Inspection report editing method, device and equipment
By automatically filtering and matching ultrasound examination report templates on the server, using word frequency vectors and TF-IDF values to calculate the similarity, the problem of low efficiency of doctors manually selecting templates in the existing technology is solved, and a fast and efficient variety of types of examination reports are achieved edit.
Patent Information
- Application Number
- CN202210516341.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-05-12
- Publication Date
- 2024-11-01
- Estimated Expiration
- 2042-05-12
AI Technical Summary
The existing ultrasound report editing method requires doctors to manually select templates. It is inefficient and limited by local machine configuration, so it cannot effectively support multiple types of examination report editing.
By obtaining the inspection location and type information in the inspection report editing instructions on the server, filtering the matching report templates, and using word frequency vectors and TF-IDF values to calculate the similarity, automatically determine the target report template, and sending it to the device side to reduce manual selection by doctors steps.
It improves the speed and efficiency of editing inspection reports, supports multiple types of inspection reports editing, without adding auxiliary equipment, and reduces the burden on doctors.
Smart Images

Figure CN114841136B_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of natural language processing, and in particular to an inspection report editing method, device and equipment. Background Art
[0002] Currently, ultrasound reports in hospitals are basically written using a stand-alone workstation, and its performance is limited by the configuration of the workstation itself. Editing reports requires data entry personnel to manually enter relevant examination information, ultrasound findings, and ultrasound conclusions, which wastes time and is prone to typos. Even if the workstation supports pre-entering templates locally, the operation of selecting and applying templates is cumbersome, and once there is a lot of template data, it is also complicated to screen. There is still a lot of room for efficiency improvement in this regard.
[0003] The existing methods for improving the speed of editing ultrasound examination reports require additional configuration of barcode scanning equipment and card reading equipment, such as Figure 1 As shown, each time a new patient is examined, the examining physician can operate the examination equipment to query the records of the parts of the patient that need to be examined by scanning the patient's barcode or reading the patient's medical card. After obtaining an image with diagnostic significance during the ultrasound examination, the image feature value of the image is calculated and compared with the pre-stored records one by one, that is, the number of identical binary bits of the two image feature values is calculated. When the feature values are determined to be consistent, the normal report template and common disease template corresponding to the part are queried based on the examination part identified from the image, and then the physician selects the corresponding template based on the examination results. The existing technical performance is limited by the local machine, requiring the doctor to intervene in the selection of templates, and supporting a single type of examination. Summary of the invention
[0004] The present application provides a method, device and equipment for editing an examination report, which solves the problem that in existing methods for editing examination reports, doctors need to manually select templates.
[0005] In a first aspect, the present application provides a method for editing an inspection report, which is applied to a server, and the method comprises:
[0006] In response to the inspection report editing instruction sent by the inspection device, obtain the inspection part, inspection type and inspection result related information carried in the inspection report editing instruction;
[0007] Filtering at least one inspection report template to be selected that matches the inspection part and the inspection type from the inspection report templates in the database;
[0008] Obtaining the importance of different participles in the pre-parsed inspection report template to be selected, parsing the frequency of occurrence of different participles in the inspection report template to be selected, obtaining a first word frequency vector, and parsing the importance and frequency of occurrence of different participles in the inspection result related information to obtain a second word frequency vector;
[0009] According to the similarity between the first word frequency vector and the second word frequency vector, a target inspection report template is determined and sent to the inspection device end.
[0010] In a possible implementation manner, in response to the inspection report editing instruction sent by the inspection device, the method further includes:
[0011] Obtain the current patient's examination report template usage record, and count the number of times the current patient uses different examination report templates;
[0012] Determining a target inspection report template according to the similarity between the first word frequency vector and the second word frequency vector includes:
[0013] A target examination report template is determined according to the similarity between the first word frequency vector and the second word frequency vector, and the number of times the current patient uses different examination report templates.
[0014] In a possible implementation manner, the information related to the inspection result includes inspection data obtained by the inspection device, and the inspection conclusion data obtained based on experience based on the inspection data.
[0015] In a possible implementation, it further includes:
[0016] Importing data matching preset items in different selected inspection report templates in the database into the data table, wherein the preset items include template identification, template name, inspection part, inspection data and inspection conclusion data;
[0017] Segmenting the inspection findings data and the inspection conclusion data in the data table and removing stop words to obtain different segmentations in the different inspection report templates;
[0018] The term frequency-inverse document frequency value TF-IDF value of the different word segments is calculated respectively to obtain the importance of the different word segments in the different inspection report templates.
[0019] In a possible implementation, parsing the importance of different segmentations in the information related to the inspection result includes:
[0020] Segmenting the information related to the inspection result to obtain different segmentations in the information related to the inspection result;
[0021] The TF-IDF values of the different participles are calculated to obtain the importance of the different participles in the information related to the inspection result.
[0022] In a possible implementation, obtaining the importance of different word segments in the pre-parsed inspection report template to be selected, and parsing the frequencies of occurrence of different word segments in the inspection report template to be selected to obtain a first word frequency vector includes:
[0023] Selecting a first preset number of participles from different participles of the inspection report template to be selected in order of importance, and selecting a second preset number of participles from different participles of the information related to the inspection result in order of importance, and forming a keyword set with the first preset number of participles and the second preset number of participles;
[0024] The frequency of each word segment in the keyword set appearing in the to-be-selected inspection report template is counted to obtain a first word frequency vector, wherein the frequency corresponds one-to-one to the word segment in the keyword set.
[0025] In a possible implementation, the importance and frequency of occurrence of different word segments in the information related to the inspection result are analyzed to obtain a second word frequency vector, including:
[0026] Selecting a first preset number of participles from different participles of the inspection report template to be selected in order of importance, and selecting a second preset number of participles from different participles of the information related to the inspection result in order of importance, and forming a keyword set with the first preset number of participles and the second preset number of participles;
[0027] The frequency of each word segment in the keyword set appearing in the information related to the inspection result is counted to obtain a second word frequency vector, wherein the frequency corresponds to the word segment in the keyword set one by one.
