Method and system for inputting electronic medical record information of Alzheimer's disease

Abstract

The invention relates to the technical field of electronic medical record data processing, in particular to an input method and system for Alzheimer's disease electronic medical record information, and the method comprises the steps: obtaining a plurality of medical record data when a medical worker inputs an electronic medical record in a preset recent period of time, and carrying out the word segmentation of the medical record data, and obtaining a plurality of word groups; obtaining the input easiness of each word group; the comprehensive working state of the medical staff at each moment is obtained; according to the comprehensive working state of the medical staff at each moment and the proportion condition of word group errors, the input easiness is corrected, and the corrected input easiness of each word group is obtained; obtaining the priority of each phrase by combining the importance of the phrases and the revised input easiness; and constructing a list of phrases to be selected when the medical personnel input the electronic medical record information. According to the method and the device, the reasonability of constructing the list of the to-be-selected phrases when the medical personnel input the electronic medical record information is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of electronic medical record data processing, and in particular to a method and system for inputting electronic medical record information for Alzheimer's disease. Background Art

[0002] Alzheimer's disease (AD) is a common disease in the elderly, and patients usually show irreversible memory loss, cognitive and language dysfunction. Due to the unclear pathogenesis of AD and the lack of drugs for causal treatment, the medical and health care of AD patients has attracted much attention. Comprehensive collection of information on AD patients is conducive to providing individualized treatment plans and improving treatment effects, but there is a related defect in the collection and collection of relevant medical history information of AD patients, such as low information utilization rate.

[0003] When entering the electronic medical record information of existing Alzheimer's patients, the importance of the phrases is judged by analyzing the frequency of use of each phrase, so as to construct a list of candidate phrases for medical staff to select when entering the electronic medical record information; since the phrase names may be wrong when obtaining the input phrases of the medical staff, resulting in different ease of entry of each phrase, and the working status of the medical staff will also affect the accuracy of phrase entry, if the list of candidate phrases is constructed only by the importance of the phrases without considering the ease of entry of the phrases, the final constructed list of candidate phrases will be poor, which will make it inconvenient for medical staff to enter the electronic medical record information through the constructed list of candidate phrases. Summary of the invention

[0004] To solve the above problems, the present invention provides a method and system for entering electronic medical record information of Alzheimer's disease.

[0005] The method and system for entering electronic medical record information of Alzheimer's disease of the present invention adopt the following technical solutions: An embodiment of the present invention provides a method for entering electronic medical record information of Alzheimer's disease, the method comprising the following steps: Obtain a number of medical record data when medical staff enter electronic medical records in a preset recent period of time, segment the medical record data, and obtain a number of phrases; obtain the medical staff's daily working hours in the preset recent period of time, the entry time of each phrase, the voice signal when entering each electronic medical record, and the pronunciation rate and intonation characteristics of the medical staff at each moment in each voice signal; According to the frequency and distribution of the phrases in the electronic medical records, the importance of each phrase is obtained; the number of different errors in each phrase in all electronic medical records is obtained; the distribution of each error is obtained according to the number of errors, and the comprehensive error rate of each phrase is obtained according to the distribution of errors; the ease of entry of each phrase is obtained according to the entry time and comprehensive error rate of the phrase; According to the working hours of the medical staff, the first working state of the medical staff at each moment in each voice signal is obtained; according to the size of the pronunciation rate and intonation characteristics, the second working state of the medical staff at each moment in each voice signal is obtained; according to the first working state and the second working state of the medical staff at the same moment, the comprehensive working state of the medical staff at each moment in each voice signal is obtained; According to the comprehensive working status of medical staff at each moment and the proportion of phrase errors, the ease of entry is corrected to obtain the corrected ease of entry of each phrase; the priority of each phrase is obtained by combining the importance of the phrase and the corrected ease of entry; and a list of candidate phrases for medical staff to select when entering electronic medical record information is constructed according to the size of the priority.

[0006] Furthermore, the importance of each phrase is obtained according to the frequency and distribution of the phrases in the electronic medical records, and the specific steps include the following: Any phrase is recorded as a target phrase; the total number of phrases and the number of target phrase occurrences obtained after word segmentation of all medical record data are obtained, and the ratio of the number of target phrase occurrences to the total number of phrases is used as the frequency of occurrence of the target phrase in the electronic medical record; the inverse proportional value of the standard deviation of the number of target phrase occurrences in all electronic medical records is used as the distribution uniformity of the target phrase in the electronic medical record; the occurrence frequency and the distribution uniformity are merged to obtain the importance of the target phrase.

[0007] Furthermore, the method of obtaining the distribution of each error according to the number of errors and obtaining the comprehensive error rate of each phrase according to the distribution of the errors includes the following specific steps: Any phrase is recorded as the target phrase; the ratio of the number of times each error occurs in the target phrase to the total number of times the error occurs in the target phrase is taken as the distribution of each error in the target phrase; important factors for each error in the target phrase are preset; the distribution of each error in the target phrase and the important factors are integrated to obtain the initial comprehensive error rate for each error in the target phrase; the initial comprehensive error rates of all types of errors in the target phrase are accumulated and normalized to obtain the comprehensive error rate of the target phrase.

