Physical examination result analysis method and system based on outpatient cases and storage medium
By automatically analyzing physical examination results through word segmentation, entity recognition and semantic matching of outpatient medical records, the problem of inefficient analysis of physical examination results is solved, the analysis efficiency is improved and manual errors are reduced.
Patent Information
- Application Number
- CN202411052141.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-01
- Publication Date
- 2024-11-15
Smart Images

Figure CN118969160A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of smart medical technology, and in particular to a method, system and storage medium for analyzing physical examination results based on outpatient cases. Background Art
[0002] With the popularization of medical electronic information, electronic medical record information has become an essential way for major hospitals to record medical information. In the existing electronic medical record information, medical workers need to fill in the patient's physical examination results in the medical record. In order to improve the accuracy of medical record data, it is necessary to analyze the physical examination results to improve the accuracy of the physical examination results.
[0003] In the existing physical examination result analysis process, the physical examination results are generally analyzed manually, resulting in low efficiency in the physical examination result analysis. Summary of the invention
[0004] The purpose of the embodiments of the present invention is to provide a method, system and storage medium for analyzing physical examination results based on outpatient cases, so as to solve the problem of low efficiency of physical examination result analysis in the prior art.
[0005] The embodiment of the present invention is implemented as follows: a method for analyzing physical examination results based on outpatient cases, the method comprising:
[0006] Obtaining the text of the history of current illness in the outpatient medical record, and segmenting the text of the history of current illness to obtain the segmentation of the history of current illness;
[0007] Performing entity recognition on the present medical history segmentation, and determining the entity type of the present medical history segmentation according to the entity recognition result;
[0008] Determine the entity object of the present medical history segmentation word according to the entity type, and classify the present medical history segmentation word according to the entity object to obtain an object set;
[0009] Matching the physical examination items in the physical examination result with the entity object to obtain a part matching result, and determining an associated set of the physical examination items in the object set according to the part matching result;
[0010] The examination conclusion of the physical examination item is semantically matched with the corresponding association set to obtain a semantic matching result, and a result analysis report is generated according to the semantic matching result.
[0011] Preferably, performing part matching between the physical examination items in the physical examination result and the physical object to obtain a part matching result includes:
[0012] Obtaining the examination identification of the physical examination item, and determining the examination site according to the examination identification;
[0013] A sub-part of the inspection part is obtained, and the inspection part and the sub-part are matched with the entity object respectively to obtain the part matching result.
[0014] Preferably, determining, according to the part matching result, an associated set of the physical examination items in the object set includes:
[0015] If any of the examination parts or the sub-parts matches the entity object, determining the object set corresponding to the entity object as the associated set of the physical examination items corresponding to the examination part or the sub-part;
[0016] If the inspection part and the corresponding sub-part do not match the physical object, the physical examination item corresponding to the inspection part is marked as an abnormal inspection item, and an analysis abnormality prompt is sent for the abnormal inspection item.
[0017] Preferably, semantic matching is performed between the examination conclusion of the physical examination item and the corresponding association set to obtain a semantic matching result, including:
[0018] Acquire the text position of the entity object in the association set, and determine the medical history description of the entity object according to the text position;
[0019] Semantically identifying the examination conclusion and the medical history description respectively to obtain conclusion semantics and medical history semantics;
[0020] The conclusion semantics are respectively combined with the medical history semantics in the corresponding association set to obtain a semantic combination, and the semantic combination is matched with the semantic list to obtain the semantic matching result.
[0021] Preferably, generating a result analysis report based on the semantic matching result includes:
[0022] If any of the semantic combinations matches the semantic list, it is determined that there is a semantic conflict between the conclusion semantics in the semantic combination and the medical history semantics, and the examination conclusion corresponding to the conclusion semantics of the semantic conflict is determined as an abnormal conclusion;
[0023] The physical examination item corresponding to the abnormal conclusion is determined as an abnormal examination item, and the result analysis report is generated according to the abnormal examination item and the abnormal conclusion.