[0028] In a possible implementation, determining a target examination report template according to the similarity between the first word frequency vector and the second word frequency vector and the number of times the current patient uses different examination report templates includes:
[0029] Calculate the cosine similarity cosθ between the first word frequency vector and the second word frequency vector;
[0030] The cosine similarity magnification value is calculated using the similarity magnification formula COSB=cosθ×(1+sin(T×π / 180)) to obtain the cosine similarity magnification value corresponding to at least one candidate examination report template, where T is the number of times the current patient uses different examination report templates;
[0031] A third preset number of target inspection report templates are selected in sequence according to the magnitude order of the cosine similarity magnification values as the target inspection report templates.
[0032] In a second aspect, the present application provides an inspection report editing method, which is applied to an inspection device, and the method includes:
[0033] In response to the information entry instruction, an information entry interface including different fields is displayed on the display interface;
[0034] Collect the inspection location and inspection type entered in different fields of the information entry interface, and collect the inspection result-related information entered in the corresponding fields of the information entry interface during the inspection process;
[0035] In response to the report editing and generating instruction, generating an inspection report editing instruction carrying the inspection part, inspection type and inspection result related information and sending the instruction to the server;
[0036] Receive the target inspection report template sent by the server, and extract the content in the target inspection report template and display it in the corresponding column of the information entry interface;
[0037] The inspection result-related information includes the inspection data obtained and the inspection conclusion data obtained based on the inspection data according to experience.
[0038] In a third aspect, the present application provides an inspection report editing device, which is applied to a server, and the device includes:
[0039] An acquisition module, configured to respond to an inspection report editing instruction sent by the inspection device end and acquire information related to the inspection part, inspection type and inspection result carried in the inspection report editing instruction;
[0040] A screening module, used to screen out at least one inspection report template to be selected that matches the inspection part and the inspection type from the inspection report templates in the database;
[0041] A parsing module, used to obtain the importance of different participles in the pre-parsed inspection report template to be selected, and parse the frequency of occurrence of different participles in the inspection report template to be selected to obtain a first word frequency vector, and parse the importance and frequency of occurrence of different participles in the inspection result related information to obtain a second word frequency vector;
[0042] A determination module is used to determine a target inspection report template according to the similarity between the first word frequency vector and the second word frequency vector and send it to the inspection device end.
[0043] In a fourth aspect, the present application provides an inspection report editing device, which is applied to an inspection device, and the device includes:
[0044] A display module, for displaying an information entry interface including different fields on a display interface in response to an information entry instruction;
[0045] The collection module is used to collect the inspection part and inspection type input in different fields of the information input interface, and collect the inspection result related information input in the corresponding fields of the information input interface during the inspection process;
[0046] A sending module, configured to generate, in response to the report editing and generating instruction, an inspection report editing instruction carrying information related to the inspection part, inspection type and inspection result and send the instruction to the server;
[0047] A receiving module, used for receiving a target inspection report template sent by a server, and extracting the content in the target inspection report template to display in a corresponding column of the information entry interface;
[0048] The inspection result-related information includes inspection data obtained through the inspection device, and the inspection conclusion data obtained based on experience based on the inspection data.
[0049] In a fifth aspect, the present application provides an inspection report editing device, the device comprising:
[0050] At least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the method as described in the first aspect or the second aspect above.
[0051] In a sixth aspect, an embodiment of the present application provides a computer storage medium, wherein the computer storage medium stores a computer program, and the computer program is used to enable a computer to execute the method in the first aspect or the second aspect as described above.
[0052] The present application provides a method, device and equipment for editing examination reports, which matches the examination result-related information entered by the doctor with the data in the template to determine the target examination report template corresponding to the current patient, thereby improving the editing speed of the examination report and eliminating the need to add new auxiliary equipment or the need for the doctor to manually select the corresponding examination report template based on experience. BRIEF DESCRIPTION OF THE DRAWINGS
[0053] Figure 1 An existing inspection report editing method according to an exemplary embodiment of the present invention is exemplified;
[0054] Figure 2 A schematic diagram of cosine similarity calculation according to an exemplary embodiment of the present invention;
[0055] Figure 3 A schematic diagram of a first application scenario of the inspection report editing method according to an exemplary embodiment of the present invention;
[0056] Figure 4 A schematic diagram of a second application scenario of the inspection report editing method according to an exemplary embodiment of the present invention;
[0057] Figure 5 A schematic diagram of a third application scenario of the inspection report editing method according to an exemplary embodiment of the present invention;
[0058] Figure 6 A schematic diagram of a fourth application scenario of the inspection report editing method according to an exemplary embodiment of the present invention;
[0059] Figure 7 A flowchart of a method for editing an inspection report according to an exemplary embodiment of the present invention;
[0060] Figure 8 A schematic diagram of a screening method according to an exemplary embodiment of the present invention;
[0061] Fig. 9 A flowchart of a matching method according to an exemplary embodiment of the present invention;
[0062] Fig.10 A schematic diagram of a preprocessing flow chart of an inspection report editing method according to an exemplary embodiment of the present invention;
[0063] Fig.11 A schematic diagram of a process of determining a first word frequency vector according to an exemplary embodiment of the present invention;
[0064] Fig.12 A schematic diagram of a process of determining a second word frequency vector according to an exemplary embodiment of the present invention;
[0065] Fig.13 A schematic diagram of a specific flow chart of a method for editing an inspection report according to an exemplary embodiment of the present invention;
[0066] Fig.14 A schematic diagram of a specific flow chart of a method for editing an inspection report according to an exemplary embodiment of the present invention;
[0067] Fig.15 A schematic diagram of a specific flow chart of a method for editing an inspection report according to an exemplary embodiment of the present invention;
[0068] Fig.16 A schematic diagram of an inspection report editing system according to an exemplary embodiment of the present invention;
[0069] Fig.17A flowchart of another inspection report editing method according to an exemplary embodiment of the present invention;
[0070] Fig.18 A schematic diagram of an interface display of a method for editing an inspection report according to an exemplary embodiment of the present invention;
[0071] Fig.19 A schematic diagram of an interface display of a method for editing an inspection report according to an exemplary embodiment of the present invention;
[0072] Fig. 20 A schematic diagram of an interface display of a method for editing an inspection report according to an exemplary embodiment of the present invention;
[0073] Fig.21 A schematic diagram of an inspection report editing device according to an exemplary embodiment of the present invention;
[0074] Fig. 22 A schematic diagram of another inspection report editing device according to an exemplary embodiment of the present invention;
[0075] Fig.23 It is a schematic diagram of an inspection report editing device according to an exemplary embodiment of the present invention. DETAILED DESCRIPTION
[0076] The technical solutions in the embodiments of the present application will be described clearly and in detail below in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0077] First, the professional terms of the embodiments of the present application are introduced:
[0078] (1) Term frequency-inverse document frequency (TF-IDF):
[0079] TF-IDF value is a common weighting technique used in information retrieval and data mining. TF-IDF is used to evaluate the importance of a word to a document set or one of the documents in a corpus. The importance of a word increases in direct proportion to the number of times it appears in the document, but it decreases in inverse proportion to the frequency of its appearance in the corpus.