[0008] Furthermore, the step of obtaining the ease of input of each phrase based on the input time and comprehensive error rate of the phrase includes the following specific steps: Any phrase is recorded as a target phrase; the average time taken by medical staff to enter the target phrase in all electronic medical records is obtained; the average time taken to enter the target phrase and the comprehensive error rate of the target phrase are combined to obtain the ease of entry of the target phrase.

[0009] Furthermore, the step of obtaining the first working state of the medical staff at each moment in each voice signal according to the working hours of the medical staff includes the following specific steps: Any voice signal is recorded as a target voice signal; any time in the target voice signal is recorded as a target moment; the working day to which the target moment belongs is recorded as a target working day; the inverse proportional value of the working hours of the medical staff on the target working day up to the target moment is taken as the first working state of the medical staff at the target moment.

[0010] Furthermore, the method of obtaining the second working state of the medical staff at each moment in each voice signal according to the size of the pronunciation rate and the intonation feature includes the following specific steps: If the pronunciation rate at the target moment in the target speech signal is less than the average of the pronunciation rates at all moments in the target speech signal, and the intonation feature at the target moment in the target speech signal is less than the average of the intonation features at all moments in the target speech signal, the product of the reciprocal of the difference between the average of the pronunciation rates at all moments and the pronunciation rate at the target moment and the reciprocal of the difference between the average of the intonation features at all moments and the intonation features at the target moment is taken as the second working state of the medical staff at the target moment; If the pronunciation rate at the target moment in the target speech signal is less than the average of the pronunciation rates at all moments in the target speech signal, and the intonation feature at the target moment in the target speech signal is greater than or equal to the average of the intonation feature at all moments in the target speech signal, the reciprocal of the difference between the average of the pronunciation rates at all moments and the pronunciation rate at the target moment is taken as the second working state of the medical staff at the target moment; If the pronunciation rate at the target moment in the target speech signal is greater than or equal to the average of the pronunciation rates at all moments in the target speech signal, and the intonation feature at the target moment in the target speech signal is less than the average of the intonation features at all moments in the target speech signal, the reciprocal of the difference between the average of the intonation features at all moments and the intonation features at the target moment is taken as the second working state of the medical staff at the target moment; If the pronunciation rate at the target moment in the target speech signal is greater than or equal to the average pronunciation rate at all moments in the target speech signal, and the intonation feature at the target moment in the target speech signal is greater than or equal to the average intonation feature at all moments in the target speech signal, let the second working state of the medical staff at the target moment be 1.

[0011] Furthermore, the method of obtaining the comprehensive working state of the medical staff at each moment in each voice signal according to the first working state and the second working state of the medical staff at the same moment includes the following specific steps: The product of the first working state and the second working state of the medical staff at the target moment is taken as the comprehensive working state of the medical staff at the target moment.

[0012] Furthermore, the ease of input is corrected according to the comprehensive working status of the medical staff at each moment and the proportion of errors in the phrases to obtain the corrected ease of input of each phrase, which includes the following specific steps: Any phrase is recorded as a target phrase; any voice signal is recorded as a target voice signal; any time in the target voice signal is recorded as a target moment; the comprehensive working status of the medical staff at the target moment is obtained; the ratio of the number of target phrase errors in the target voice signal up to the target moment to the total number of target phrase errors in the target voice signal in the corresponding electronic medical record is used as the proportion of errors in the target phrase; the inverse proportional value of the product of the comprehensive working status of the medical staff at the target moment and the proportion of errors in the target phrase is used as the comprehensive correction factor at the target moment in the target voice signal, and the ease of input of the target phrase is weightedly corrected according to the accumulated value of the comprehensive correction factor at all moments in all voice signals to obtain the corrected ease of input of the target phrase.

[0013] Furthermore, the step of obtaining the priority of each phrase according to the importance of each phrase and the corrected ease of entry includes the following specific steps: Any phrase is recorded as a target phrase; the corrected ease of entry of the target phrase is obtained, and the corrected ease of entry of the target phrase is input into a negative exponential function with a natural constant as the base; the product of the output result and the importance of the target phrase is recorded as the priority of the target phrase.

[0014] The present invention also proposes a system for entering electronic medical record information of Alzheimer's disease, including a memory and a processor, wherein the processor executes a computer program stored in the memory to implement the steps of the aforementioned method.