[0024] Preferably, entity recognition is performed on the present medical history segmentation, and the entity type of the present medical history segmentation is determined according to the entity recognition result, including:
[0025] Performing phoneme conversion on the present medical history segmentation to obtain present medical history phonemes, and performing phoneme encoding on the present medical history phonemes to obtain phoneme features;
[0026] Performing vector conversion on the phoneme feature to obtain a phoneme vector, and calculating the similarity between the phoneme vector and a preset entity vector to obtain a vector similarity;
[0027] The type corresponding to the maximum vector similarity is determined as the entity type of the current medical history segmentation corresponding to the phoneme vector.
[0028] Preferably, entity recognition is performed on the present medical history segmentation, and the entity type of the present medical history segmentation is determined according to the entity recognition result, including:
[0029] Calculate the hash value of the current medical history segmentation to obtain the current medical history hash value, and map the current medical history hash value to a preset hash ring to obtain a hash position;
[0030] A target hash bucket is determined according to the hash position and the ring direction of the preset hash ring, and a type corresponding to the target hash bucket is determined as the entity type of the present medical history segmentation.
[0031] Another object of an embodiment of the present invention is to provide a physical examination result analysis system based on outpatient cases, the system comprising:
[0032] A text segmentation module is used to obtain the current medical history text in the outpatient medical record and segment the current medical history text to obtain the current medical history segmentation;
[0033] An entity recognition module, used for performing entity recognition on the present medical history segmentation, and determining the entity type of the present medical history segmentation according to the entity recognition result;
[0034] An object classification module, used for determining the entity object of the present medical history segmentation word according to the entity type, and classifying the present medical history segmentation word according to the entity object to obtain an object set;
[0035] A set determination module, used for matching the physical examination items in the physical examination results with the entity objects to obtain a part matching result, and determining an associated set of the physical examination items in the object set according to the part matching result;
[0036] The report generation module is used to perform semantic matching on the examination conclusion of the physical examination item and the corresponding association set to obtain a semantic matching result, and generate a result analysis report according to the semantic matching result.
[0037] Preferably, the set determination module is further used for:
[0038] Obtaining the examination identification of the physical examination item, and determining the examination site according to the examination identification;
[0039] A sub-part of the inspection part is obtained, and the inspection part and the sub-part are matched with the entity object respectively to obtain the part matching result.
[0040] The embodiments of the present invention can effectively obtain the current medical history segmentation by performing segmentation on the current medical history text, effectively determine the entity type of the current medical history segmentation by performing entity recognition on the current medical history segmentation, effectively determine the entity object of the current medical history segmentation based on the entity type, effectively obtain the object set by classifying the current medical history segmentation through the entity object, effectively obtain the part matching result by matching the physical examination items with the entity objects, effectively determine the associated set of the physical examination items in the object set based on the part matching result, effectively detect whether there is a semantic conflict between the examination conclusion and the current medical history text by performing semantic matching on the examination conclusion of the physical examination item with the corresponding associated set, so as to achieve the effect of automatically analyzing whether there is an error in the physical examination result, thereby improving the analysis efficiency of the physical examination result. BRIEF DESCRIPTION OF THE DRAWINGS
[0041] Figure 1 is a flow chart of a method for analyzing physical examination results based on outpatient cases provided by a first embodiment of the present invention;
[0042] Figure 2 is a structural schematic diagram of a physical examination result analysis system based on outpatient cases provided by a second embodiment of the present invention;
[0043] Figure 3 It is a schematic diagram of the structure of a terminal device provided in the third embodiment of the present invention. DETAILED DESCRIPTION
[0044] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.
[0045] In order to illustrate the technical solution of the present invention, a specific embodiment is provided below for illustration.
[0046] Embodiment 1
[0047] See also Figure 1 , is a flow chart of a method for analyzing physical examination results based on outpatient cases provided by a first embodiment of the present invention. The method for analyzing physical examination results based on outpatient cases can be applied to any device or system. The method for analyzing physical examination results based on outpatient cases comprises the steps of:
[0048] Step S10, obtaining the current medical history text in the outpatient medical record, and segmenting the current medical history text to obtain the current medical history segmentation;
[0049] The current medical history text is segmented according to a preset segmentation interval to obtain the current medical history segmentations, and the preset segmentation interval can be set according to demand.