[0080] TF-IDF has two meanings, one is "Term Frequency" (TF), and the other is "Inverse Document Frequency" (IDF).
[0081] TF-IDF algorithm steps:
[0082] The first step is to calculate the word frequency:
[0083] Term frequency (TF) = the number of times a word appears in an article. Considering that articles are of different lengths, in order to facilitate the comparison of different articles, "term frequency" standardization is performed, which can be achieved through formula 1:
[0084] Word frequency (TF) = the number of times a word appears in an article / the total number of words in the article Formula 1;
[0085] In the embodiment of the present application, the article refers to an inspection report template.
[0086] The second step is to calculate the inverse document frequency:
[0087] At this time, a corpus is needed to simulate the language usage environment, which can be achieved through formula 2:
[0088] Inverse Document Frequency (IDF) = log(total number of documents in the corpus / (number of documents containing the word + 1))
[0089] Formula 2
[0090] It can be seen that TF-IDF is proportional to the number of times a word appears in a document, and inversely proportional to the number of times the word appears in the entire corpus. That is to say, the more times a word appears in a document, the more important the word is in the document; if the number of times a word appears in the entire corpus is higher, it means that the contents of the documents in the corpus are all related to the word, and at this time it is impossible to distinguish the types of documents in the corpus.
[0091] (2) Cosine similarity:
[0092] Cosine similarity uses the cosine value of the angle between two vectors in the vector space as a measure of the difference between two individuals. The closer the cosine value is to 1, the closer the angle is to 0 degrees, that is, the more similar the two vectors are, which is called "cosine similarity".
[0093] like Figure 2 As shown, taking two-dimensional space as an example, vector a is (x1, y1) and vector b is (x2, y2), and we want to calculate their angle θ. The cosine similarity can be obtained by formula 3-5:
[0094]
[0095]
[0096]
[0097] like Figure 3 The diagram shows an application scenario of a method for editing an examination report provided in an embodiment of the present application, and the application scenario includes: a server 301, a database 302, and at least one ultrasonic examination device (ultrasonic examination device 303_1, ultrasonic examination device 303_2, and ultrasonic examination device 303_N in the figure). The ultrasonic examination device is used to collect ultrasonic images of the patient and upload them to the server 301. The server 301 assists the doctor in editing the examination report by using the method for editing an examination report provided in the present application, and the database 302 stores the data and programs required for the server 301 to execute the method for editing an examination report.
[0098] In a possible implementation, the inspection report editing method provided in the embodiment of the present application can be used for editing a single type of inspection report, or can be used for editing multiple types of inspection reports. In addition to ultrasonic inspection, it is also applicable to editing electronic computed tomography inspection reports, such as Figure 4 The application scenario shown includes at least one electronic computer tomography inspection device (the electronic computer tomography inspection device 403_1, the electronic computer tomography inspection device 403_2, and the electronic computer tomography inspection device 403_N shown in the figure); it is also suitable for editing a computer ray inspection report, such as Figure 5 The application scenario shown includes at least one computer ray inspection device (the computer ray inspection device 503_1, the computer ray inspection device 503_2, and the computer ray inspection device 503_N shown in the figure); in addition, it can also be applied to the editing of multiple types of inspection reports at the same time, such as Figure 6 The application scenario shown also includes ultrasonic inspection equipment 603_1, electronic computer tomography inspection equipment 603_2, and computer ray inspection equipment 603_3.
[0099] The core of an inspection report editing method provided in an embodiment of the present application is to determine a target inspection report template by means of natural language retrieval and matching. However, the image recognition and feature comparison matching methods in the prior art are only applicable to imaging devices and cannot be used to edit non-imaging inspection reports. The method provided in an embodiment of the present application can solve the above problem.
[0100] The present application provides a method for editing an inspection report, such as Figure 7 As shown, the method includes:
[0101] S701: In response to an inspection report editing instruction sent by an inspection device, obtain information related to an inspection part, inspection type, and inspection result carried in the inspection report editing instruction.
[0102] When editing the examination report, the doctor needs to enter the patient information, examination site, examination type and related information of the examination results. The patient information includes the patient's name, gender, age, number and other information; the examination site is such as thyroid, heart, etc.; the examination type is specifically the type of examination equipment, such as: ultrasound examination equipment, electronic computer tomography examination equipment, etc., and can also be different examination components of the same type of examination equipment. For example, the ultrasound probes used by ultrasound examination equipment to examine different parts are different. In addition, ultrasound examination equipment includes many types, such as A-type ultrasound diagnostic instrument, B-type ultrasound diagnostic instrument, D-type ultrasound diagnostic instrument, M-type ultrasound diagnostic instrument, ultrasound Doppler diagnostic instrument, etc. Doctors can fill in the type of equipment according to the requirements of the hospital, such as B-type ultrasound diagnostic instrument, or fill in the equipment number including numbers and letters, such as SIEMENS Oxana3.
[0103] In a possible implementation, the information related to the examination result includes the examination findings data obtained by the examination device, and the examination conclusion data obtained based on the examination findings data according to experience. After the doctor examines the patient through the examination device, the doctor can obtain the examination findings and the examination conclusion. The doctor will fill in the keywords that he thinks are more important in a targeted manner, rather than filling them in randomly. If the doctor obtains an accurate matching result when filling in the examination findings, there is no need to write the examination conclusion.
[0104] S702: Filter out at least one inspection report template to be selected that matches the inspection part and the inspection type from the inspection report templates in the database.
[0105] Since doctors will fill in the examination site and examination equipment (type) in advance, the system will screen all the examination report templates in the database according to the entered examination site and examination type. Figure 8 As shown:
[0106] S801: Filtering the examination report templates that meet the requirements from the examination report templates in the database according to the examination part filled in by the doctor;
[0107] S802: From the multiple examination report templates screened according to the examination site, screen an examination report template that meets the requirements according to the examination type filled in by the doctor.