[0015] The beneficial effect of the technical solution of the present invention is as follows: after obtaining a number of medical record data when medical staff enter electronic medical records in a preset recent period of time, the present invention first obtains the importance of each phrase by analyzing the frequency and distribution of phrases in the electronic medical records; then obtains the ease of entry of each phrase by analyzing the entry time and comprehensive error rate of the phrases, and corrects the ease of entry in combination with the comprehensive working status of the medical staff and the proportion of phrase errors to obtain the corrected ease of entry of each phrase, so that the ease of entry of the phrase is more accurate, thereby better ensuring the accuracy of the medical record. Determine the priority of the phrase; when obtaining the comprehensive working status of the medical staff, analyze the working hours of the medical staff, the pronunciation rate and intonation characteristics of the voice signals of the medical staff when working, so that the quantification of the working status of the medical staff is more accurate; finally, by combining the importance of the phrase and the corrected ease of input, obtain the priority of each phrase, and construct a list of candidate phrases for medical staff when entering electronic medical record information according to the size of the priority, which improves the rationality of constructing the list of candidate phrases and facilitates medical staff to better enter electronic medical record information through the constructed list of candidate phrases. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0017] Figure 1 A flowchart of the steps of a method for entering electronic medical record information of Alzheimer's disease provided by one embodiment of the present invention; Figure 2 A flowchart of obtaining a list of phrases to be selected is provided in accordance with an embodiment of the present invention. DETAILED DESCRIPTION

[0018] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following is a detailed description of the method and system for entering electronic medical record information for Alzheimer's disease proposed in accordance with the present invention, its specific implementation method, structure, features and effects, in conjunction with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0019] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0020] The specific scheme of the method and system for entering electronic medical record information of Alzheimer's disease provided by the present invention is described in detail below with reference to the accompanying drawings.

[0021] See also Figure 1 and Figure 2 , which shows a flowchart of the steps of a method for entering electronic medical record information of Alzheimer's disease provided by an embodiment of the present invention and a flowchart of obtaining a list of phrases to be selected, the method comprising the following steps: Step S001, obtain a number of medical record data when the medical staff enters the electronic medical record in a preset recent period of time, segment the medical record data, and obtain a number of phrases; obtain the medical staff's daily working hours in the preset recent period of time, the entry time of each phrase, the voice signal when entering each electronic medical record, and the pronunciation rate and intonation characteristics of the medical staff at each moment in each voice signal.

[0022] It should be noted that the main purpose of this embodiment is to determine the priority of phrase arrangement when medical staff enter electronic medical record information, and then construct a list of candidate phrases for medical staff when entering electronic medical record information, thereby improving the efficiency of medical staff in entering electronic medical record information; relevant data is first obtained before starting analysis.

[0023] Specifically, a number of medical record data entered by medical staff into the electronic medical record in a preset recent period of time is obtained, and the medical record data is segmented to obtain a number of phrases, as follows: Through the device disclosed in patent CN221079634U, all medical record data entered by medical staff into electronic medical records within the past week are counted; it should be noted that each piece of medical record data is specifically input data once by medical staff when entering medical records, for example, when entering the name of a disease, the entered disease name is a piece of medical record data; Jieba segmentation is performed on each piece of medical record data to obtain several phrases of each piece of medical record data, and all phrases after Jieba segmentation of all medical record data are obtained; it should be noted that Jieba segmentation is performed on medical record data to obtain several phrases of medical record data is an existing method and will not be repeated in this embodiment.

[0024] Furthermore, the daily working hours of medical staff in a preset recent period of time, the entry time of each phrase, the voice signal when entering each electronic medical record, and the pronunciation rate and intonation characteristics of the medical staff at each moment in each voice signal are obtained; it should be noted that obtaining the entry time of each phrase and the voice signal when entering each electronic medical record is an existing method, which will not be repeated in this embodiment; CN104735461B discloses a method for obtaining pronunciation rate and intonation characteristics. This embodiment obtains the pronunciation rate and intonation characteristics of medical staff at each moment in the voice signal through this method, and the specific details will not be repeated or limited.

[0025] At this point, several phrases are obtained, and the input duration of each phrase and the voice signal when each electronic medical record is input are obtained.

[0026] Step S002: Obtain the importance of each phrase based on the frequency and distribution of the phrases in the electronic medical records; obtain the number of different errors in each phrase in all electronic medical records; obtain the distribution of each error based on the number of errors, and obtain the comprehensive error rate of each phrase based on the distribution of the errors; obtain the ease of entry of each phrase based on the entry time and comprehensive error rate of the phrase.

[0027] It should be noted that the clinical manifestations of Alzheimer's patients at different stages of the disease are also different. For example, early patients have more obvious memory and judgment, and as the disease worsens, patients will gradually show symptoms such as inability to take care of themselves, limb stiffness, and even coma. The manifestations of different patients are also described differently in the electronic medical records. Therefore, in order to better determine the priority of the disease phrases when arranging them, it is necessary to first determine the importance of each phrase. The more important the phrase, the higher the priority.

[0028] Since the importance of a phrase is mainly related to the frequency of occurrence of the phrase in the electronic medical record and its distribution in multiple electronic medical records, the greater the frequency of occurrence and the more uniform the distribution in multiple electronic medical records, the more frequently the phrase appears and the more important the phrase is. Therefore, the importance of each phrase is obtained by analyzing the frequency of occurrence and distribution of the phrase in the electronic medical record.