[0050] Step S20, performing entity recognition on the present medical history segmentation, and determining the entity type of the present medical history segmentation according to the entity recognition result;
[0051] Among them, by performing entity recognition on the current medical history segmentation, the entity type of the current medical history segmentation can be effectively determined. Optionally, in this step, performing entity recognition on the current medical history segmentation, and determining the entity type of the current medical history segmentation according to the entity recognition result, includes:
[0052] The present medical history segmentation is converted into phonemes to obtain present medical history phonemes, and the present medical history phonemes are encoded into phonemes to obtain phoneme features; wherein the present medical history segmentation is converted into phonemes to obtain present medical history phonemes, the present medical history phonemes are segmented, the segmented present medical history phonemes are subjected to phoneme embedding processing, and the present medical history phonemes subjected to the phoneme embedding processing are input into a phoneme encoder for encoding to obtain phoneme features;
[0053] The phoneme feature is vector-converted to obtain a phoneme vector, and the similarity between the phoneme vector and a preset entity vector is calculated to obtain vector similarity; wherein the phoneme feature is input into a vector encoder for vector conversion to obtain a phoneme vector, and the distance between the phoneme vector and a preset entity vector is calculated according to a Euclidean distance formula to obtain vector similarity, and the preset entity vector can be set as required;
[0054] The type corresponding to the maximum vector similarity is determined as the entity type of the phoneme vector corresponding to the current medical history participle; among them, the type corresponding to the maximum vector similarity is closest to the type of the phoneme vector corresponding to the current medical history participle, so the type corresponding to the maximum vector similarity is determined as the entity type corresponding to the current medical history participle.
[0055] Further, entity recognition is performed on the present medical history segmentation, and the entity type of the present medical history segmentation is determined according to the entity recognition result, including:
[0056] Calculate the hash value of the present medical history segmentation to obtain the present medical history hash value, and map the present medical history hash value to a preset hash ring to obtain a hash position; wherein the present medical history hash value is matched with a hash mapping table to obtain a hash position, a hash bucket is stored on the preset hash ring, and the hash bucket is used to characterize the corresponding entity type, and the hash mapping table stores the corresponding relationship between different present medical history hash values and corresponding hash positions;
[0057] The target hash bucket is determined according to the hash position and the ring direction of the preset hash ring, and the type corresponding to the target hash bucket is determined as the entity type of the current medical history segmentation; wherein the ring direction of the preset hash ring can be set according to demand, the first hash bucket in the ring direction is determined as the target hash bucket, and the bucket information of the target hash bucket is obtained, and the type stored in the bucket information is determined as the entity type of the current medical history segmentation.
[0058] Step S30, determining the entity object of the present medical history segmentation word according to the entity type, and classifying the present medical history segmentation word according to the entity object to obtain an object set;
[0059] Among them, the entity type is matched with the object query table to obtain the entity object, and the object query table stores the correspondence between different entity types and corresponding entity objects. In this step, the object identifier of the entity object is obtained, the object identifier is matched with the collection list to obtain the collection address, and the entity object is stored in the corresponding collection address to obtain an object set. The collection list stores the correspondence between different object identifiers and corresponding collection addresses, and multiple different object identifiers can correspond to the same collection address.
[0060] Step S40, matching the physical examination items in the physical examination result with the entity object to obtain a part matching result, and determining an associated set of the physical examination items in the object set according to the part matching result;
[0061] Among them, by matching the physical examination items with the entity objects, the effect of part alignment between the physical examination items and the entity objects is achieved. Based on the part matching results, the associated set of the physical examination items in the object set can be effectively determined, and the part descriptions between the physical examination items and the corresponding associated sets are the same or similar.
[0062] Optionally, performing part matching on the physical examination items in the physical examination result and the entity object to obtain a part matching result includes:
[0063] Obtaining the examination identifier of the physical examination item, and determining the examination part according to the examination identifier; wherein the examination identifier is matched with a part query table to obtain the examination part, and the part query table stores the correspondence between different examination identifiers and corresponding examination parts;
[0064] Obtain a sub-part of the inspection part, and match the inspection part and the sub-part with the physical object respectively to obtain the part matching result; wherein, by matching the inspection part and the sub-part with the physical object respectively, it is determined whether the parts between the physical examination item and the physical object are the same or similar. For example, the inspection part may include the head, hand or leg, etc. When the inspection part is the hand, the sub-part may include the arm and finger, etc.