[0108] Specifically, S801 may be executed first, ie, screening of the examination site, and then S802, ie, screening of the examination type; or S802 may be executed first, and then S801. The steps may be determined according to the order of input by the doctor.
[0109] Then, match the examination report template corresponding to the examination result information entered by the doctor in the screened examination report template, such as Fig. 9 As shown:
[0110] S901: Obtaining the entered information related to the inspection result;
[0111] S902: Match the examination report template corresponding to the examination result-related information entered by the doctor among the screened examination report templates.
[0112] The above method ensures that when there are multiple types of inspection report templates in the database, there is no need to match inspection report templates corresponding to irrelevant inspection parts and inspection report templates of irrelevant types, thereby improving the matching efficiency and saving server resources.
[0113] S703: Obtain the importance of different participles in the pre-parsed inspection report template to be selected, and parse the frequency of occurrence of different participles in the inspection report template to be selected to obtain a first word frequency vector, and parse the importance and frequency of occurrence of different participles in the inspection result-related information to obtain a second word frequency vector.
[0114] S703 is the process of processing the acquired data. Before matching the entered information with the inspection report template, the data needs to be processed into information that can be recognized by the computer. The importance of different word segments is obtained by parsing the inspection report template and the data in the inspection result related information, so as to give priority to matching the word segments with high importance in the subsequent matching process, so as to improve the matching efficiency.
[0115] The first word frequency vector and the second word frequency vector are composed of the frequencies of different word segments. When the frequency is 0, the corresponding position in the word frequency vector is 0. Among them, the first word frequency vector includes at least one, which corresponds to at least one inspection report template to be selected, and the second word frequency vector is obtained by parsing the relevant information of the current inspection result, so there is only one.
[0116] S704: Determine a target inspection report template based on the similarity between the first word frequency vector and the second word frequency vector and send it to the inspection device.
[0117] After parsing to obtain the first word frequency vector and the second word frequency vector, respectively calculate the similarity between the first word frequency vector and the second word frequency vector corresponding to at least one candidate inspection report template, and specifically calculate the cosine similarity between the two. According to the matching result, select the inspection report template with the highest matching degree in the matching result, or provide the doctor with multiple inspection report templates with high matching degrees for the doctor to choose.
[0118] An embodiment of the present application provides a method for editing an inspection report, which obtains the inspection site, inspection type and inspection result related information carried by the inspection report editing instruction; then, from the inspection report templates in the database, screens out at least one inspection report template that matches the inspection site and inspection type; then obtains the importance of different participles in the inspection report template that has been pre-parsed, and parses the frequency of occurrence of different participles in the inspection report template to obtain a first word frequency vector, and parses the importance and frequency of occurrence of different participles in the inspection result related information to obtain a second word frequency vector; finally, based on the similarity between the first word frequency vector and the second word frequency vector, determines the target inspection report template and sends it to the inspection device. According to the above-mentioned editing method, not only can the editing efficiency be improved and the burden on doctors be reduced, but also multiple types of inspection reports can be supported, not just limited to inspection reports of imaging devices.
[0119] The importance of different segmented words in the selected inspection report template analyzed in S703 can be specifically implemented as follows:
[0120] In a possible implementation, data matching preset items in different selected inspection report templates in the database are imported into the data table, wherein the preset items include template identification, template name, inspection site, inspection findings and inspection conclusion data;
[0121] The above data table is shown in Table 1:
[0122] Table 1
[0123] id name organ comment conclusion
[0124] The id in Table 1 is the template identifier, through which the data in the data table can be linked to the examination report template; name is the template name, which does not simply refer to the examination site as the template name. Taking ultrasound examination as an example, the template name is superficial organ color Doppler ultrasound examination; organ is the examination site, such as mesenteric lymph nodes. It should be noted that different examination sites may correspond to the same template name. For example, the template names corresponding to the examination sites abdominal wall incision and thyroid can both be superficial organ color Doppler ultrasound examination; comment is the examination findings; conclusion is the examination conclusion. The embodiment of the present application mainly matches the examination findings and the examination conclusions.
[0125] After importing the data table, preprocess the data in the data table. The specific process is as follows: Fig.10 As shown:
[0126] 1) Participle.
[0127] Segmenting the inspection findings data and the inspection conclusion data in the data table and removing stop words to obtain different segmentations in the different inspection report templates;
[0128] There are many word segmentation algorithms, for example, dictionary-based word segmentation algorithms include: forward maximum matching method, backward maximum matching method; statistics-based word segmentation algorithms include: language model-based, hidden Markov model, conditional random field; neural network-based word segmentation algorithms include: long short-term memory neural network, pre-training model + knowledge distillation. The appropriate word segmentation algorithm can be selected according to the actual situation, and no specific limitation is made in the embodiments of the present application.
[0129] 2) Calculate the TF-IDF value.
[0130] The term frequency-inverse document frequency value TF-IDF value of the different word segments is calculated respectively to obtain the importance of the different word segments in the different inspection report templates.
[0131] 3) Sorting of results.
[0132] Sort different words in descending order according to their TF-IDF values.
[0133] 4) Storage.
[0134] The sorted TF-IDF values are stored in correspondence with different word segments.
[0135] In the embodiment of the present application, word segmentation and calculation of word frequency-inverse document frequency value TF-IDF value of different word segmentations are mainly aimed at the inspection findings and inspection conclusions in the inspection report template.
[0136] It is determined that the addition, modification and deletion of any template data requires updating the entire database through the above preprocessing operations.
[0137] The importance of different word segments in the information related to the inspection result in S703 can be analyzed in the following manner:
[0138] Segmenting the information related to the inspection result to obtain different segmentations in the information related to the inspection result;
[0139] The TF-IDF values of the different participles are calculated to obtain the importance of the different participles in the information related to the inspection result.
[0140] The word segmentation and calculation of TF-IDF values of different word segments can be performed using the above algorithm and formula.
[0141] The determination of the first word frequency vector in S703 may be specifically carried out as follows:
[0142] like Fig.11As shown, determining the first word frequency vector specifically includes two steps:
[0143] Unlike word segmentation and calculation of TF-IDF values of different word segmentations, statistical frequency of occurrence of different word segmentations in the examination report template requires the combination of relevant information of the examination results entered by the doctor, which is a real-time process; while word segmentation of the data in the selected examination report template and calculation of TF-IDF values of different word segmentations require pre-parsing.
