JP2015194974A - Home/visit medical service system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a home/visit medical service system for supporting visit medical care.SOLUTION: The home/visit medical service system extracts plural text data to generate a piece of medical chart information, a piece of medical service fee information and a piece of doctor knowledge information by using a first identifier for identifying items of the medical chart information, and a second identifier for identifying items of the medical service fee information and a third identifier for identifying knowledge of a doctor who performs the medical care of a patient from medical care text data in which medical care voice data at visit homes are recorded in a text form. With this, a system to store (pick up) doctor knowledge texts based on doctors experiences and know-how other than the medical chart information and the medical service fee information can be established, and a circumstance for obtaining a medical care support using the experiences and know-how of plural doctors is achieved.

Description

  Embodiments described herein relate generally to a technique for supporting visiting medical care.

  Conventionally, as described in Patent Document 1, there is a technique for creating and storing an electronic medical record by performing speech processing by recognizing text data of speech data of a doctor who is undergoing medical treatment. At this time, in Patent Document 1, a receipt (medical fee data) is automatically created from the electronic medical record.

JP 2005-267358 A Japanese Patent No. 489894 Japanese Patent No. 4197271

  Doctor's voice data in visiting medical care is converted into text, doctor knowledge text based on doctor's experience other than medical record information and medical fee information is extracted from the textualized medical treatment contents, and the experience and know-how of multiple doctors are utilized Provide a home / visit medical service system that can realize an environment for providing medical assistance.

  The home / visit medical service system according to the embodiment performs text processing on medical speech data input by speech at a visited site, generates a medical text data, and a first for identifying each item of medical record information A first storage unit for storing an identifier, a second identifier for identifying each item of medical fee information, and a third identifier for identifying knowledge of a doctor who treats a patient; and the medical text data includes the first storage unit Based on the first identifier and the second identifier, a medical record / receipt processing unit that generates medical record text data corresponding to the medical record information and medical remuneration text data corresponding to the medical remuneration information from the medical text data, Based on the third identifier included in the medical text data, the doctor corresponding to the doctor knowledge information from the medical text data It includes a knowledge data processor for generating an identification text data, wherein the the medical voice data, together with the medical record information and the medical fee information, extracts the respective text data to generate the physician knowledge information.

1 is a system configuration diagram of a home / visit medical service according to a first embodiment. FIG. 1 is a configuration block diagram of a mobile terminal device, a management device, and a medical care system of a first embodiment. It is a figure which shows the example of a process which produces | generates each data of a medical chart text, a receipt text, and doctor knowledge text from the text-ized medical treatment content (medical practice text) of 1st Embodiment. It is a figure which shows the medical assistance data generation example for every patient using the medical statistics data of 1st Embodiment. It is a figure which shows an example of the determination result pattern and case occurrence probability data of 1st Embodiment. (A) is an onset condition result pattern, (b) is a prevention condition result pattern, and (c) is an example of occurrence probability data of aspiration pneumonia based on a determination result. It is a figure which shows the example of a matching with the determination reference | standard of each determination of medical statistical data and knowledge data of 1st Embodiment, and a patient's chart information. It is a figure which shows the process example which produces | generates knowledge data from the doctor knowledge text of 1st Embodiment. It is the flowchart which showed the processing flow of the home and visiting medical service system of 1st Embodiment. It is the flowchart which showed the processing flow of the home and visiting medical service system of 1st Embodiment.

  Hereinafter, embodiments will be described with reference to the drawings.

  The home / visit medical service system according to the embodiment is a system that supports a visit medical treatment in which a doctor visits a residence, a facility, or a hospital. As shown in FIG. 1, for example, a doctor (for example, a dentist) visits a patient's home, a nursing home, a care facility, another hospital different from the hospital where the doctor is in full duty, and the doctor is in each place. The patient is treated (for example, dental practice).

  At this time, the doctor can visit the patient with the portable terminal device 100. The mobile terminal device 100 can be connected to a medical care system 500 that manages and provides a patient's electronic medical record, and can confirm the patient's medical record information on the mobile terminal device 100.

  When a doctor treats a patient, he / she inputs medical treatment contents and care contents to the mobile terminal device 100 by voice. The medical voice data input to the mobile terminal device 100 is transmitted to the management device (home / visit medical service support device) 300. The management apparatus 300 analyzes medical voice data and converts it into text. Each data of the medical record text, the receipt text, and the doctor knowledge text is extracted from the medicalized medical treatment contents (medical treatment text).

  Here, in the present embodiment, the doctor's knowledge and experience associated with the medical treatment are individually extracted from the medical treatment text as a doctor knowledge text (doctor knowledge data). By accumulating and analyzing doctor knowledge texts and performing statistical processing, doctor knowledge data based on how symptoms are improved and worsened, and the effects of treatment and prevention are generated.

  Conventionally, there has been a technology for generating electronic medical records from audio data, but the point of view is to facilitate receipts (medical fees) and not to provide medical assistance by accumulating doctor experience and know-how. It was. In addition, even if doctor's experience and know-how were included in the chart information, it was managed on a patient-by-patient basis, so only “knowledge” for a specific patient could be grasped, and the experience and know-how of multiple doctors were utilized. I was unable to provide medical assistance.