[0029] Preferably, the importance of each phrase is obtained according to the frequency and distribution of the phrase in the electronic medical record, as follows: Any phrase is recorded as a target phrase; the total number of phrases and the number of target phrase occurrences obtained after word segmentation of all medical record data are obtained, and the ratio of the number of target phrase occurrences to the total number of phrases is used as the frequency of occurrence of the target phrase in the electronic medical record; the inverse proportional value of the standard deviation of the number of target phrase occurrences in all electronic medical records is used as the distribution uniformity of the target phrase in the electronic medical record; the occurrence frequency and the distribution uniformity are merged to obtain the importance of the target phrase.

[0030] As a specific example, the specific method for obtaining the importance is as follows: In the formula, For the The frequency of occurrence of a phrase in the electronic medical record is obtained by obtaining the total number of phrases obtained after word segmentation of all medical record data and the The number of occurrences of the phrase The ratio of the number of occurrences of a phrase to the total number of phrases is taken as the The frequency of occurrence of the phrase in the electronic medical record; For the The distribution uniformity of the phrases in the electronic medical record is obtained by: obtaining the first The number of occurrences of the phrase in all electronic medical records The inverse proportional value of the standard deviation of the number of occurrences of the first phrase is taken as the The distribution uniformity of the phrases in the electronic medical record. It should be noted that in this embodiment, The model shows the inverse proportional relationship. As the input of the model, the specific implementation can be set to other inverse proportional functions; For the The importance of a phrase.

[0031] It should be noted that the importance of a phrase is mainly related to its frequency of occurrence in the electronic medical record and its distribution in multiple electronic medical records. The greater the frequency of occurrence and the more evenly distributed in multiple electronic medical records, the more important the phrase is. Therefore, by analyzing the first The frequency of occurrence of the phrases in the electronic medical records and The distribution uniformity of the phrases in the electronic medical records is combined to obtain the The importance of a phrase.

[0032] It should be noted that in order to better obtain the priority of phrases when arranging, it is necessary to first obtain the number of different errors in each phrase in the electronic medical record for better subsequent analysis.

[0033] Specifically, the number of different errors that occur in each phrase in all electronic medical records is obtained; it should be noted that different errors that occur in phrases include typos, extra characters, missing characters, etc. Obtaining the number of different errors that occur in each phrase in all electronic medical records is an existing method and will not be repeated in this embodiment.

[0034] It should be noted that the priority of a phrase during arrangement is not only related to the importance of the phrase, but also to the ease of entering the phrase. If the phrase is highly important and difficult to enter, the phrase will have a higher priority during arrangement so that medical staff can enter information better. The ease of entering a phrase is mainly related to the errors that occur in the phrase. When a phrase often has errors during entry, that is, there are more errors in the electronic medical record, which means that the phrase is more difficult to enter; therefore, the comprehensive error rate of each phrase is obtained through the distribution of errors.

[0035] Preferably, the distribution of each error is obtained according to the number of errors, and the comprehensive error rate of each phrase is obtained according to the distribution of the errors, as follows: The ratio of the number of times each error occurs in the target phrase to the total number of times the error occurs in the target phrase is used as the distribution of each error in the target phrase; important factors for each error in the target phrase are preset; the distribution of each error in the target phrase and the important factors are integrated to obtain the initial comprehensive error rate for each error in the target phrase; the initial comprehensive error rates of all types of errors in the target phrase are accumulated and normalized to obtain the comprehensive error rate of the target phrase.

[0036] As a specific example, the specific method for obtaining the error rate is as follows: In the formula, For the Different types of errors when errors occur in a phrase; For the The phrase appears It should be noted that, in this embodiment, the preset important factor for each error in each phrase is obtained by manual evaluation, and the value range of the preset important factor is ; For the The phrase appears The number of errors; For the The total number of errors in the phrase; is a linear normalization function; For the The comprehensive error rate of phrases; it should be noted that if the There is no error in the phrase, then , to prevent the denominator from being unable to calculate.

[0037] It should be noted that since the error rate of a phrase is mainly related to the distribution of the number of times different errors occur in the phrase, The proportion of each error in each phrase is combined with the preset important factors to obtain the The comprehensive error rate of the phrases. When a certain error is distributed more frequently, that is, it accounts for a large proportion, the severity of the error in the phrase is evaluated, that is, the important factors are preset, and the first The combined error rate of the phrases.

[0038] It should be noted that the above analysis analyzes the comprehensive error rate of the phrase. The ease of inputting a phrase is not only related to the comprehensive error rate, but also to the time taken to input the phrase. When the comprehensive error rate is low and the time taken to input the phrase is short, it means that the phrase is easier to input. Therefore, the ease of inputting the phrase is obtained based on the comprehensive error rate analyzed above and the time taken to input the phrase.

[0039] Preferably, the ease of input of each phrase is obtained according to the input time and comprehensive error rate of the phrase, as follows: The average time taken by medical staff to enter the target phrase in all electronic medical records is obtained; the average time taken to enter the target phrase and the comprehensive error rate of the target phrase are combined to obtain the ease of entry of the target phrase.

[0040] As a specific example, the specific method for obtaining the ease of input is as follows: In the formula, Provide medical staff with the first The average duration of the phrases; For the The overall error rate of the phrases; For the The ease of input of a phrase; it should be noted that if the If the combined error rate of the phrase is 0, add 1 to the denominator for calculation purposes.

