KR102297466B1 - Appatus and method for asking a patient about his condition - Google Patents

Abstract

A paperweight device is provided. The questionnaire includes an analysis unit that outputs the patient's pain information when the patient's answer is input, and the pain information includes a pain region, a pain expression vocabulary, a pain occurrence situation, pain intensity, pain frequency, and pain period. It may include at least one.

Description

APPATUS AND METHOD FOR ASKING A PATIENT ABOUT HIS CONDITION

The present invention relates to a questionnaire and a method for asking a patient about the patient's condition.

When the patient visits the hospital, he or she indicates his/her current condition on a paper questionnaire in a predetermined form and answers various items. However, since symptoms and treatment processes are different for each patient, it is difficult to expect an efficient questionnaire with only one standardized paper form.

Korean Patent Publication No. 1584308 discloses a smart electronic medical questionnaire system that can collect medical questionnaire data through a mobile terminal and utilize it for customized treatment for each patient. However, since a preset questionnaire is provided to the patient terminal, there is a limit in obtaining only a short-answer answer from the patient.

Korean Patent Publication No. 1584308

An object of the present invention is to provide an interview apparatus and method for performing an interview on behalf of a doctor and collecting patient status information.

The questionnaire apparatus of the present invention includes an analysis unit that outputs pain information of the patient when a patient's answer is input, and the pain information includes a pain area, a pain expression vocabulary, a pain occurrence situation, pain intensity, pain frequency, and pain. It may include at least one of the periods.

The questionnaire method of the present invention extracts a characteristic word related to pain from an answer that a patient explains about pain in a sentence, and extracts a matching word matching the extracted characteristic word from a pre-registered questionnaire, and in the questionnaire dictionary Pain information including medical information that the matching word means may be generated.

The questionnaire method of the present invention includes a questionnaire step of asking a patient a question related to pain and obtaining an answer from the patient; a detection step of analyzing the patient's answer to detect a characteristic word; a matching step of extracting a matching word matching the characteristic word from a pre-registered questionnaire; a selection step of selecting a next question item using the matching word; an open question step of obtaining free answers from the patient related to the pain; A confirmation step of reconfirming the patient's answer obtained in the questionnaire step; may include.

According to the questionnaire apparatus and method of the present invention, when the patient's answer is input, an analysis unit that outputs the patient's pain information including the pain occurrence situation may be provided.

The pain information is intended to replace the medical staff's interview including the doctor, and the key point is to include enough information (including the symptom analysis result) that the doctor does not need to perform a re-interview.

In order to provide a wealth of pain information that does not require a doctor to re-examine the patient, the questionnaire device of the present invention may ask the patient a subjective question instead of a multiple-choice question.

According to the present invention, the patient can express his or her pain in various ways, and the analysis result of the variously expressed pain can be useful to the doctor.

In addition, since the present invention can analyze the patient's answer through utterance (voice) as well as text information, an environment similar to an actual questionnaire performed by a doctor can be provided. This allows the patient to perform their pain expression more realistically.

The questionnaire apparatus and method of the present invention may detect a characteristic word included in a patient's answer by using a questionnaire in which pain expressions based on consultation with a doctor or medical information are registered. Medical information related to the patient is derived using the detected characteristic word and the questionnaire dictionary, and the medical information may be included in the pain information output from the questionnaire.

In accordance with the present invention, open-ended answers allow specific answers to future questions to be obtained now. In this case, it may be inconvenient for the user to re-perform a future question that requires a specific answer already answered by the user even though the user's turn has arrived. According to the questionnaire apparatus and method of the present invention, when various characteristic words included in an answer are analyzed, if a specific characteristic word corresponds to an answer to a future question, the future question can be excluded from the list of additional questions for the patient. have. As a result, the patient may not be asked absurd questions again asking for answers to which they have answered in the past.