  Therefore, in the present embodiment, first, a mechanism for accumulating (sucking up) doctor knowledge texts based on medical record information and doctor experiences and know-how other than receipts from the medical texts, and utilizing the experiences and know-how of multiple doctors is provided. Provide an environment for providing support. Then, through the creation of knowledge data using the accumulated doctor knowledge text, symptoms that will occur in the future, their occurrence probability, treatment method, occurrence probability of symptoms after treatment, and the like are generated as medical assistance data. That is, each doctor himself / herself generates “knowledge” of a medical condition for each patient from doctor knowledge texts accumulated know-how through medical treatment and the like, and provides medical assistance data using the doctor's “knowledge”. The medical assistance data is stored as information accompanying the medical chart information. Each doctor can check current medical record information and medical assistance data, and can diagnose current symptoms of each patient, develop a medical treatment policy, etc., and alert the patient.

  FIG. 2 is a configuration block diagram of the mobile terminal device 100, the management device 300, and the medical care system 500.

  The portable terminal device 100 is a portable computer device by a doctor or the like. The information terminal device 100 includes, for example, a portable information terminal such as a multi-function mobile phone or a PDA (Personal Digital Assistant) and a mobile information terminal having a calculation function (CPU or the like), a notebook computer, a tablet computer, or the like.

  The mobile terminal device 100 includes a communication unit 101, a control unit 102, a display unit 103, and a storage unit 104. The communication unit 101 is a communication control device that provides a data communication function via a wireless communication network such as a public wireless LAN. The mobile terminal device 100 can connect to a wireless communication network by wireless communication through a predetermined access point through the communication unit 101 and perform data communication with the management device 300 and the medical care system 500. The communication unit 101 can also be connected to a communication network such as a wired LAN. In this case, the mobile terminal device 100 can include a connection interface such as a LAN cable.

  The control unit 102 performs display control for displaying a screen provided by the management apparatus 300 or the medical treatment system 500 on the display unit 103. Further, when the display unit 103 has an operation input function such as a touch panel, the control unit 102 can perform display control of various information and screens based on user operations input from the display unit 103. Display control can be performed via a browser or a predetermined screen, for example. When the portable terminal device 100 includes an input key and an input button separately from the display unit 103, the control unit 102 can perform display control according to a user operation via the input key or the like.

  The mobile terminal device 100 according to the present embodiment further includes an audio input unit 105 and an audio data generation unit 106. The voice input unit 105 is a microphone (sound collecting device). The audio data generation unit 106 digitizes the audio input through the audio input unit 105 in a predetermined audio file format and generates audio data. The control unit 102 controls the voice input unit 105 and the voice data generation unit 106.

  The management apparatus 300 includes a control unit 310, a storage unit 320, and a communication control unit 330. The communication control unit 330 controls data communication between the mobile terminal device 100 and the medical care system 500 via a communication network. The management device 300 is a server device, for example.

  The control unit 310 includes a voice recognition unit 311, a medical record / receipt processing unit 312, a knowledge data processing unit 313, and a knowledge extraction unit 314. The voice recognition unit 311 converts voice data of medical treatment contents output from the mobile terminal device 100 to text (character information) by applying voice recognition technology. A known method can be applied to the speech recognition technology (text processing). Detailed description is omitted.

  FIG. 3 is a diagram showing a processing example for generating each data of a medical record text, a receipt text, and a doctor knowledge text from the medicalized medical contents. Note that the textualized medical contents are hereinafter referred to as “medical text”.

  When a doctor who visits a patient starts medical care, an operation for starting medical care for the mobile terminal device 100 (for example, obtains patient chart information from the medical care system 500 and selects a medical care start button) To do). Note that it is possible to automatically select a medical treatment start button by voice input. For example, when the doctor speaks “start of medical care”, the control unit 102 can control the voice input unit 105 and the voice data generation unit 106 so as to start collection of medical voice data.

  As shown in the example of FIG. 3, doctors are divided into each medical treatment category of chief complaint (S), findings (O), screening (A1), screening results (A2), diagnostic content (A3), and treatment plan (P). Then, the contents of the examination can be input by voice. For example, the doctor describes the medical treatment category “main complaint” at the beginning of the utterance, utters the symptoms heard from the patient, and inputs them to the mobile terminal device 100.

  In addition, the mobile terminal device 100 determines each of the main complaint (S), findings (O), examination (A1), examination result (A2), diagnosis contents (A3), and treatment plan (P) according to these medical treatment categories. A selection button can be displayed on the display unit 103. After selecting the “main complaint” button displayed on the display unit 103, the doctor can utter a symptom heard from the patient and perform voice input.

  In the present embodiment, a medical treatment division (identifier) is set in advance in consideration of collected information or information to be collected for voice input during medical treatment by a doctor. With this configuration, the voice recognition unit 311 can identify and extract each piece of information corresponding to a medical treatment category from the medical treatment content that has been voiced and generate medical text.

  The voice recognition unit 311 can perform a cleansing process on a text sentence generated with reference to the analysis dictionary group 325. At this time, the analysis dictionary group 325 can include a dictionary that unifies a plurality of synonyms into one term, a dictionary for removing unnecessary sentences and phrases (such as stagnation, companion, polite language), and the like. The speech recognition unit 311 can refer to the analysis dictionary group 325 and perform data cleansing processing to unify terms and remove unnecessary sentences and phrases when text or text after text conversion is converted. it can.

  In addition, the text sentences in the medical text are sentences corresponding to the medical departments as described above, and in each of these sentences, the terms and phrases used for each medical department are often determined in advance. Therefore, the analysis dictionary group 325 can include a dictionary that defines text analysis rules for each medical treatment category. The voice recognition unit 311 can generate a text sentence of the medical treatment category “main complaint” with reference to the dictionary corresponding to the medical treatment category “main complaint”.

