pathological images, characterized in that, include: Obtain the location and specimen type corresponding to the pathological image, wherein the specimen type represents the physical processing method of the specimen and is taken from a preset set of specimen types; Based on the location, obtain the corresponding list of mandatory items to be checked, which includes items to be excluded that need to be checked in the location; For each item in the list of mandatory items, determine whether each item is a naturally unevaluable item of the specimen type, wherein a naturally unevaluable item refers to an item that cannot be evaluated due to the physical processing properties of the specimen type; Items deemed naturally unevaluable are filtered out from the list of mandatory items to obtain a list of mandatory items applicable to the specimen type, and the filtered items are marked as specimen-related limitations. Output the list of mandatory items applicable to the specimen type and the specimen-related limitations.
2. The method for filtering mandatory items in pathological images according to claim 1, characterized in that, The determination of whether each item belongs to the naturally unevaluable items of the specimen type includes: When the description of each item matches a preset range-dependent keyword, and the specimen type belongs to a preset range-restricted specimen type set, it is determined to be a naturally unevaluable item of the specimen type.
3. The method for filtering mandatory items in pathological images according to claim 2, characterized in that, The range-dependent keywords include at least one of cutting edge, depth of infiltration, capsule, vascular, range, and stage; The limited set of specimen types includes at least one of core needle biopsy, curettage, and cytology.
4. The method for filtering mandatory items in pathological images according to claim 1, characterized in that, The preset set of specimen types includes at least one of the following: core needle biopsy, excisional biopsy, surgical biopsy, radical resection, curettage, cytology, frozen section, and unknown.
5. The method for filtering mandatory items in pathological images according to claim 1, characterized in that, For the same site, when the specimen type belongs to the range-limited specimen type set, the list of mandatory items applicable to the specimen type is obtained, and the number of items in the list of mandatory items is no more than the number of items in the list of mandatory items when the specimen type is radical resection.
6. The method for filtering mandatory items in pathological images according to claim 1, characterized in that, The specimen-related limitations are presented as limitation declarations in subsequent outputs and are distinguished from the identifiers indicating that the image quality is unassessable in the data structure.
7. The method for filtering mandatory items in pathological images according to claim 1, characterized in that, The list of mandatory items is obtained by filtering the union of multiple heterogeneous ontologies after deduplication and aggregation of concept clusters, according to the specified location.
8. A pathological image mandatory search item filtering device, characterized in that, include: The attribute acquisition module is used to acquire the location and specimen type corresponding to the pathological image, wherein the specimen type represents the physical processing method of the specimen and is taken from a preset specimen type set; The list reading module is used to obtain the corresponding list of mandatory items based on the location, and the list of mandatory items includes items to be excluded that need to be checked in the location; The unevaluable determination module is used to determine whether each item in the list of mandatory items belongs to the naturally unevaluable items of the specimen type, wherein the naturally unevaluable items refer to items that cannot be evaluated due to the physical processing properties of the specimen type; The filtering and labeling module is used to filter out items that are determined to be naturally unevaluable from the list of mandatory items, obtain a list of mandatory items applicable to the specimen type, and label the filtered items as specimen-related limitations. The output module is used to output the list of mandatory items applicable to this specimen type and the specimen-related limitations.
9. A computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the pathological image mandatory search item filtering method according to any one of claims 1-7.
10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the pathological image mandatory search item filtering method according to any one of claims 1-7.