[0065] Further, determining, according to the part matching result, an associated set of the physical examination items in the object set, includes:
[0066] If any of the inspection parts or sub-parts matches the entity object, the object set corresponding to the entity object is determined as the associated set of the physical examination items corresponding to the inspection part or sub-part; wherein, when any of the inspection parts or sub-parts matches the entity object, it is determined that the parts between the physical examination items and the entity object are the same or similar, and therefore, the object set corresponding to the entity object is determined as the associated set of the corresponding physical examination items;
[0067] If the inspection part and the corresponding sub-part do not match the physical object, the physical examination item corresponding to the inspection part is marked as an abnormal inspection item, and an analysis abnormality prompt is sent for the abnormal inspection item.
[0068] Step S50, semantically matching the examination conclusion of the physical examination item with the corresponding association set to obtain a semantic matching result, and generating a result analysis report according to the semantic matching result;
[0069] Among them, by semantically matching the examination conclusion of the physical examination item with the corresponding association set, it is possible to effectively detect whether there is a semantic conflict between the examination conclusion and the current medical history text, so as to achieve the effect of automatically analyzing whether there are errors in the physical examination results, thereby improving the efficiency of physical examination result analysis. In this embodiment, the physical examination results can be analyzed based on the description of the current medical history to see whether there is a contradiction between the patient's symptoms mentioned in the current medical history, so as to automatically determine whether the physical examination results are compliant, thereby helping hospital doctors improve the quality of medical records.
[0070] Optionally, semantic matching is performed on the examination conclusion of the physical examination item and the corresponding association set to obtain a semantic matching result, including:
[0071] Acquire the text position of the entity object in the association set, and determine the medical history description of the entity object according to the text position;
[0072] Semantically identifying the examination conclusion and the medical history description respectively to obtain conclusion semantics and medical history semantics;
[0073] Respectively combining the conclusion semantics with the medical history semantics in the corresponding association set to obtain a semantic combination, and matching the semantic combination with the semantic list to obtain the semantic matching result;
[0074] Among them, characters are acquired with the text position as the center and the preset number of characters or the preset area distance as the radius to obtain the medical history description. By performing semantic recognition on the examination conclusion and the medical history description respectively, the semantic information corresponding to the examination conclusion and the medical history description can be effectively obtained. By matching the semantic combination with the semantic list, it is determined whether there is a semantic conflict between the conclusion semantics and the medical history semantics in the semantic combination. The semantic list stores pre-set semantic combinations with semantic conflicts.
[0075] For example, when the medical history semantics corresponding to the current medical history text is "coma, unclear consciousness, consciousness disorder, blurred consciousness", and the conclusion semantics in the physical examination results are "clear consciousness, mentally sound, cooperative in physical examination, self-position, normal gait, and walking into the ward", then it is determined that the semantic combination matches the semantic list;
[0076] When the semantic meaning of the medical history corresponding to the current medical history text is "paralysis", and the semantic meaning of the conclusion in the physical examination result is "cooperation in physical examination, self-position, normal gait, and entering the ward", it is determined that the semantic combination matches the semantic list;
[0077] When the medical history semantics corresponding to the current medical history text is "crooked mouth corners", and the conclusion semantics in the physical examination results is "normal face, normal face and expression", then it is determined that the semantic combination matches the semantic list;
[0078] When the medical history semantics corresponding to the current medical history text is "ecchymosis and petechiae", and the conclusion semantics in the physical examination results is "no ecchymosis and petechiae, no subcutaneous bleeding", it is determined that the semantic combination matches the semantic list;
[0079] When the medical history semantics corresponding to the current medical history text is that hemoglobin is lower than 60g / L, and the conclusion semantics in the physical examination results is "no anemia", it is determined that the semantic combination matches the semantic list.
[0080] When the medical history semantics corresponding to the current medical history text is palpable lymphadenopathy, and the conclusion semantics in the physical examination results is "no palpable lymphadenopathy throughout the body", it is determined that the semantic combination matches the semantic list.