[0041] It should be noted that the ease of inputting a phrase is not only related to the overall error rate, but also to the time it takes to input the phrase. The average duration of the phrases and The comprehensive error rate of the phrases is obtained The ease of input of each phrase, taking into account the deviation in the time it takes for medical staff to input the same phrase, the average of the input time is taken here to reduce the impact of different time lengths. When the overall error rate is low, the shorter the time it takes to input the phrase, the easier it is to input the phrase, and the easier it is to input the phrase.

[0042] At this point, the ease of input of each phrase is obtained.

[0043] Step S003: obtain the first working status of the medical staff at each moment in each voice signal according to the working hours of the medical staff; obtain the second working status of the medical staff at each moment in each voice signal according to the pronunciation rate and the size of the intonation characteristics; obtain the comprehensive working status of the medical staff at each moment in each voice signal according to the first working status and the second working status of the medical staff at the same moment.

[0044] It should be noted that the ease of inputting phrases is also affected by the working status of medical staff. Since medical staff need to input a lot of electronic medical records in outpatient clinics, when the working status of medical staff is poor, it is easy to make mistakes when inputting phrases, and the ease of inputting phrases is low. Therefore, the working status of medical staff needs to be considered to reduce the impact of poor working status of medical staff on the priority of phrase arrangement. By analyzing the working hours of medical staff, the working status of medical staff is obtained. The longer the working hours of medical staff, the worse the working status.

[0045] Preferably, according to the working hours of the medical staff, the first working state of the medical staff at each moment in each voice signal is obtained, as follows: Any voice signal is recorded as a target voice signal; any time in the target voice signal is recorded as a target moment; the working day to which the target moment belongs is recorded as a target working day; the inverse proportional value of the working hours of the medical staff on the target working day up to the target moment is taken as the first working state of the medical staff at the target moment.

[0046] As a specific example, the specific method for obtaining the first working state is: Any time in any voice signal is recorded as the target time; the working day to which the target time belongs is recorded as the target working day.

[0047] In the formula, The working hours of medical staff on the target working day up to the target time; It is the first working state of medical staff at the target moment.

[0048] It should be noted that due to the easy input of phrases and the fact that they are also affected by the working status of medical staff, the first working status of medical staff at the target moment is obtained by analyzing the working hours of medical staff on the target working day up to the target moment. The longer the working hours, the worse the working status of the medical staff, that is, the smaller the first working status.

[0049] It should be noted that the first working status analyzed above is analyzed from the dimension of the working time of the medical staff. There is a large error in analyzing the working status of the staff only from the dimension of working time. Therefore, it is also necessary to analyze the voice signals of the medical staff at work. Since the working status of the medical staff is mainly related to the pronunciation rate and intonation characteristics at each moment in the voice signal, when the medical staff is in good working spirit, the speed of speaking is faster and the intonation characteristics are also greater. Therefore, the second working status of the medical staff is obtained by analyzing the changes in the pronunciation rate and intonation characteristics.

[0050] Specifically, according to the size of the pronunciation rate and intonation features, the second working state of the medical staff at each moment in each voice signal is obtained, as follows: If the pronunciation rate at the target moment in the target speech signal is less than the average of the pronunciation rates at all moments in the target speech signal, and the intonation feature at the target moment in the target speech signal is less than the average of the intonation features at all moments in the target speech signal, the product of the reciprocal of the difference between the average of the pronunciation rates at all moments and the pronunciation rate at the target moment, and the reciprocal of the difference between the average of the intonation features at all moments and the intonation features at the target moment is taken as the second working state of the medical staff at the target moment.

[0051] If the pronunciation rate at the target moment in the target speech signal is less than the average of the pronunciation rates at all moments in the target speech signal, and the intonation feature at the target moment in the target speech signal is greater than or equal to the average of the intonation feature at all moments in the target speech signal, the inverse of the difference between the average of the pronunciation rates at all moments and the pronunciation rate at the target moment is taken as the second working state of the medical staff at the target moment.

[0052] If the pronunciation rate at the target moment in the target speech signal is greater than or equal to the average pronunciation rate at all moments in the target speech signal, and the intonation feature at the target moment in the target speech signal is less than the average intonation feature at all moments in the target speech signal, the inverse of the difference between the average intonation feature at all moments and the intonation feature at the target moment is taken as the second working state of the medical staff at the target moment.

[0053] If the pronunciation rate at the target moment in the target speech signal is greater than or equal to the average pronunciation rate at all moments in the target speech signal, and the intonation feature at the target moment in the target speech signal is greater than or equal to the average intonation feature at all moments in the target speech signal, let the second working state of the medical staff at the target moment be 1.