[0144] In a possible implementation, obtaining the importance of different word segments in the pre-parsed inspection report template to be selected, and parsing the frequencies of occurrence of different word segments in the inspection report template to be selected to obtain a first word frequency vector includes:
[0145] S1101: Selecting a first preset number of participles from different participles of the inspection report template to be selected in order of importance, and selecting a second preset number of participles from different participles of the inspection result related information in order of importance, and forming a keyword set with the first preset number of participles and the second preset number of participles;
[0146] S1102: Count the frequency of each word in the keyword set appearing in the inspection report template to be selected, and obtain a first word frequency vector.
[0147] The frequencies correspond one-to-one to the participles in the keyword set.
[0148] The first preset number and the second preset number are not specifically limited here. Only implementation methods are given in the embodiment, and specific methods can be selected according to actual circumstances.
[0149] For example: from the different participles of the inspection report template to be selected, select a first preset number of participles [ABCD] in order of importance, and from the different participles of the information related to the inspection result, select a second preset number of participles [AE] in order of importance, and combine [ABCD] and [AE] to form a keyword set [AB CDE], wherein the elements in the keyword set are different and not a list of all participles. After obtaining the keyword set, count the frequency of each participle in the keyword set [ABCDE] in the inspection report template to be selected [10 86 3 0] to obtain the first word frequency vector, that is, the frequency of A in the inspection report template to be selected is 10; the frequency of B in the inspection report template to be selected is 8; the frequency of C in the inspection report template to be selected is 6; the frequency of D in the inspection report template to be selected is 3; the frequency of E in the inspection report template to be selected is 0.
[0150] The second word frequency vector can be determined in S703 as follows:
[0151] like Fig.12 As shown, determining the first word frequency vector specifically includes two steps:
[0152] Analyze the importance and frequency of different word segments in the information related to the inspection result to obtain a second word frequency vector, including:
[0153] S1201: Selecting a first preset number of participles from different participles of the inspection report template to be selected in order of importance, and selecting a second preset number of participles from different participles of the inspection result related information in order of importance, and forming a keyword set with the first preset number of participles and the second preset number of participles;
[0154] S1202: Count the frequency of each word in the keyword set appearing in the information related to the inspection result to obtain a second word frequency vector.
[0155] Among them, the frequency corresponds one-to-one to the word segments in the keyword set, and the specific steps of obtaining the keyword set are the same as the steps of determining the keyword set in determining the first word frequency vector, which will not be repeated here.
[0156] Taking the above-mentioned key set [ABCDE] as an example, the frequency of each word in the keyword set appearing in the information related to the inspection result is counted. The frequency of A appearing in the information related to the inspection result is 1; the frequency of B appearing in the information related to the inspection result is 0; the frequency of C appearing in the information related to the inspection result is 0; the frequency of D appearing in the information related to the inspection result is 0; the frequency of E appearing in the information related to the inspection result is 1, and the corresponding second word frequency vector [1 0 0 0 1] is obtained.
[0157] It should be noted that when the doctor enters information related to the examination results, he can enter one or more. There is no specific limitation here, because the process of matching the examination report template is a real-time process, that is, real-time matching is performed based on the examination result-related information entered by the doctor. The doctor may get a matching target examination report template when he enters one word, or he may need to enter multiple words to get a matching target examination report template.
[0158] In the above S704, determining the target inspection report template according to the similarity between the first word frequency vector and the second word frequency vector and sending it to the inspection device end can be specifically implemented as follows:
[0159] In a possible implementation, a usage record of the examination report template of the current patient is obtained, and the number of times the current patient uses different examination report templates is counted;
[0160] Determining a target inspection report template according to the similarity between the first word frequency vector and the second word frequency vector includes:
[0161] A target examination report template is determined according to the similarity between the first word frequency vector and the second word frequency vector, and the number of times the current patient uses different examination report templates.
[0162] Each time the target inspection report template is determined, a corresponding usage record will be generated, which includes the patient ID and the inspection report template ID used this time. When the inspection report is edited next time, the inspection report template usage record corresponding to the patient will be taken out as the basis for matching the inspection report template this time. The corresponding relationship record between patients and inspection report templates is shown in Table 2:
[0163] Table 2
[0164] Patient identification Template Logo Use times
[0165] After the doctor finishes writing and saving the examination report, the server records the corresponding template used by the patient and accumulates the number of times it is used, and saves the result in Table 2. The purpose of recording the use of the patient's examination report template is to deal with the situation where a patient needs to undergo multiple reexaminations for the same disease. In other words, the content of the examination report of the same patient in the follow-up visit is very similar to that of the previous examination report. The probability that the corresponding report template is the same will be greater. Therefore, the correlation between the cosine value and the number of times the patient uses the current report template can be amplified, which not only improves the matching efficiency, but also improves the matching accuracy.
[0166] Afterwards, according to the similarity between the first word frequency vector and the second word frequency vector, and the number of times the current patient uses different examination report templates, determining a target examination report template includes:
[0167] Calculate the cosine similarity cosθ between the first word frequency vector and the second word frequency vector;
[0168] The cosine similarity magnification value is calculated using the similarity magnification formula COSB=cosθ×(1+sin(T×π / 180)) to obtain the cosine similarity magnification value corresponding to at least one candidate examination report template, where T is the number of times the current patient uses different examination report templates;
[0169] A third preset number of target inspection report templates are selected in sequence according to the magnitude order of the cosine similarity magnification values as the target inspection report templates.
[0170] Among them, the range of T value is [0,90]. When the actual value of T is greater than 90, the result is not much different when T=90. The third preset number is not specifically limited here. If the number of a certain type of inspection report templates in the database is small, there may be only one target inspection report template obtained in the end. It is also possible to obtain the target inspection report template in the screening stage.
[0171] The following is an introduction to the specific inspection report editing method provided in the embodiment of the present application. The specific process is as follows: Fig.13 , Fig.14 and Fig.15 As shown:
[0172] S1301: Obtaining information related to the examination results input by the doctor;
[0173] S1302: Segment the input inspection result related information into words, and calculate the TF-IDF value of each word respectively;
[0174] S1303: Combining the word segmentation in S1302 and m word segmentations in at least one inspection report template to be selected into a vector Jn;
[0175] S1304: Obtain a first word frequency vector and a second word frequency vector according to the vector Jn. The specific steps are as described above.