[0081] Furthermore, a result analysis report is generated according to the semantic matching result, including:
[0082] If any of the semantic combinations matches the semantic list, it is determined that there is a semantic conflict between the conclusion semantics in the semantic combination and the medical history semantics, and the examination conclusion corresponding to the conclusion semantics of the semantic conflict is determined as an abnormal conclusion;
[0083] The physical examination item corresponding to the abnormal conclusion is determined as an abnormal examination item, and the result analysis report is generated according to the abnormal examination item and the abnormal conclusion; wherein, by determining the examination conclusion corresponding to the semantics of the conclusion with semantic conflict as an abnormal conclusion, it is effectively convenient for users to view the abnormal examination conclusions in the physical examination results.
[0084] In this embodiment, by performing word segmentation on the current medical history text, the current medical history segmentation can be effectively obtained, by performing entity recognition on the current medical history segmentation, the entity type of the current medical history segmentation can be effectively determined, based on the entity type, the entity object of the current medical history segmentation can be effectively determined, by classifying the current medical history segmentation through the entity object, the object set can be effectively obtained, by matching the physical examination items with the entity objects, the part matching results can be effectively obtained, based on the part matching results, the associated set of the physical examination items in the object set can be effectively determined, by performing semantic matching of the examination conclusions of the physical examination items with the corresponding associated sets, it can be effectively detected whether there is a semantic conflict between the examination conclusions and the current medical history text, so as to achieve the effect of automatically analyzing whether there are errors in the physical examination results, thereby improving the efficiency of analyzing the physical examination results.
[0085] Embodiment 2
[0086] See also Figure 2 , is a schematic diagram of the structure of a physical examination result analysis system 100 based on outpatient cases provided in a second embodiment of the present invention, comprising:
[0087] The text segmentation module 10 is used to obtain the current medical history text in the outpatient medical record, and segment the current medical history text to obtain the current medical history segmentation.
[0088] The entity recognition module 11 is used to perform entity recognition on the present medical history segmentation and determine the entity type of the present medical history segmentation according to the entity recognition result.
[0089] Optionally, the entity recognition module 11 is further used to: perform phoneme conversion on the present medical history segmentation to obtain present medical history phonemes, and perform phoneme encoding on the present medical history phonemes to obtain phoneme features;
[0090] Performing vector conversion on the phoneme feature to obtain a phoneme vector, and calculating the similarity between the phoneme vector and a preset entity vector to obtain a vector similarity;
[0091] The type corresponding to the maximum vector similarity is determined as the entity type of the current medical history segmentation corresponding to the phoneme vector.
[0092] Furthermore, the entity recognition module 11 is also used to: calculate the hash value of the current medical history segmentation to obtain the current medical history hash value, and map the current medical history hash value to a preset hash ring to obtain a hash position;
[0093] A target hash bucket is determined according to the hash position and the ring direction of the preset hash ring, and a type corresponding to the target hash bucket is determined as the entity type of the present medical history segmentation.
[0094] The object classification module 12 is used to determine the entity object of the present medical history segmentation according to the entity type, and classify the present medical history segmentation according to the entity object to obtain an object set.
[0095] The set determination module 13 is used to match the physical examination items in the physical examination results with the entity objects to obtain a part matching result, and determine the associated set of the physical examination items in the object set according to the part matching result.
[0096] Optionally, the set determination module 13 is further used to: obtain the examination identifier of the physical examination item, and determine the examination site according to the examination identifier;
[0097] A sub-part of the inspection part is obtained, and the inspection part and the sub-part are matched with the entity object respectively to obtain the part matching result.
[0098] Further, the set determination module 13 is also used for: if any of the examination parts or the sub-parts matches the entity object, determining the object set corresponding to the entity object as the associated set of the examination parts or the sub-parts corresponding to the physical examination items;
[0099] If the inspection part and the corresponding sub-part do not match the physical object, the physical examination item corresponding to the inspection part is marked as an abnormal inspection item, and an analysis abnormality prompt is sent for the abnormal inspection item.
[0100] The report generating module 14 is used to perform semantic matching on the examination conclusion of the physical examination item and the corresponding association set to obtain a semantic matching result, and generate a result analysis report according to the semantic matching result.
[0101] Optionally, the report generating module 14 is further used to: obtain a text position of the entity object in the association set, and determine a medical history description of the entity object according to the text position;
[0102] Semantically identifying the examination conclusion and the medical history description respectively to obtain conclusion semantics and medical history semantics;
[0103] The conclusion semantics are respectively combined with the medical history semantics in the corresponding association set to obtain a semantic combination, and the semantic combination is matched with the semantic list to obtain the semantic matching result.