  In the text processing by the voice recognition unit 311, the accuracy of text conversion of medical voice data can be improved by using a dedicated dictionary for each text sentence corresponding to the medical classification or removing unnecessary sentences and phrases. it can.

  In addition, as shown in FIG. 3, it can also comprise so that the information division may be further included in one medical treatment division. In the example of FIG. 3, information categories such as “treatment results”, “state change”, “explanation results”, and “comments” are included in the medical treatment category “treatment plan”. The information section is also used as an identifier included in the medical text, like the medical section, so that each piece of information corresponding to the information section can be identified and extracted from the speech-input medical contents.

  The above-described medical treatment categories can be set mainly according to medical treatment items and medical claim items for medical record information and receipt information. Information categories “treatment results”, “state changes”, etc. in the medical treatment category “treatment plan” can also be set based on medical record information and receipt information.

  On the other hand, the information categories “explanation record”, “comment” and the like are information input based on the medical experience and knowledge of a doctor regardless of the medical record information and the receipt information. By setting the classification for identifying the doctor knowledge text input based on the doctor's medical experience and knowledge, the doctor knowledge text other than the medical record information and the receipt information can be extracted from the medical treatment contents.

  The relation (relational) between the medical treatment section and the information section can be arbitrarily set. For example, in the example of FIG. 3, the information category related to the doctor knowledge text is set in the same row (same hierarchy) as one of the medical treatment categories, or is set as a division for further dividing information as a lower hierarchy within one medical treatment category. be able to.

  The medical text generated by the voice recognition unit 311 can be corrected by the operator from the information processing apparatus 400. The operator can check the medical text generated by the text processing of the voice recognition unit 311 and the operator can correct a mistake in the context, a typo, or a missing letter. Using the medical text corrected by the operator, it is possible to generate a medical record text and the like described below.

  In addition, the operator can update the analysis dictionary group 325 through the information processing apparatus 400. The operator can add to the text analysis rule or change the existing text analysis rule based on the correction contents of the medical text. Also, unnecessary sentences and phrases can be updated.

  The medical record / receipt processing unit 312 extracts the medical record text and the receipt text from the medical text processed into text by the speech recognition unit 311 using the analysis dictionary group 325.

  First, as described above, the medical treatment items and the like of the medical record information and the receipt information are associated with the medical treatment division in advance. The medical record / receipt processing unit 312 extracts one or more sentences associated with the medical treatment category “main complaint” from the medical text as sentences corresponding to the medical treatment item “main complaint” of the medical record text.

  Similarly, the medical record / receipt processing unit 312 converts one or a plurality of sentences associated with each of the medical treatment categories “findings”, “test results”, and “consideration (diagnosis content)” from the medical texts in the corresponding medical record text. Extracted as sentences corresponding to each medical item.

  At this time, the medical record / receipt processing unit 312 can convert the sentence corresponding to each medical treatment section into the data format of the medical treatment item of the medical record text using the analysis dictionary group 325.

  For example, in the examination result of the dental examination, each tooth is numbered in the order of upper right → upper left → lower left → lower right, etc., and the doctor says, “Upper right, No. 1 to No. 6, C1, No. 7 C3 "Read out." In the medical chart information, tooth numbers are listed, and it is necessary to enter the progress degree “C1” and “C3” of the caries in each tooth number. Therefore, the medical record / receipt processing unit 312 changes “from No. 1 to No. 6” to “Upper right No. 1: C1, upper right No. 2: C1, upper right No. 3: C1, upper right No. 4: C1, upper right No. 5: C1. , Upper right number 6: C1 ”.

  The analysis dictionary group 325 includes a dictionary in which an analysis rule that divides a sentence “from No. to No. □” into a plurality of pieces of data from No. 1 to No. 6 is defined, and the chart / receipt processing unit 312. Based on the analysis rule, the text sentence corresponding to the “test result” of the medical text can be converted into a format corresponding to the data format of the test result of the medical record information.

  The medical record / receipt processing unit 312 extracts one or more sentences associated with the medical examination category “examination (A1)” from the medical text as sentences (phrases) corresponding to the medical claim item “examination item” of the receipt text. To do. Also at this time, the analysis dictionary group 325 includes a dictionary in which a plurality of medical claim items are defined in advance, and the medical record / reception processing unit 312 refers to the analysis dictionary group 325 while identifying the corresponding text sentence. It is possible to extract words and the like necessary for reception from the text sentence through morphological analysis and the like.

  Similarly, the medical record / receipt processing unit 312 corresponds to one or a plurality of sentences associated with the medical treatment category “treatment results” from the medical text to the medical claim items “treatment item” and “prescription item” of the receipt text. Extracted as sentences (phrases). At this time, if the medical record / receipt processing unit 312 includes two medical claim items in one text sentence, for example, morphological analysis or the like is used to separate “calculus removal” and “XX medicine ○ ml medication”. Each text can be extracted as a corresponding medical claim item.

  In addition, in the analysis dictionary group 325, receipt related data can be stored in advance. For example, the storage unit 320 holds data in which a medical fee is specified for each medical claim item. The medical record / receipt processing unit 312 can automatically calculate a medical remuneration score for each medical claim item of the medical receipt text generated from the medical text and generate a receipt text.

  The receipt text can also be generated from the medical record text. That is, since the medical record that records the medical claim item is recorded in the medical record text, the medical record / receipt processing unit 312 generates the medical record text as described above or in parallel with the generation process of the medical record text. The receipt text can also be generated.