On the other hand, if the patient provides an answer omitting a characteristic word necessary for generating pain information, repeating the same question to obtain the characteristic word may be inappropriate for both the patient and the questionnaire. According to the present invention, a plurality of questions for acquiring the same type of characteristic word may be provided in various ways. As a result, when a specific type of characteristic word is not obtained through a specific question, the questionnaire may provide the patient with another question that induces a specific type of characteristic word.

On the other hand, when a patient uses a pain expression that is common to a plurality of symptoms, it is necessary to collect 'other pain expression' rather than the corresponding pain expression. According to the present invention, a 'other question' to obtain a 'different pain expression' may be performed. At this time, the corresponding 'other question' may include 'expression of another pain', for example, "Please express pain differently except for 'sting'.

According to the present invention, the content or category of the next question may be changed according to the patient's current answer. According to the present method, which is similar to feedback control at a glance, the patient can be queried with the least optimal question to understand the patient's pain.

The present invention is capable of examining muscle pain. When examining muscle pain, the questionnaire device of the present invention can intensively collect answers to specific postures that cause muscle pain from the patient. Information about the specific postures that cause pain can be of great help to doctors who treat muscle pain.

When examining muscle pain, the medical device can classify whether muscle pain is continuous regardless of posture or appears only in a specific posture, and can ask the patient different questions according to the classification result. As a result, an interview environment optimized for muscle pain can be provided.

On the other hand, according to the present invention, since the medical examination of the patient is performed only when the patient is present in the setting area, unnecessary waste of energy can be prevented.

In addition, the medical questionnaire device of the present invention can automatically write pain information based on the patient's answer on the medical sheet used by the doctor. According to the automatic entry, it can save the doctor the trouble of having to write down pain information separately. In order for the doctor to directly check the contents of the questionnaire, a recorded file in which the questions and answers of the patient are recorded may be delivered to the doctor or the hospital server.

In particular, according to the present invention, it is possible to provide a medical questionnaire for patients with muscle pain instead of medical staff to collect the patient's status information. Unlike the service robot that only performs hospital guidance or reservations, the questionnaire robot can be applied as a part of a robot system that can replace the rounds of actual doctors. If robots replace patient questionnaires, medical staff will be able to focus on treatment and increase the number of patients they can respond to per day.

1 is a schematic diagram showing a paperweight device of the present invention.
2 is a block diagram showing the analysis unit of the present invention.
3 is a schematic diagram illustrating an expression vocabulary for muscle pain.
4 is a flow chart showing the paper-weighting method of the present invention.
5 is a flowchart illustrating a process in which a specific question is selected by the selection unit of the present invention.
6 is a diagram illustrating a computing device according to an embodiment of the present invention.

Hereinafter, with reference to the accompanying drawings, embodiments of the present invention will be described in detail so that those of ordinary skill in the art to which the present invention pertains can easily implement them. However, the present invention may be embodied in many different forms and is not limited to the embodiments described herein. And in order to clearly explain the present invention in the drawings, parts irrelevant to the description are omitted, and similar reference numerals are attached to similar parts throughout the specification.

In the present specification, duplicate descriptions of the same components will be omitted.

Also, in this specification, when it is mentioned that a certain element is 'connected' or 'connected' to another element, it may be directly connected or connected to the other element, but another element in the middle. It should be understood that there may be On the other hand, in this specification, when it is mentioned that a certain element is 'directly connected' or 'directly connected' to another element, it should be understood that the other element does not exist in the middle.

In addition, the terms used herein are used only to describe specific embodiments, and are not intended to limit the present invention.

Also, in this specification, the singular expression may include the plural expression unless the context clearly dictates otherwise.

Also in this specification, terms such as 'include' or 'have' are only intended to designate that the features, numbers, steps, operations, components, parts, or combinations thereof described in the specification exist, and one or more It should be understood that the existence or addition of other features, numbers, steps, operations, components, parts or combinations thereof is not precluded in advance.

Also in this specification, the term 'and/or' includes a combination of a plurality of listed items or any of a plurality of listed items. In this specification, 'A or B' may include 'A', 'B', or 'both A and B'.