[0176] S1305: Calculate the cosine similarity of two word frequency vectors;
[0177] S1306: Calculate the amplified value cosB of the cosine similarity;
[0178] S1307: Determine whether all inspection report templates have been compared, if so, execute S1308, otherwise execute S1303;
[0179] S1308: Get the inspection report template corresponding to the preset maximum number of cosB values.
[0180] Fig.14 To compare all inspection report templates; Fig.15 This is the case where not all inspection report templates have been compared.
[0181] In a possible implementation, before the above step S1303, at least one inspection report template to be selected can be divided into multiple groups, and then the multiple groups concurrently execute steps S1303 to S1308. The specific grouping method is not specifically limited, and can be divided equally or grouped according to preset categories.
[0182] In a possible implementation, the present application also provides an inspection report editing system, such as Fig.16As shown, it includes an inspection device end S1601 for doctors to use, a storage module S1602, a software module S1603, a display module S1604, and a server S1605. Among them, the patient information and the information related to the inspection result are obtained through the inspection device end S1601 and stored in the storage module S1602; the data in the storage module S1602 is extracted by the software module S1603, and the software module 1603 uploads the input information related to the inspection result and the patient information to the server S1605 through the calling interface; the server S1605 performs calculations, returns the matching results to the display module S1604, and records the patient and the corresponding inspection report template data, and accumulates the number of times and other operations; the display module S1604 is used to display the matching results to the doctor.
[0183] Based on the same inventive concept, the embodiment of the present application also provides an inspection report editing method, which is applied to the inspection device side, such as Fig.17 As shown, the method includes:
[0184] S1701: In response to an information entry instruction, an information entry interface including different fields is displayed on the display interface.
[0185] Specifically, the doctor can click the "Information Entry" button on the display page to check whether the device will respond to the information entry instruction. Fig.18 The blank inspection report is displayed on the information input interface. The inspection report consists of two parts. The first part is the information input interface with editable and different columns; the second part is the matching result preview interface including the matching result list and related content. Before entering information, the content of all columns is empty.
[0186] S1702: Collecting the inspection part and inspection type input in different fields of the information input interface, and collecting the inspection result-related information input in the corresponding fields of the information input interface during the inspection process.
[0187] like Fig.19 As shown in the figure, when entering information, the doctor first fills in the fields above the "horizontal line", including the patient's identity information, as well as the examination equipment (examination type) and the examination site. Filling in the examination type and examination site is to facilitate the server to select the examination report template that meets the requirements and speed up the matching process.
[0188] After filling in the above information, the doctor starts to use the inspection device to conduct the inspection. After the inspection, the doctor can determine the relevant information of the inspection result based on the inspection results such as images, wherein the inspection result related information includes the inspection data obtained by the inspection device, and the inspection conclusion data obtained based on the inspection data based on experience. Then the doctor fills in the contents in the inspection findings and inspection conclusion columns. In the process of filling in, it is only necessary to fill in at least one keyword related to the inspection result in a targeted manner, and there is no need to enter all the required content. The number of keywords can be determined by the doctor according to the inspection process, and the number of keywords is not limited here.
[0189] S1703: In response to the report editing and generating instruction, generate an inspection report editing instruction carrying information related to the inspection part, inspection type and inspection result and send it to the server;
[0190] After the doctor has finished entering the information, he can click the "Match" button in the upper left corner to match the target examination report template.
[0191] In a possible implementation, the name of the button may also be “query”, “search” or “→”, which may be specifically set according to the user's habits and is not specifically limited here.
[0192] In a possible implementation, the process of matching the target inspection report template may not be performed by clicking a button such as "match". The inspection device may send the information entered by the doctor to the server in real time, and the server will parse the entered information. After the doctor enters the information, no result is displayed on the matching result preview interface, that is, the target inspection report template is not matched. At this time, the doctor may need to continue to enter information related to the inspection result. The inspection device will automatically collect the newly entered information, integrate the information into the previously entered information, and send it together to the server.
[0193] S1704: Receive the target inspection report template sent by the server, and extract the content in the target inspection report template to display it in the corresponding column of the information entry interface.
[0194] like Fig. 20The following is the display interface of the target inspection report template. The matching result list includes 5 results. The number of matching results is not limited here. When result 1 is selected, its corresponding content will be displayed in the columns corresponding to the inspection findings and inspection conclusions of the matching result preview interface. The content is previewed to facilitate doctors to select the most appropriate inspection report template. The inspection device will add its content to the columns corresponding to the inspection findings and inspection conclusions of the information entry interface. Due to individual differences between patients, their inspection findings will not be exactly the same. Doctors need to manually fill in the measurement data. Fig. 20 The contents in “()” are filled in by the doctor based on the actual measurement data.
[0195] The specific process of determining the target inspection report template can be obtained through the inspection report editing method of the above-mentioned server, which will not be repeated here.
[0196] Based on the same inventive concept, the present application embodiment provides an inspection report editing device, such as Fig.21 As shown, the device 2100 includes:
[0197] The acquisition module 2101 is used to respond to the inspection report editing instruction sent by the inspection device end and obtain the inspection part, inspection type and inspection result related information carried in the inspection report editing instruction;
[0198] A screening module 2102 is used to screen out at least one inspection report template to be selected that matches the inspection part and the inspection type from the inspection report templates in the database;
[0199] The parsing module 2103 is used to obtain the importance of different participles in the pre-parsed inspection report template to be selected, and parse the frequency of occurrence of different participles in the inspection report template to be selected to obtain a first word frequency vector, and parse the importance and frequency of occurrence of different participles in the inspection result related information to obtain a second word frequency vector;
[0200] The determination module 2104 is used to determine a target inspection report template according to the similarity between the first word frequency vector and the second word frequency vector, and send the template to the inspection device.
[0201] In a possible implementation, the acquisition module 2101 is further configured to respond to the inspection report editing instruction sent by the inspection device, and further include:
[0202] Obtain the current patient's examination report template usage record, and count the number of times the current patient uses different examination report templates;
[0203] Determining a target inspection report template according to the similarity between the first word frequency vector and the second word frequency vector includes:
[0204] A target examination report template is determined according to the similarity between the first word frequency vector and the second word frequency vector, and the number of times the current patient uses different examination report templates.