  The basic information (name, age, insurance type, etc.) of the patient can be registered in advance or can be input by a doctor from a predetermined screen through the mobile terminal device 100 during a visit medical treatment. In addition, patient basic information may be input by voice. In this case, for example, an information category such as “basic information” is provided, and the doctor speaks “basic information” first, and the basic information of the patient can be extracted from the text sentence associated with the “basic information”.

  The medical record / receipt processing unit 312 stores the generated medical record text and receipt text in the storage unit 320 and transmits them to the medical care system 500 to register the medical record information and the receipt information. As illustrated in FIG. 2, the medical care system 500 includes a control unit 510, a storage unit 520, and a communication control unit 530 that performs data communication with the mobile terminal device 100 and the management device 300.

  The control unit 510 receives the medical record text transmitted from the management device 300 (the medical record / reception processing unit 312) and registers it in the electronic medical record, and the management device 300 (the medical record / reception processing unit 312). And a receipt control unit 512 that receives the received receipt text and registers it in the receipt data.

  The electronic medical chart control unit 511 and the receipt control unit 512 have an approval function that the corresponding doctor approves the electronic medical record and the receipt data of the patient who has been treated through a computer device that can be connected to the portable terminal device 100 or the medical treatment system 500. Yes. The electronic medical chart control unit 511 provides the mobile terminal apparatus 100 with the electronic medical chart of the corresponding patient based on the electronic medical chart request from the mobile terminal apparatus 100.

  The knowledge data processing unit 313 extracts doctor knowledge text from the medical text. As described above, “explanation record”, “comment”, and the like can be set as information input based on the medical experience and knowledge of a doctor. The knowledge data processing unit 313 extracts, as doctor knowledge text, one or more sentences associated with the information categories “explanation record” and “comment” from the medical text. The extracted doctor knowledge text is stored in the storage unit 320.

  The sentence extracted as the doctor knowledge text may be extracted as a medical record text. That is, it is only necessary to extract the medical experience and knowledge of a doctor recorded as medical chart information as a doctor knowledge text. For example, a medical treatment item “state change” in the medical chart text can be extracted as a doctor knowledge text.

  And the doctor knowledge text of this embodiment is utilized for the production | generation of the medical assistance data extracted for every patient. The doctor knowledge text is information for centrally managing the experience and knowledge of the doctor according to the medical treatment contents for each patient, unlike the medical record information and the receipt information managed independently for each patient.

  FIG. 4 is a diagram illustrating an example of generating medical assistance data for each patient using medical statistical data. The occurrence probability of aspiration pneumonia based on medical statistical data will be described as an example. The storage unit 320 stores medical statistical data. The medical statistical data is, for example, medical data of a certain symptom (case) or death cause statistical data. In the example of FIG. 4, in order to determine the occurrence probability of aspiration pneumonia, a plurality of check sheets (determination blocks) of determinations A to F are provided in advance based on medical statistical data. These check sheets are symptom determination information for determining the risk of developing aspiration pneumonia, and are stored in the storage unit 320 as medical statistical data.

  Each check sheet of determination A to determination D is a check item for grasping the risk (onset condition) of causing aspiration pneumonia through the analysis of the current state of the patient's medical history and symptoms. Each check sheet of determination E and determination F is a check item for grasping prevention conditions for reducing the risk of developing aspiration pneumonia through preventive analysis of a patient's preventive behavior and preventive medical care (preventive treatment) by a doctor. In the present embodiment, onset conditions and prevention conditions for aspiration pneumonia are grasped from the patient's medical chart information, and medical support data for each patient is generated.

  Age is the risk of causing aspiration pneumonia. This corresponds to the check sheet for determination A. For example, the age at which the risk of developing aspiration pneumonia increases can be set in advance using medical statistical data that has developed aspiration pneumonia.

  There is a symptom as a risk corresponding to the check sheet of determination B. The medical condition is “It is difficult to swallow things”, “There is pain and pain when swallowed”, “Food spills from the mouth or remains in the mouth.”, “Drowsing”, “Thirsty. There are a plurality of check items such as “It takes time to eat.”, “Chews when eating, coughs.”, “Eating food flows backward.” Each item of the medical condition is previously extracted from past medical statistical data in analyzing the risk of aspiration pneumonia.

  There is a medical history as a risk corresponding to the check sheet of determination C. The medical history has multiple items such as cerebrovascular disorders (cerebral infarction, intracerebral hemorrhage), Parkinson's syndrome, Alzheimer's dementia, stroke, dysphagia, esophageal cancer, and care status. Each item of medical history is also extracted in advance from past medical statistical data in analyzing the risk of aspiration pneumonia.

  Moreover, there is nursing care as a risk corresponding to the check sheet of determination D. Whether or not a care recipient is required is previously extracted from past medical statistical data in analyzing the risk of aspiration pneumonia.

  These age, medical condition, medical history, and long-term care are check items for analyzing the current state of the patient as described above. It is possible to grasp whether or not aspiration pneumonia easily develops by performing a matching process between the patient's medical record information and the check item of the current state analysis.

  On the other hand, there are oral care and preventive care as prevention (prevention conditions) for reducing the risk of developing aspiration pneumonia. The oral care corresponding to the check sheet of judgment E includes a plurality of check items such as “daily tooth brushing”, “gum gum massage”, “salivary gland massage”, “eat training exercise”, and “continuation of oral cleaning”. Each check item of oral care is previously extracted from past medical statistical data in order to prevent aspiration pneumonia (to reduce the risk of developing aspiration pneumonia).