Also, in this specification, detailed descriptions of well-known functions and configurations that may obscure the gist of the present invention will be omitted.

1 is a schematic diagram showing a paperweight device of the present invention. 2 is a block diagram showing the analysis unit 150 of the present invention.

The paper-weighting device shown in the drawing may include a sensing unit 110 , a paper-weighting unit 130 , an analysis unit 150 , and a communication unit 170 .

The medical examination apparatus of the present invention may be applied to a robot system in which an automatic robot performs an examination performed by a doctor on the patient 10 instead. Alternatively, the medical examination apparatus of the present invention may be mounted in the form of software in a terminal provided in a hospital, a terminal carried by a user, or the like.

An interview may mean that a doctor asks a patient about the patient's and family medical history, the time of onset, and the progress of the disease. In order to alleviate the patient's condition, it is necessary to identify the cause of the patient's symptoms or pain.

Since the object of the questionnaire is a patient, the patient must exist for the normal operation of the questionnaire.

The detection unit 110 may include an input button for recognizing a patient, a camera, a human body detection sensor, and the like.

The input button may be operated by the patient. When the manipulation of the input button is sensed, the sensing unit 110 may estimate or determine that the patient is present at a position where an interview is possible.

The camera may photograph the position where the paperweight is made. The detection unit 110 may analyze the video image of the camera to determine whether a patient is present in the interviewable area.

The human body detection sensor may include an infrared sensor, a passive infrared (PIR) sensor, a pressure sensor sensing the weight of a human body, a piezo sensor, and the like. The pressure sensor measures the weight of a chair disposed at the interview position, and when the weight of the human body is added, it can be determined that the patient is present.

The detection unit 110 may provide the corresponding information to the interview unit 130 , the analysis unit 150 , and the communication unit 170 when it is detected that the patient is present at the interviewable position. The information provided from the sensing unit 110 to another unit may include a trigger signal for instructing an operation start. Alternatively, when it is determined that a patient is present in a position capable of being interviewed, the sensing unit 110 may provide driving power to each unit.

The detection unit 110 may stop the operation of the interview unit 130 , the analysis unit 150 , and the communication unit 170 when the patient does not exist in the position where the patient can be interviewed for a set time period. In summary, the sensing unit 110 can prevent unnecessary operation of each unit in the absence of a patient, and can prevent unnecessary waste of energy.

A speaker for asking a question to the patient by voice and a microphone for acoustically acquiring the patient's answer may be provided.

An interview area (examination possible position) in which the patient can recognize the voice output from the speaker and at the same time propagate the patient's answer to the microphone may be defined. For example, when there is a chair disposed to face the speaker and the microphone, the chair may be set as a paperweight area. In this case, the sensing unit 110 may detect a patient who has entered the questionnaire area.

The analysis unit 150 may operate only when the patient enters the questionnaire area. If the current patient leaves the interview area before completion of the pain information and the time the current patient leaves the interview area satisfies the set time (continues), the analysis unit 150 provides questions about the current patient and pain information for the next patient. can be initialized.

The questionnaire apparatus of the present invention may include a questioning unit 131 and an acquiring unit 133 . The conversion unit 135 may be additionally provided according to circumstances. In this case, it is preferable that the interrogation unit 131 and the acquisition unit 133 are provided in the paperweight unit 130 . The conversion unit 135 may be formed in the paperweight unit 130 or may be formed in the analysis unit 150 .

The questionnaire unit 130 may induce a subjective answer that the patient directly speaks with his or her voice. The patient's answer includes the patient's own utterance (voice), and in addition, the patient's notes, the patient's typing on the terminal, etc. may be included. The questionnaire unit 130 may ask a question to the patient for obtaining a characteristic word necessary for generating pain information. The questionnaire unit 130 may ask the patient a plurality of questions in order to obtain a plurality of types of characteristic words necessary for generating pain information. The questionnaire unit 130 may visually or aurally ask a specific patient a question, obtain an answer to the question, and provide it to the analysis unit 150 .