[0205] In a possible implementation, the acquisition module 2101 is used to acquire information related to the inspection result, wherein the information related to the inspection result includes inspection data obtained by the inspection device and the inspection conclusion data obtained based on experience based on the inspection data.
[0206] In a possible implementation, the parsing module 2103 is used to import data matching preset items in different selected inspection report templates in the database into a data table, wherein the preset items include template identification, template name, inspection site, inspection findings, and inspection conclusion data;
[0207] Segmenting the inspection findings data and the inspection conclusion data in the data table and removing stop words to obtain different segmentations in the different inspection report templates;
[0208] The term frequency-inverse document frequency value TF-IDF value of the different word segments is calculated respectively to obtain the importance of the different word segments in the different inspection report templates.
[0209] In a possible implementation, the parsing module 2103 is further configured to parse the importance of different segmentations in the information related to the inspection result, including:
[0210] Segmenting the information related to the inspection result to obtain different segmentations in the information related to the inspection result;
[0211] The TF-IDF values of the different participles are calculated to obtain the importance of the different participles in the information related to the inspection result.
[0212] In a possible implementation, the parsing module 2103 is further used to obtain the importance of different participles in the pre-parsed inspection report template to be selected, and parse the frequency of occurrence of different participles in the inspection report template to be selected to obtain a first word frequency vector, including:
[0213] Selecting a first preset number of participles from different participles of the inspection report template to be selected in order of importance, and selecting a second preset number of participles from different participles of the information related to the inspection result in order of importance, and forming a keyword set with the first preset number of participles and the second preset number of participles;
[0214] The frequency of each word segment in the keyword set appearing in the to-be-selected inspection report template is counted to obtain a first word frequency vector, wherein the frequency corresponds one-to-one to the word segment in the keyword set.
[0215] In a possible implementation, the parsing module 2103 is further configured to parse the importance and frequency of occurrence of different word segments in the information related to the inspection result to obtain a second word frequency vector, including:
[0216] Selecting a first preset number of participles from different participles of the inspection report template to be selected in order of importance, and selecting a second preset number of participles from different participles of the information related to the inspection result in order of importance, and forming a keyword set with the first preset number of participles and the second preset number of participles;
[0217] The frequency of each word segment in the keyword set appearing in the information related to the inspection result is counted to obtain a second word frequency vector, wherein the frequency corresponds to the word segment in the keyword set one by one.
[0218] In a possible implementation, the determination module 2104 is used to determine the target examination report template according to the similarity between the first word frequency vector and the second word frequency vector, and the number of times the current patient uses different examination report templates, including:
[0219] Calculate the cosine similarity cosθ between the first word frequency vector and the second word frequency vector;
[0220] The cosine similarity magnification value is calculated using the similarity magnification formula COSB=cosθ×(1+sin(T×π / 180)) to obtain the cosine similarity magnification value corresponding to at least one candidate examination report template, where T is the number of times the current patient uses different examination report templates;
[0221] A third preset number of target inspection report templates are selected in sequence according to the magnitude order of the cosine similarity magnification values as the target inspection report templates.
[0222] Based on the same inventive concept, the embodiment of the present application provides an inspection report editing device 2200, which is applied to the inspection device end, such as Fig. 22 As shown, the device comprises:
[0223] Display module 2201, for displaying an information entry interface including different fields on a display interface in response to an information entry instruction;
[0224] The collection module 2202 is used to collect the inspection part and inspection type input in different fields of the information input interface, and collect the inspection result related information input in the corresponding fields of the information input interface during the inspection process;
[0225] The sending module 2203 is used to generate an inspection report editing instruction carrying the inspection part, inspection type and inspection result related information in response to the report editing generation instruction and send it to the server;
[0226] The receiving module 2204 is used to receive the target inspection report template sent by the server, and extract the content in the target inspection report template to display it in the corresponding column of the information input interface;
[0227] The inspection result-related information includes inspection data obtained through the inspection device, and the inspection conclusion data obtained based on experience based on the inspection data.
[0228] Based on the same inventive concept, the present application embodiment provides an inspection report editing device, such as Fig. 22 As shown, the device includes at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute any inspection report editing method in the above-mentioned embodiments.
[0229] Refer to the following Fig.23 The electronic device 130 according to this embodiment of the present application is described. Fig.23 The electronic device 130 shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.
[0230] like Fig.23 As shown, the electronic device 130 is in the form of a general electronic device. The components of the electronic device 130 may include but are not limited to: the at least one processor 131, the at least one memory 132, and a bus 133 connecting different system components (including the memory 132 and the processor 131).
[0231] The processor 131 is used to read and execute instructions in the memory 132, so that the at least one processor can execute the inspection report editing method provided in the above embodiment.
[0232] Bus 133 represents one or more of several types of bus structures, including a memory bus or memory controller, a peripheral bus, a processor, or a local bus using any of a variety of bus architectures.
[0233] The memory 132 may include a readable medium in the form of a volatile memory, such as a random access memory (RAM) 1321 and / or a cache memory 1322 , and may further include a read-only memory (ROM) 1323 .
[0234] The memory 132 may also include a program / utility 1325 having a set (at least one) of program modules 1324, such program modules 1324 including but not limited to: an operating system, one or more application programs, other program modules, and program data, each of which or some combination may include an implementation of a network environment.
[0235] The electronic device 130 may also communicate with one or more external devices 134 (e.g., keyboards, pointing devices, etc.), may also communicate with one or more devices that enable a user to interact with the electronic device 130, and / or communicate with any device that enables the electronic device 130 to communicate with one or more other electronic devices (e.g., routers, modems, etc.). Such communication may be performed via an input / output (I / O) interface 135. Furthermore, the electronic device 130 may also communicate with one or more networks (e.g., a local area network (LAN), a wide area network (WAN), and / or a public network, such as the Internet) via a network adapter 136. As shown, the network adapter 136 communicates with other modules for the electronic device 130 via a bus 133. It should be understood that, although not shown in the figure, other hardware and / or software modules may be used in conjunction with the electronic device 130, including but not limited to: microcode, device drivers, redundant processors, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.