[0054] It should be noted that if the pronunciation rate of the medical staff at the target moment is less than the average value, it means that the medical staff's work energy is poor, speaking is slow, and the corresponding second working state is also poor. Therefore, the second working state of the medical staff is reflected by the inverse of the difference between the average pronunciation rate of all moments and the pronunciation rate of the target moment; similarly, if the medical staff's intonation characteristics at the target moment are less than the average value, it also means that the medical staff's work energy is poor, and the corresponding second working state is poor. Therefore, the second working state of the medical staff is reflected by the inverse of the difference between the average intonation characteristics of all moments and the intonation characteristics of the target moment; when there is a situation greater than or equal to the mean, this embodiment directly sets the pronunciation rate or intonation characteristic part to 1, so as to better determine the second working state of the medical staff at the target moment. The second working state reflects the poor working state of the medical staff, that is, the larger the second working state, the worse the working state of the medical staff.

[0055] It should be noted that the above analysis of the working status of medical staff is carried out from the working time dimension and the voice signal dimension respectively. By combining the first working status and the second working status of the medical staff, a more accurate comprehensive working status is obtained to reduce the impact of the working status on the ease of input of the phrase.

[0056] Specifically, according to the first working state and the second working state of the medical staff at the same time, the comprehensive working state of the medical staff at each time in each voice signal is obtained, which is as follows: The product of the first working state and the second working state of the medical staff at the target moment is taken as the comprehensive working state of the medical staff at the target moment.

[0057] At this point, the comprehensive working status of the medical staff at each moment in each voice signal is obtained.

[0058] Step S004: According to the comprehensive working status of the medical staff at each moment and the proportion of errors in the phrases, the ease of entry is corrected to obtain the corrected ease of entry of each phrase; the priority of each phrase is obtained by combining the importance of the phrase and the corrected ease of entry; and a list of candidate phrases for medical staff to select when entering electronic medical record information is constructed according to the priority.

[0059] It should be noted that since the ease of input of a phrase will also be affected by the working status of the medical staff, the above analysis analyzes the comprehensive working status of the medical staff at each moment in each voice signal, and corrects the ease of input of the phrase based on the comprehensive working status. At the same time, the proportion of errors in the phrase is considered during the correction to make the correction result more accurate. Since the working status of this embodiment reflects the poor working status of the medical staff, the greater the comprehensive working status, the worse the status of the medical staff, and the lower the ease of input of the phrase. If there are still many errors in the phrase at the same time, it also means that the worse the status of the medical staff, the lower the ease of input of the phrase. By combining the comprehensive working status of the medical staff at each moment and the proportion of errors in the phrase, the corrected ease of input of each phrase is obtained.

[0060] Preferably, the ease of input is corrected according to the comprehensive working status of the medical staff at each moment and the proportion of errors in the phrases, and the corrected ease of input of each phrase is obtained, as follows: The comprehensive working status of the medical staff at the target moment is obtained; the ratio of the number of errors in the target phrase in the target voice signal up to the target moment to the total number of errors in the target phrase in the target voice signal in the corresponding electronic medical record is used as the proportion of errors in the target phrase; the inverse proportional value of the product of the comprehensive working status of the medical staff at the target moment and the proportion of errors in the target phrase is used as the comprehensive correction factor at the target moment in the target voice signal, and the ease of input of the target phrase is weightedly corrected according to the accumulated value of the comprehensive correction factor at all moments in all voice signals to obtain the corrected ease of input of the target phrase.

[0061] As a specific example, the specific method for obtaining the ease of input of the modified phrase is as follows: In the formula, For the The ease of inputting the phrase; is the number of speech signals; is the number of moments in the speech signal; For the The first The comprehensive working status of medical staff at each moment; For the The speech signal ends at At that moment, the first The number of errors in the phrase; For the The voice signal is in the corresponding electronic medical record The total number of errors in the phrase; is a linear normalization function; For the The ease of input of the phrase after correction.

[0062] It should be noted that, since the working status of this embodiment reflects the poor working status of the medical staff, the greater the comprehensive working status, the worse the status of the medical staff, and the lower the ease of input after the phrase is corrected, so In the denominator, there are still many errors in the phrases at the same time, that is, The larger it is, the worse the condition of the medical staff is, and the lower the ease of input of the phrase after correction is. Considering multiple recent voice signals, the ease of input of the phrase after correction is obtained.

[0063] It should be noted that the above-mentioned corrected ease of input of each phrase is obtained. By combining the corrected ease of input with the importance of the phrase, the priority of each phrase in the arrangement can be obtained, thereby facilitating the input of electronic medical record information.

[0064] Specifically, the priority of each phrase is obtained by combining the importance of the phrase and the corrected ease of input, as follows: For any phrase, the importance of the phrase and the inverse proportional value of the corrected ease of input are combined to obtain the priority of the phrase.

[0065] As a specific example, the specific method for obtaining the priority is as follows: In the formula, For the The importance of each phrase; For the The ease of input of the modified phrases; is an exponential function with a natural constant as base; For the The priority of each phrase.

[0066] It should be noted that the easier it is to enter a phrase after correction, the easier it is to enter the phrase, and its priority should be lower, and it should be ranked later. The model is used to realize the inverse proportional relationship; the greater the importance of a phrase, the more important the phrase is, and its priority should be greater, and it should be ranked at the front. Therefore, the priority of each phrase is obtained through the above.