  Further, the preventive care corresponding to the check sheet of determination F includes a plurality of check items such as “food”, “how to eat”, “post-meal posture”, and “preventive drug”. Each check item of preventive care is also extracted in advance from past medical statistical data in order to prevent aspiration pneumonia.

  These oral care and preventive care are check items for performing preventive analysis of patients as described above. It is possible to grasp whether or not the risk of developing aspiration pneumonia can be reduced by performing matching processing between the patient's medical record information and the check items of the preventive analysis.

  In the present embodiment, as shown in FIG. 4, according to the results of the current analysis and the preventive analysis, the current probability of becoming aspiration pneumonia is generated as medical assistance data for each patient, and the treatment to be performed in the future The probability of occurrence of aspiration pneumonia that will decrease is generated as medical assistance data.

  The probability of aspiration pneumonia can be calculated based on, for example, each determination result from determination A to determination F. FIG. 5 is a diagram illustrating an example of a determination result pattern and case occurrence probability data. 5A is an example of the onset condition result pattern, FIG. 5B is an example of the prevention condition result pattern, and FIG. 5C is an example of aspiration pneumonia occurrence probability data based on the determination result.

  As shown in FIG. 5A, four determination result patterns of determination check sheets A to D are defined in advance as onset condition result patterns corresponding to the current analysis of aspiration pneumonia. Similarly, as shown in FIG. 5B, two determination result patterns of determination check sheets E and F are defined in advance as prevention condition result patterns corresponding to the prevention analysis of aspiration pneumonia.

  Then, as shown in FIG. 5 (c), for each combination of the onset condition result pattern and the prevention condition result pattern, the probability of occurrence of aspiration pneumonia after XX years is calculated in advance from medical statistical data. At this time, the occurrence probability of aspiration pneumonia is divided into two categories: the current state and after treatment, and a specific treatment method is associated in advance with the occurrence probability after treatment.

  The probability of occurrence of aspiration pneumonia according to the combination of determination result patterns increases as the number corresponding to each determination A to D increases, for example, in the onset condition result pattern. On the other hand, the greater the number corresponding to each determination E and F in the prevention condition result pattern, the lower the probability of occurrence.

  Therefore, in both the onset condition and the prevention condition, for example, as shown in FIG. 5 (c), the occurrence probability of aspiration pneumonia when it does not correspond to each determination of the onset condition is the lowest, and each determination of the prevention condition The more likely it is, the lower the probability of occurrence so that the risk of aspiration pneumonia is reduced. Such a combination pattern and occurrence probability of the determination result can be calculated from medical statistical data. That is, as described above, each determination of the determination check sheets A to F is extracted in advance based on past medical statistical data, and the occurrence probability of aspiration pneumonia is determined from the medical statistical data for each determination. Can be calculated. For this reason, the probability that aspiration pneumonia will occur for each combination pattern of determination results can be calculated from the medical statistical data using each occurrence probability for each determination.

  The occurrence probability after treatment indicates, for example, the occurrence probability when a predetermined treatment is performed with respect to the current occurrence probability. That is, the probability of occurrence when left untreated without predetermined treatment is the current probability of occurrence of aspiration pneumonia. Therefore, the medical statistics data includes the probability of occurrence in the past history when left untreated without predetermined treatment and the probability of occurrence with predetermined treatment, respectively. By processing, the probability of occurrence before and after treatment can be calculated for each combination pattern of the determination results.

  FIG. 6 is a diagram illustrating an example of matching between determination criteria for each determination of medical statistical data and knowledge data and patient chart information. The generation of the medical assistance data can be performed through the determination process using the check sheets A to F based on the patient's medical record information as described above. The medical assistance data generation process is performed by the knowledge extraction unit 314 and is performed automatically or at an arbitrary timing. For example, when the doctor refers to the medical record information via the mobile terminal device 100, when the doctor wants to refer to the medical assistance data, or at a specific timing when the knowledge data processing unit 313 generates the knowledge data. it can.

  The knowledge extraction unit 314 acquires medical record information of the corresponding patient from the medical care system 500 in accordance with the medical assistance data generation process. Based on the medical record information acquisition request from the knowledge extraction unit 314, the electronic medical record control unit 511 extracts the corresponding patient's electronic medical record 521 from the storage unit 520 and transmits the extracted electronic medical record 521 to the management apparatus 300.

  The knowledge extraction unit 314 matches the patient's medical chart information with the above-described determination check sheets. First, regarding the check sheet of determination A, the knowledge extraction unit 314 determines whether or not the patient's age is 65 years or older using “patient basic data (date of birth, age)” included in the medical record information. To do. When the patient's age is 65 years or older, the determination result of the determination A is “◯ (applicable)”, and when the patient is less than 65 years old, the determination result is “× (not applicable)” in the storage unit 320. Remember.

  Subsequently, regarding the check sheet for determination B, the knowledge extraction unit 314 corresponds to a plurality of check items for determination B using “diagnosis data (main complaint, findings, comments from family members, etc.)” included in the chart information. It is determined whether or not. At this time, the knowledge extraction unit 314 counts the number corresponding to the check item. For example, if the determination criterion based on the medical statistics data corresponds to five or more check items in the determination item B, If the determination result is “◯ (applicable)” and does not correspond to five or more check items, the determination result is stored in the storage unit 320 as “× (not applicable)”.