[0104] Further, the report generation module 14 is also used for: if any of the semantic combinations matches the semantic list, determining the semantic conflict between the conclusion semantics in the semantic combination and the medical history semantics, and determining the examination conclusion corresponding to the conclusion semantics of the semantic conflict as an abnormal conclusion;
[0105] The physical examination item corresponding to the abnormal conclusion is determined as an abnormal examination item, and the result analysis report is generated according to the abnormal examination item and the abnormal conclusion.
[0106] In this embodiment, by performing word segmentation on the current medical history text, the current medical history segmentation can be effectively obtained, by performing entity recognition on the current medical history segmentation, the entity type of the current medical history segmentation can be effectively determined, based on the entity type, the entity object of the current medical history segmentation can be effectively determined, by classifying the current medical history segmentation through the entity object, the object set can be effectively obtained, by matching the physical examination items with the entity objects, the part matching results can be effectively obtained, based on the part matching results, the associated set of the physical examination items in the object set can be effectively determined, by performing semantic matching of the examination conclusions of the physical examination items with the corresponding associated sets, it can be effectively detected whether there is a semantic conflict between the examination conclusions and the current medical history text, so as to achieve the effect of automatically analyzing whether there are errors in the physical examination results, thereby improving the efficiency of analyzing the physical examination results.
[0107] Embodiment 3
[0108] Figure 3 2 is a block diagram of a terminal device 2 provided in the third embodiment of the present application. Figure 3 As shown, the terminal device 2 of this embodiment includes: a processor 20, a memory 21, and a computer program 22 stored in the memory 21 and executable on the processor 20, such as a program of a method for analyzing physical examination results based on outpatient cases. When the processor 20 executes the computer program 22, the steps in each embodiment of the method for analyzing physical examination results based on outpatient cases are implemented.
[0109] Exemplarily, the computer program 22 may be divided into one or more modules, which are stored in the memory 21 and executed by the processor 20 to complete the present application. The one or more modules may be a series of computer program instruction segments capable of completing specific functions, which are used to describe the execution process of the computer program 22 in the terminal device 2. The terminal device may include, but is not limited to, a processor 20 and a memory 21.
[0110] The processor 20 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc.
[0111] The memory 21 may be an internal storage unit of the terminal device 2, such as a hard disk or memory of the terminal device 2. The memory 21 may also be an external storage device of the terminal device 2, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. equipped on the terminal device 2. Further, the memory 21 may also include both an internal storage unit and an external storage device of the terminal device 2. The memory 21 is used to store the computer program and other programs and data required by the terminal device. The memory 21 may also be used to temporarily store data that has been output or is to be output.
[0112] In addition, each functional module in each embodiment of the present application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The above integrated unit can be implemented in the form of hardware or in the form of software functional unit.
[0113] If the integrated module is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Among them, the computer-readable storage medium can be non-volatile or volatile. Based on this understanding, the present application implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable storage medium may include: any entity or device capable of carrying computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium, etc. It should be noted that the content contained in computer-readable storage media can be appropriately increased or decreased according to the requirements of legislation and patent practices in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practices, computer-readable storage media do not include electrical carrier signals and telecommunication signals.
[0114] The embodiments described above are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, a person skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.
Claims
1. A method for analyzing physical examination results based on outpatient cases, characterized in that: The method comprises: Obtaining the text of the history of current illness in the outpatient medical record, and segmenting the text of the history of current illness to obtain the segmentation of the history of current illness; Performing entity recognition on the present medical history segmentation, and determining the entity type of the present medical history segmentation according to the entity recognition result; Determine the entity object of the present medical history segmentation word according to the entity type, and classify the present medical history segmentation word according to the entity object to obtain an object set; Matching the physical examination items in the physical examination result with the entity object to obtain a part matching result, and determining an associated set of the physical examination items in the object set according to the part matching result; The examination conclusion of the physical examination item is semantically matched with the corresponding association set to obtain a semantic matching result, and a result analysis report is generated according to the semantic matching result.