[0067] It should be noted that the priority of each phrase is obtained above, and a list of phrases to be selected by medical staff when entering electronic medical record information can be constructed based on the priority of each phrase.

[0068] Furthermore, a list of phrases to be selected by medical staff when entering electronic medical record information is constructed according to the priority level, as follows: The priority of each phrase is obtained, and they are arranged in descending order of priority, so as to construct a list of phrases to be selected by medical staff when entering electronic medical record information. It should be noted that the list of phrases to be selected is subsequently optimized regularly to ensure that the phrases in the list of phrases to be selected can be well updated and optimized, so as to facilitate medical staff to enter electronic medical record information.

[0069] Through the above steps, the method for entering electronic medical record information for Alzheimer's disease is completed.

[0070] The embodiment of the present invention further provides a system for entering electronic medical record information of Alzheimer's disease, the system comprising a memory and a processor, and when the processor executes a computer program stored in the memory, the processor performs the following operations: Obtain a number of medical record data when medical staff enter electronic medical records in a preset recent period of time, segment the medical record data, and obtain a number of phrases; obtain the daily working hours of medical staff in the preset recent period of time, the entry time of each phrase, the voice signal when entering each electronic medical record, and the pronunciation rate and intonation characteristics of the medical staff at each moment in each voice signal; obtain the importance of each phrase based on the frequency and distribution of the phrase in the electronic medical record; obtain the number of different errors in each phrase in all electronic medical records; obtain the distribution of each error based on the number of errors, and obtain the comprehensive error rate of each phrase based on the distribution of the errors; obtain the ease of entry of each phrase based on the entry time and comprehensive error rate of the phrase According to the working hours of the medical staff, the first working state of the medical staff at each moment in each voice signal is obtained; according to the size of the pronunciation rate and intonation features, the second working state of the medical staff at each moment in each voice signal is obtained; according to the first working state and the second working state of the medical staff at the same moment, the comprehensive working state of the medical staff at each moment in each voice signal is obtained; according to the comprehensive working state of the medical staff at each moment and the proportion of errors in the phrases, the ease of input is corrected to obtain the corrected ease of input of each phrase; combining the importance of the phrase and the corrected ease of input, the priority of each phrase is obtained; according to the size of the priority, a list of candidate phrases for medical staff when entering electronic medical record information is constructed.

[0071] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for entering electronic medical record information of Alzheimer's disease, characterized in that: The method comprises the following steps: Obtain a number of medical record data when medical staff enter electronic medical records in a preset recent period of time, segment the medical record data, and obtain a number of phrases; obtain the medical staff's daily working hours in the preset recent period of time, the entry time of each phrase, the voice signal when entering each electronic medical record, and the pronunciation rate and intonation characteristics of the medical staff at each moment in each voice signal; According to the frequency and distribution of the phrases in the electronic medical records, the importance of each phrase is obtained; the number of different errors in each phrase in all electronic medical records is obtained; the distribution of each error is obtained according to the number of errors, and the comprehensive error rate of each phrase is obtained according to the distribution of errors; the ease of entry of each phrase is obtained according to the entry time and comprehensive error rate of the phrase; According to the working hours of the medical staff, the first working state of the medical staff at each moment in each voice signal is obtained; according to the size of the pronunciation rate and intonation characteristics, the second working state of the medical staff at each moment in each voice signal is obtained; according to the first working state and the second working state of the medical staff at the same moment, the comprehensive working state of the medical staff at each moment in each voice signal is obtained; According to the comprehensive working status of medical staff at each moment and the proportion of errors in phrases, the ease of entry is corrected to obtain the corrected ease of entry of each phrase; according to the importance of each phrase and the corrected ease of entry, the priority of each phrase is obtained; according to the priority, a list of phrases to be selected by medical staff when entering electronic medical record information is constructed; The specific steps of obtaining the importance of each phrase are as follows: Any phrase is recorded as the target phrase; the total number of phrases obtained after word segmentation of all medical record data and the number of occurrences of the target phrase are obtained, and the ratio of the number of occurrences of the target phrase to the total number of phrases is used as the frequency of occurrence of the target phrase in the electronic medical record; the inverse proportional value of the standard deviation of the number of occurrences of the target phrase in all electronic medical records is used as the distribution uniformity of the target phrase in the electronic medical record; the occurrence frequency and the distribution uniformity are combined to obtain the importance of the target phrase; the importance is expressed by the formula: In the formula, For the The frequency of occurrence of the phrase in the electronic medical record, For the The distribution uniformity of the phrases in the electronic medical record; The specific steps for obtaining the ease of input are as follows: Any phrase is recorded as the target phrase; the average time taken by medical staff to enter the target phrase in all electronic medical records is obtained; the average time taken by the target phrase and the comprehensive error rate of the target phrase are combined to obtain the ease of entry of the target phrase; the ease of entry is expressed by the formula: In the formula, Provide medical staff with the first The average duration of the phrases; For the The overall error rate of the phrases; For the The ease of inputting the phrase; The specific steps for obtaining the comprehensive working status are as follows: Record any speech signal as the target speech signal; record any time in the target speech signal as the target moment; The product of the first working state and the second working state of the medical staff at the target moment is used as the comprehensive working state of the medical staff at the target moment; The specific steps of obtaining the priority of each phrase are as follows: Any phrase is recorded as a target phrase; the corrected ease of entry of the target phrase is obtained, and the corrected ease of entry of the target phrase is input into a negative exponential function with a natural constant as the base; the product of the output result and the importance of the target phrase is recorded as the priority of the target phrase.