  The knowledge extraction unit 314 determines whether or not the check sheet of the determination C corresponds to a plurality of check items of the determination C by using “historical history (disease name, treatment period, outcome, etc.)” included in the chart information. To do. Also at this time, the knowledge extraction unit 314 counts the number corresponding to the check item of the determination C, for example, when the determination criterion based on the medical statistical data corresponds to two or more check items in the check item of the determination C When the determination result of determination C is “◯ (applicable)” and does not correspond to two or more check items, the determination result is stored as “× (not applicable)” in the storage unit 320.

  The knowledge extraction unit 314 determines whether or not the patient is in a care state, using the “care state” included in the chart information regarding the check sheet of determination D. When the patient is in the care state, the determination result of determination D is “◯ (applicable)”, and when the patient is not in the care state, the determination result is stored as “× (not applicable)” in the storage unit 320.

  The knowledge extraction unit 314 determines whether or not the check sheet of determination E corresponds to a plurality of check items of determination E, using “treatment history (treatment content, state change)” included in the chart information. Also at this time, the knowledge extraction unit 314 counts the number corresponding to the check item of the determination E, for example, when the determination criterion based on the medical statistical data corresponds to two or more check items in the check item of the determination E When the determination result of determination E is “◯ (applicable)” and does not correspond to two or more check items, the determination result is stored as “× (not applicable)” in the storage unit 320.

  The knowledge extraction unit 314 determines whether or not the check sheet of the determination F corresponds to a plurality of check items of the determination F by using “treatment (living environment, medications, and eating habits)” included in the medical record information. To do. Also at this time, the knowledge extraction unit 314 counts the number corresponding to the check item of the determination F, for example, when the determination criterion based on the medical statistical data corresponds to one or more check items in the check item of the determination F When the determination result of determination F is “◯ (applicable)” and does not correspond to one or more check items, the determination result is stored as “× (not applicable)” in the storage unit 320.

  In this manner, the knowledge extraction unit 314 matches the patient chart information and the determinations A to F of each determination check sheet, and grasps (calculates) each determination result. Then, the onset condition result pattern is specified from the onset condition result pattern shown in FIG. 5A, and the preventive condition result pattern is specified from the preventive condition result pattern shown in FIG. 5B. Based on the combination of the onset condition result pattern and the prevention condition result pattern, the knowledge extraction unit 314 refers to the occurrence probability data of aspiration pneumonia based on the determination result shown in FIG. Extract data.

  In the present embodiment, the risk of developing aspiration pneumonia itself (whether or not there is a risk of onset) is determined through the determination process from determination A to determination F of each check sheet, and the determination result of the risk of onset The probability of symptom occurrence is calculated from the data and is generated as medical support data.

  Here, the criterion (determination criterion) corresponding to each of the determinations A to F of each determination check sheet is based on medical statistical data as described above, and how many check items correspond to each determination, Although it can be set in advance, in the present embodiment, a determination criterion is set using knowledge data based on doctor knowledge text extracted from medical sound data.

  FIG. 7 is a diagram illustrating a processing example for generating knowledge data from doctor knowledge text. As shown in FIG. 7, the knowledge data processing unit 313 performs statistical processing on the doctor knowledge text accumulated in the storage unit 320 to generate knowledge data. For example, it is possible to generate the knowledge data A that ranks the “medical condition history” corresponding to the determination B. For example, as shown in FIG. 2, it is possible to grasp that there is “risk of aspiration pneumonia” when “it is difficult to swallow things” as doctor knowledge text. Thus, the doctor knows empirically the medical condition history of the patient who has a high probability of developing aspiration pneumonia, and can extract it as “knowledge” of the doctor from the doctor knowledge text.

  By statistically processing the doctor's “knowledge” from the doctor's knowledge text in this way, as shown in FIG. 7, in the medical history, the importance of the risk of developing aspiration pneumonia can be ranked and weighted. it can. In the example of FIG. 6, as a determination criterion for determination B, when five or more check items correspond to the medical statistical data as described above, the determination result of determination B is processed as “corresponding”. As a determination criterion based on “knowledge” corresponding to determination B, when two or more symptoms belonging to rank A of knowledge data A are applicable, the determination result of determination B is processed as “applicable”.

  Further, the knowledge data processing unit 313 performs statistical processing on the doctor knowledge text accumulated in the storage unit 320, and knowledge data B corresponding to the determination C, knowledge data C corresponding to the determination E, and knowledge corresponding to the determination F Data D can be generated and stored in the storage unit 320. The knowledge extraction unit 314 can individually perform the determinations A to F with reference to the determination criteria based on the knowledge data stored in the storage unit 320.

  The knowledge extraction unit 314 stores the generated medical support data in the storage unit 320 and transmits the medical support data to the medical system 500 to register the medical support data in the electronic medical record 521 stored in the medical system 500. When receiving the medical support data, the electronic medical chart control unit 511 stores the received medical support data in the electronic medical record 521 of the corresponding patient.

  In addition, as shown in the example of FIG. 6, when the determination criterion based on the knowledge data and the determination criterion based on the medical statistical data coexist, if any one of the determination criteria is met, You may comprise so that it may judge using only the criteria based on judgment or knowledge data.

  8 and 9 are flowcharts showing the processing flow of the home / visit medical service system of this embodiment.