[0236] In some possible implementations, various aspects of an inspection report editing method provided by the present application may also be implemented in the form of a program product, which includes a program code. When the program product is run on a computer device, the program code is used to enable the computer device to execute the steps of an inspection report editing method according to various exemplary implementations of the present application described above in this specification.
[0237] In addition, the present application also provides a computer-readable storage medium, wherein the computer storage medium stores a computer program, wherein the computer program is used to enable a computer to execute any of the methods described in the above embodiments.
[0238] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1A function specified in one or more boxes.
[0239] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process in the computer or other programmable device. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.
[0240] Although the preferred embodiments of the present application have been described, those skilled in the art may make additional changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications falling within the scope of the present application.
[0241] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalents, the present application is also intended to include these modifications and variations.
Claims
1. A method for editing an inspection report, characterized in that: Applied to the server, including: In response to the inspection report editing instruction sent by the inspection device, obtain the inspection site, inspection type and inspection result related information carried by the inspection report editing instruction, obtain the inspection report template usage record of the current patient, and count the number of times the current patient uses different inspection report templates; Filtering at least one inspection report template to be selected that matches the inspection part and the inspection type from the inspection report templates in the database; Selecting a first preset number of participles from different participles of the inspection report template to be selected in order of importance, and selecting a second preset number of participles from different participles of the information related to the inspection result in order of importance, and forming a keyword set with the first preset number of participles and the second preset number of participles, obtaining a first word frequency vector based on the frequency of each participle in the keyword set appearing in the inspection report template to be selected, and obtaining a second word frequency vector based on the frequency of each participle in the keyword set appearing in the information related to the inspection result; Calculate the cosine similarity cosθ between the first word frequency vector and the second word frequency vector, and use the similarity amplification formula COSB=cosθ×(1+sin(T×π / 180)) to calculate the cosine similarity amplification value to obtain at least one cosine similarity amplification value corresponding to the selected examination report template, determine the target examination report template according to the size of the cosine similarity amplification value, and send it to the examination device end, so that the doctor can fill in the actual measurement data on the information entry interface of the examination device end, where T is the number of times the current patient uses different examination report templates.
2. The method according to claim 1, characterized in that The inspection result related information includes inspection data obtained by the inspection device, and the inspection conclusion data obtained based on the inspection data according to experience.
3. The method according to any one of claims 1 to 2, characterized in that: Determine the importance of different words in the inspection report template, including: Importing data matching preset items in different selected inspection report templates in the database into the data table, wherein the preset items include template identification, template name, inspection part, inspection findings and inspection conclusion data; Segmenting the inspection findings data and the inspection conclusion data in the data table and removing stop words to obtain different segmentations in the different inspection report templates; The term frequency-inverse document frequency value TF-IDF value of the different word segments is calculated respectively to obtain the importance of the different word segments in the different inspection report templates.
4. The method according to any one of claims 1 to 2, characterized in that: Determine the importance of different segmentations in the information related to the inspection result, including: Segmenting the information related to the inspection result to obtain different segmentations in the information related to the inspection result; The TF-IDF values of the different participles are calculated to obtain the importance of the different participles in the information related to the inspection result.
5. A method for editing an inspection report, characterized in that: Applied to the inspection equipment side, including: In response to the information entry instruction, an information entry interface including different fields is displayed on the display interface; Collect the inspection location and inspection type entered in different fields of the information entry interface, and collect the inspection result-related information entered in the corresponding fields of the information entry interface during the inspection process; In response to the report editing and generating instruction, generating an inspection report editing instruction carrying the inspection part, inspection type and inspection result related information and sending the instruction to the server; Receive the target inspection report template sent by the server, extract the content in the target inspection report template and display it in the corresponding column of the information entry interface, and obtain the actual measurement data filled in by the doctor on the information entry interface of the inspection device; The information related to the examination result includes the obtained examination findings data and the examination conclusion data obtained based on the examination findings data according to experience. The target examination report template is determined based on the size of the cosine similarity magnification value of at least one examination report template to be selected. The cosine similarity magnification value can be obtained using the formula COSB=cosθ×(1+sin(T×π / 180)), cosθ is the cosine similarity between the first word frequency vector and the second word frequency vector, T is the number of times the current patient uses different examination report templates, the first word frequency vector is determined based on the frequency of each participle in the keyword set in the examination report template to be selected, the second word frequency vector is determined based on the frequency of each participle in the keyword set in the examination result related information, the keyword set is constructed based on a first preset number of participles and a second preset number of participles, the first preset number of participles are selected from different participles in the examination report template to be selected, in order of importance, and the second preset number of participles are selected from different participles in the examination result related information, in order of importance.
6. An inspection report editing device, characterized in that: Applied to the server, including: An acquisition module is used to respond to the inspection report editing instruction sent by the inspection device, obtain the inspection site, inspection type and inspection result related information carried in the inspection report editing instruction, obtain the inspection report template usage record of the current patient, and count the number of times the current patient uses different inspection report templates; A screening module, used to screen out at least one inspection report template to be selected that matches the inspection part and the inspection type from the inspection report templates in the database; A parsing module, configured to select a first preset number of participles from different participles of the inspection report template to be selected in order of importance, and select a second preset number of participles from different participles of the information related to the inspection result in order of importance, and form a keyword set with the first preset number of participles and the second preset number of participles, obtain a first word frequency vector based on the frequency of each participle in the keyword set appearing in the inspection report template to be selected, and obtain a second word frequency vector based on the frequency of each participle in the keyword set appearing in the information related to the inspection result; A determination module is used to calculate the cosine similarity cosθ between the first word frequency vector and the second word frequency vector, and use the similarity amplification formula COSB=cosθ×(1+sin(T×π / 180)) to calculate the cosine similarity amplification value to obtain the cosine similarity amplification value corresponding to at least one to-be-selected examination report template, and determine the target examination report template according to the size of the cosine similarity amplification value and send it to the examination device end, so that the doctor fills in the actual measurement data on the information entry interface of the examination device end, wherein T is the number of times the current patient uses different examination report templates.
Citation Information
Patent Citations
Endoscopy intelligent decision support system for minimally invasive therapy
CN107633156A
Method and device for automatically obtaining medical record template and storage medium
CN111696640A
System and method for dynamically generating structured report in imaging diagnosis service
CN112614563A
Inspection report generation method and system, equipment and storage medium
CN112632944A
Cited By
Method and device for generating CRF form for rare disease
EP4793960A1