2. The method for entering electronic medical record information of Alzheimer's disease according to claim 1, characterized in that: The specific steps of obtaining the distribution of each error according to the number of errors and obtaining the comprehensive error rate of each phrase according to the distribution of the errors are as follows: Any phrase is recorded as the target phrase; the ratio of the number of times each error occurs in the target phrase to the total number of times each error occurs in the target phrase is recorded as the distribution of each error in the target phrase; The important factors of each error in the target phrase are preset; the distribution of each error in the target phrase and the important factors are integrated to obtain the initial comprehensive error rate of each error in the target phrase; the initial comprehensive error rates of all types of errors in the target phrase are accumulated and normalized to obtain the comprehensive error rate of the target phrase; the comprehensive error rate is expressed by the formula: In the formula, For the The overall error rate of the phrases; For the Different types of errors when errors occur in a phrase; For the The phrase appears erroneous assumptions about the importance of factors; For the The phrase appears The number of errors; For the The total number of errors in the phrase; For the The phrase appears The distribution of errors; For the The phrase appears The initial combined error rate of the errors; is a linear normalization function.

3. The method for entering electronic medical record information of Alzheimer's disease according to claim 1, characterized in that: The specific steps of obtaining the first working state of the medical staff at each moment in each voice signal according to the working hours of the medical staff are as follows: The working day to which the target time belongs is recorded as the target working day; the inverse proportional value of the working hours of the medical staff on the target working day up to the target time is taken as the first working status of the medical staff at the target time.

4. The method for entering electronic medical record information of Alzheimer's disease according to claim 1, characterized in that: The method of obtaining the second working state of the medical staff at each moment in each voice signal according to the pronunciation rate and the size of the intonation feature includes the following specific steps: If the pronunciation rate at the target moment in the target speech signal is less than the average of the pronunciation rates at all moments in the target speech signal, and the intonation feature at the target moment in the target speech signal is less than the average of the intonation features at all moments in the target speech signal, the product of the reciprocal of the difference between the average of the pronunciation rates at all moments and the pronunciation rate at the target moment and the reciprocal of the difference between the average of the intonation features at all moments and the intonation features at the target moment is taken as the second working state of the medical staff at the target moment; If the pronunciation rate at the target moment in the target speech signal is less than the average of the pronunciation rates at all moments in the target speech signal, and the intonation feature at the target moment in the target speech signal is greater than or equal to the average of the intonation feature at all moments in the target speech signal, the reciprocal of the difference between the average of the pronunciation rates at all moments and the pronunciation rate at the target moment is taken as the second working state of the medical staff at the target moment; If the pronunciation rate at the target moment in the target speech signal is greater than or equal to the average of the pronunciation rates at all moments in the target speech signal, and the intonation feature at the target moment in the target speech signal is less than the average of the intonation features at all moments in the target speech signal, the reciprocal of the difference between the average of the intonation features at all moments and the intonation features at the target moment is taken as the second working state of the medical staff at the target moment; If the pronunciation rate at the target moment in the target speech signal is greater than or equal to the average pronunciation rate at all moments in the target speech signal, and the intonation feature at the target moment in the target speech signal is greater than or equal to the average intonation feature at all moments in the target speech signal, let the second working state of the medical staff at the target moment be 1.

5. The method for entering electronic medical record information of Alzheimer's disease according to claim 1, characterized in that: The step of correcting the ease of input according to the comprehensive working status of the medical staff at each moment and the proportion of errors in the phrases to obtain the corrected ease of input of each phrase includes the following specific steps: Record any phrase as a target phrase; record any voice signal as a target voice signal; record any time in the target voice signal as a target moment; obtain the comprehensive working status of the medical staff at the target moment; The ratio of the number of times the target phrase has errors in the target speech signal up to the target moment to the total number of times the target phrase has errors in the target speech signal in the corresponding electronic medical record is used as the error ratio of the target phrase; the inverse proportional value of the product of the comprehensive working status of the medical staff at the target moment and the error ratio of the target phrase is used as the comprehensive correction factor at the target moment in the target speech signal, and the ease of input of the target phrase is weightedly corrected according to the accumulated value of the comprehensive correction factor at all moments in all speech signals to obtain the corrected ease of input of the target phrase.

6. A system for entering electronic medical record information of Alzheimer's disease, the system comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the computer program is executed by a processor, the steps of the method for entering electronic medical record information of Alzheimer's disease as described in any one of claims 1 to 5 are implemented.

Images