  As shown in FIG. 8, the doctor visits the patient's home, facility, hospital, etc. with the portable terminal device 100 and treats the patient at the visited site. When starting a medical treatment, the doctor connects to the medical treatment system 500 via the portable terminal device 100 and transmits a medical record information acquisition request. The medical care system 500 provides medical record information of the corresponding patient to the mobile terminal device 100 (S101, S501).

  The doctor starts the medical treatment of the patient, and inputs the medical treatment contents and the care contents to the portable terminal device 100 by voice (S102). The portable terminal device 100 generates medical voice data from the medical contents and the care contents inputted by voice (S103). After the medical treatment is completed (S104), the mobile terminal device 100 outputs the generated medical voice data to the management device 300 (S105).

  When the management apparatus 300 receives the medical voice data from the mobile terminal apparatus 100, the management apparatus 300 converts the medical voice data into a text and generates a medical text (S301). The management apparatus 300 extracts / generates each data of medical record text, receipt text, and doctor knowledge text from the generated medical text (S302, S303, S304).

  Next, the management apparatus 300 transmits the generated medical record text and receipt text to the medical care system 500, and performs a medical record / receipt registration process (S305). The medical care system 500 registers the medical record text received from the management apparatus 300 in the electronic medical record 521 (S502). Further, the receipt text received from the management apparatus 300 is registered in the receipt data 523 (S502).

  The doctor performs an approval process for the patient's electronic medical record 521 and the receipt data 523 registered via the mobile terminal device 100 (S106). The medical care system 500 transmits the registered electronic medical record 521 and the receipt data 523 to the mobile terminal device 100 based on the approval request from the mobile terminal device 100. The mobile terminal device 100 displays the electronic medical record 521 and the receipt data 523 on the display unit 103. The doctor refers to the electronic medical record 521 and the receipt data 523, confirms the patient's medical record information and the receipt information, and selects an approval button. The doctor can correct the electronic medical record 521 and the receipt data 523 by an input operation or the like via the display unit 103 as necessary. The mobile terminal device 100 can perform a correction process on the electronic medical record 521 and the receipt data 523 through a predetermined screen.

  When the approval button is selected, the mobile terminal device 100 transmits the approval information or / and the modified electronic medical record 521 and the receipt data 523 to the medical care system 500. The medical treatment system 500 updates the statuses of the electronic medical record 521 and the receipt data 523 to “approved” based on the approval information received from the mobile terminal device 100 (S503). At this time, when the modified electronic medical record 521 or / and the receipt data are received, each data is updated to the corrected data while the status is updated to “approved”. The medical care system 500 can perform a predetermined medical fee claim process based on the approved receipt data.

  Next, the management apparatus 300 generates knowledge data using the doctor knowledge text 324 (S306). For example, as described in the example of FIG. 7, the doctor knowledge text 324 is classified into a plurality of categories (knowledge), and knowledge data is generated through predetermined statistical processing for each classification. The classification can be set in advance corresponding to each of the determinations A to F as described above.

  Each knowledge data generated can be used as a judgment criterion for each judgment A to F. The management device 300 stores the determination reference data of the determinations A to F in the determination check sheet in the storage unit 320 (S307).

  In generating medical assistance data, the management apparatus 300 first acquires patient chart information from the medical care system 500 (S308). The management apparatus 300 matches the patient chart information and the determinations A to F of each determination check sheet (S309). The management apparatus 300 identifies the onset condition result pattern shown in FIG. 5A and the preventive condition result pattern shown in FIG. 5B based on each determination result, and the onset condition result pattern and the preventive condition result pattern Based on this combination, the medical support data for the patient is generated with reference to the occurrence probability data of aspiration pneumonia shown in FIG. 5C (S310).

  The management apparatus 300 stores the generated medical support data in the storage unit 320 and registers the medical support data in the medical system 500 (S311). The medical care system 500 registers (stores) the medical assistance data 522 received from the management apparatus 300 in the electronic medical record 521 (S504). When receiving the medical record information request or the medical support data request from the mobile terminal device 100, the medical system 500 transmits the medical support data 522 and the electronic medical record 521 of the corresponding patient to the mobile terminal device 100 (S107, S505).

  As described above, according to the present embodiment, the medical treatment voice data in the visit medical treatment is converted into text, and the experience and know-how of the doctor are individually extracted as the doctor knowledge text in addition to the information used for the chart information and the receipt information. Therefore, it is possible to provide an environment for providing medical assistance utilizing the experiences and know-how of a plurality of doctors. Then, the extracted doctor knowledge text is accumulated to generate knowledge data, and medical support data for each patient reflecting the knowledge data is generated. For this reason, it is possible to support the alerting of the patient and the formulation of the treatment policy and treatment content for the patient based on the experience and know-how of the doctor.

  The voice data input by the mobile terminal device 100 has been described as an example in which the voice data is output to the management device 300 after the end of the medical treatment. However, the voice data is input to the management device 300 each time the voice data is input without waiting for the medical treatment to end. The text may be output in parallel with the text processing.

  In addition, although the above-described management apparatus 300 has been described as an example in which the management apparatus 300 is separated from the medical care system 500, the management apparatus 300 and the medical care system 500 may have an integrated system configuration. In addition, the management apparatus 300 includes a system configuration in which each processing unit is constructed by software in one server device, or a system configuration in which each processing unit is configured by an individual computer device (for example, the control unit 310 is a control device, The hardware configuration in which the storage unit 320 is independent of each other as a storage device) and a configuration in which a part of each processing unit is included in the mobile terminal device 100 can be appropriately employed. The same applies to the medical care system 500.