2. The method for analyzing physical examination results based on outpatient cases according to claim 1, characterized in that: Partially matching the physical examination items in the physical examination results with the entity object to obtain a part matching result includes: Obtaining the examination identification of the physical examination item, and determining the examination site according to the examination identification; A sub-part of the inspection part is obtained, and the inspection part and the sub-part are matched with the entity object respectively to obtain the part matching result.
3. The method for analyzing physical examination results based on outpatient cases according to claim 2, characterized in that: Determining, according to the part matching result, an associated set of the physical examination items in the object set, including: If any of the examination parts or the sub-parts matches the entity object, determining the object set corresponding to the entity object as the associated set of the physical examination items corresponding to the examination part or the sub-part; If the inspection part and the corresponding sub-part do not match the physical object, the physical examination item corresponding to the inspection part is marked as an abnormal inspection item, and an analysis abnormality prompt is sent for the abnormal inspection item.
4. The method for analyzing physical examination results based on outpatient cases according to claim 1, characterized in that: Performing semantic matching on the examination conclusion of the physical examination item and the corresponding association set to obtain a semantic matching result includes: Acquire the text position of the entity object in the association set, and determine the medical history description of the entity object according to the text position; Semantically identifying the examination conclusion and the medical history description respectively to obtain conclusion semantics and medical history semantics; The conclusion semantics are respectively combined with the medical history semantics in the corresponding association set to obtain a semantic combination, and the semantic combination is matched with the semantic list to obtain the semantic matching result.
5. The method for analyzing physical examination results based on outpatient cases according to claim 4, characterized in that: Generate a result analysis report based on the semantic matching result, including: If any of the semantic combinations matches the semantic list, it is determined that there is a semantic conflict between the conclusion semantics in the semantic combination and the medical history semantics, and the examination conclusion corresponding to the conclusion semantics of the semantic conflict is determined as an abnormal conclusion; The physical examination item corresponding to the abnormal conclusion is determined as an abnormal examination item, and the result analysis report is generated according to the abnormal examination item and the abnormal conclusion.
6. The method for analyzing physical examination results based on outpatient cases according to claim 1, characterized in that: Performing entity recognition on the present medical history segmentation, and determining the entity type of the present medical history segmentation according to the entity recognition result, including: Performing phoneme conversion on the present medical history segmentation to obtain present medical history phonemes, and performing phoneme encoding on the present medical history phonemes to obtain phoneme features; Performing vector conversion on the phoneme feature to obtain a phoneme vector, and calculating the similarity between the phoneme vector and a preset entity vector to obtain a vector similarity; The type corresponding to the maximum vector similarity is determined as the entity type of the current medical history segmentation corresponding to the phoneme vector.
7. The method for analyzing physical examination results based on outpatient cases according to claim 1, characterized in that: Performing entity recognition on the present medical history segmentation, and determining the entity type of the present medical history segmentation according to the entity recognition result, including: Calculate the hash value of the current medical history segmentation to obtain the current medical history hash value, and map the current medical history hash value to a preset hash ring to obtain a hash position; A target hash bucket is determined according to the hash position and the ring direction of the preset hash ring, and a type corresponding to the target hash bucket is determined as the entity type of the present medical history segmentation.
8. A physical examination result analysis system based on outpatient cases, characterized in that: The system comprises: A text segmentation module is used to obtain the current medical history text in the outpatient medical record and segment the current medical history text to obtain the current medical history segmentation; An entity recognition module, used for performing entity recognition on the present medical history segmentation, and determining the entity type of the present medical history segmentation according to the entity recognition result; An object classification module, used for determining the entity object of the present medical history segmentation word according to the entity type, and classifying the present medical history segmentation word according to the entity object to obtain an object set; A set determination module, used for matching the physical examination items in the physical examination results with the entity objects to obtain a part matching result, and determining an associated set of the physical examination items in the object set according to the part matching result; The report generation module is used to perform semantic matching on the examination conclusion of the physical examination item and the corresponding association set to obtain a semantic matching result, and generate a result analysis report according to the semantic matching result.
9. The physical examination result analysis system based on outpatient cases according to claim 8, characterized in that: The set determination module is also used for: Obtaining the examination identification of the physical examination item, and determining the examination site according to the examination identification; A sub-part of the inspection part is obtained, and the inspection part and the sub-part are matched with the entity object respectively to obtain the part matching result.
10. A computer-readable storage medium storing a computer program, 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 7 are implemented.