  Further, the mobile terminal device 100 described above may be configured not to include a data communication function (communication unit 101). For example, medical voice data generated through the voice input unit 105 and the voice data generation unit 106 is stored in the storage unit 104. After returning to a predetermined medical institution after completing the visit medical treatment, the doctor connects the portable terminal device 100 to a separate computer device having a data communication function installed in the medical institution. The doctor can be configured to transmit (output) the medical voice data to the management apparatus 300 via a separate computer device.

  In addition, each process of the management apparatus 300 that configures the home / visit medical service system of the present embodiment can be realized as a program that can be executed by a computer. It is possible to operate as an information processing apparatus that performs each process of the system. For example, the program is stored in an auxiliary storage device (not shown), and a control unit such as a CPU reads the program stored in the auxiliary storage device to the main storage device, and the control unit reads the program read to the main storage device. It is possible to execute and cause the computer to operate each process according to the embodiment. The same applies to each process of the mobile terminal device 100.

  The program can be applied to a computer in a state where the program is recorded on a computer-readable recording medium, or can be downloaded to a computer through a network such as the Internet. Computer-readable recording media include optical disks such as CD-ROM, phase change optical disks such as DVD-ROM, magneto-optical disks such as MO (Magnet Optical) and MD (Mini Disk), floppy (registered trademark) disks, Examples include magnetic disks such as removable hard disks, memory cards such as compact flash (registered trademark), smart media, SD memory cards, and memory sticks. A hardware device such as an integrated circuit (IC chip or the like) specially designed and configured is also included as a recording medium.

  Moreover, although some embodiment of this invention was described, these embodiment is shown as an example and is not intending limiting the range of invention. These novel embodiments can be implemented in various other forms, and various omissions, replacements, and changes can be made without departing from the scope of the invention. These embodiments and modifications thereof are included in the scope and gist of the invention, and are included in the invention described in the claims and the equivalents thereof.

DESCRIPTION OF SYMBOLS 100 Portable terminal device 101 Communication part 102 Control part 103 Display part 104 Storage part 105 Voice input part 106 Voice data generation part 300 Management apparatus (at-home / visit medical service support apparatus)
310 Control Unit 311 Voice Recognition Unit 312 Medical Record / Receive Processing Unit 313 Knowledge Data Processing Unit 314 Knowledge Extraction Unit 320 Storage Unit 321 Medical Text 322 Medical Record Text 323 Receipt Text 324 Doctor Knowledge Text 325 Analysis Dictionary Group 326 Medical Statistical Data 327 Knowledge Data 328 Medical support data 330 Communication control unit 400 Information terminal device 500 Medical system 510 Control unit 511 Electronic medical record control unit 512 Receipt control unit 520 Storage unit 521 Electronic medical record 522 Medical support data 523 Receipt data

Claims (4)

A speech recognition unit that converts the medical speech data input by voice at the site to text processing and generates medical text data;
A first identifier for identifying each item of the chart information, a second identifier for identifying each item of the medical fee information, and a third identifier for identifying the knowledge of the doctor treating the patient A storage unit;
Based on the first identifier and the second identifier included in the medical text data, medical text data corresponding to the medical chart information and medical fee text data corresponding to the medical fee information are generated from the medical text data. A medical record / receipt processing unit;
A knowledge data processing unit that generates doctor knowledge text data corresponding to doctor knowledge information from the medical text data based on the third identifier included in the medical text data,
The home / visit medical service system, wherein the text data for generating the doctor knowledge information is extracted together with the medical record information and the medical fee information from the medical sound data. A second storage unit that stores symptom determination information including a plurality of determination items for determining a risk of developing a predetermined symptom created from predetermined medical statistical data;
A matching process between the chart information and each determination item is performed, and when the number corresponding to each determination item exceeds a predetermined number, it is determined that there is the onset risk, and the determination result of the onset risk is used as medical support data A knowledge extraction unit to generate,
The knowledge data processing unit calculates the importance for each determination item using the doctor knowledge text data,
The home / visit medical service according to claim 1, wherein the knowledge extraction unit determines that there is a risk of the onset when the number corresponding to the determination item with high importance exceeds the predetermined number. system. The symptom determination information includes a plurality of determination blocks, and the determination block includes one or a plurality of the determination items,
The knowledge extraction unit calculates a number corresponding to the determination item for each determination block by performing a matching process between the medical record information of the patient and the determination item for each determination block,
Further, the knowledge extraction unit calculates a determination result of each determination block based on whether or not the predetermined number set in each determination block is exceeded, and for each combination pattern of the determination results of each determination block, Calculating the probability based on the combination of the determination results of each determination block with reference to medical statistical data in which the probability of the onset risk is obtained in advance, and generating the calculated probability as the medical support data The home / visit medical service system according to claim 2, wherein The plurality of determination blocks are each divided into an onset condition for the predetermined symptom and a preventive condition for preventing the onset of the predetermined symptom,
The medical statistical data, the probability of the onset risk is determined in advance for each combination pattern of the determination result of the determination block corresponding to the onset condition and the determination result corresponding to the prevention condition,
The knowledge extraction unit performs a first matching process between the patient's medical record information and the determination item in the determination block corresponding to the onset condition, and the patient's medical record information in the determination block corresponding to the prevention condition. And a second matching process between the determination item and the determination item, and a probability of the onset risk is calculated based on a combination of a determination result of the first matching process and a determination result of the second matching process. 3. Home / visit medical service system described in 3.