The questioning unit 131 may ask the patient a question by text or voice. As a text questioning means, the questioning unit 131 may include various displays in which text is visually displayed. As an audio question means, the questioning unit 131 may include a speaker for outputting audio, an audio output terminal, a headphone, and the like.

The acquisition unit 133 may acquire the patient's answer visually or acoustically. Since the visual answer corresponds to an image, the acquisition unit 133 may include a mouse, an electronic pen, a keypad, a keyboard, and the like, as a visual acquisition means. The acquisition unit 133 for acquiring an auditory answer may include a microphone or the like.

When the questioning unit 131 performs a voice question and the acquiring unit 133 obtains a voice answer, the converting unit 135 may be used.

The conversion unit 135 may convert the auditory answer obtained through the microphone into text information corresponding to digital information (STT, Speech to Text). Alternatively, the conversion unit 135 may convert a question corresponding to text information into a voice output through a speaker or the like (TTS, Text to Speech).

In this case, the analysis unit 150 may analyze and process the text information to generate the patient's pain information. The analysis unit 150 may output the patient's pain information ③ when the patient's answer ② is input. The analysis unit 150 may function as a model function that converts the patient's answer ② into the patient's pain information ③.

The 'pain information' described herein may include at least one of a pain site, a pain expression vocabulary, a pain occurrence situation, pain intensity, pain frequency, and pain duration.

The questionnaire apparatus of the present invention may be for generating or outputting pain information. It is good that the patient's answer input to the questionnaire helps to derive pain information. For this purpose, the question ① that leads to an answer ② that is helpful in the derivation of pain information ③ should be given to the patient, and the example is shown in Table 1 below.

Information Collected Questions pain area Where are you sick? Please tell us where your back, shoulder, or neck hurts. pain expression vocabulary What are your symptoms? How are you sick? situation of pain How do I make the pain worse? pain intensity On a scale of 10 as the most severe pain imaginable, how much pain do you have at your worst? If 10 is the most severe pain imaginable, how much pain do you have on a regular basis? pain frequency How many days a week are you sick? pain period How long have you been sick?

The pain region may mean a body part where pain is felt. In order to derive the pain region, an unspecified question that does not include a body part, such as 'where', or a question that includes a specific body part, such as 'waist, shoulder, or neck' may be given. As an example of the former, 'Where does it hurt?' is possible. As an example of the latter, it is possible to say 'Tell me where it hurts in the back, why, or in the neck'. To this question, the patient can answer something like 'My back hurts'. In this case, the 'waist' may be generated or output by the analysis unit 150 as a pain region. The pain expression vocabulary may mean a way in which pain is felt. In order to derive pain expression vocabulary, pain expression question phrases such as 'symptom', 'how are you?', 'how', and 'are you sick?' may be included in the question. For example, 'How are your symptoms?', 'How are you sick?' You can ask questions such as To this question, the patient can answer, such as 'It hurts', 'It hurts', 'It stings', and so on. As the pain expression vocabulary, 'sting' and the like may be generated or output by the analysis unit 150 .

The pain occurrence situation may include environmental information in which pain is generated. In order to derive the situation in which the pain occurs, the question may include postural expression phrases such as 'how', 'when', and 'what posture', as well as situational expression phrases. For example, 'How do I make the pain worse?' You can ask questions such as To this question, the patient can answer such as 'it hurts when I bend my back'. In this case, 'when bending the back' as the patient's pain occurrence situation may be generated and output by the analysis unit 150 .

Pain intensity may indicate the severity of the pain. The way of expressing the intensity of pain can be seriously different from patient to patient. In order to ensure the maximum objectivity of pain intensity, the question may include a statement of the highest pain level and a confirmation statement asking which level is within the range of the highest pain level. For example, 'If 10 is the most severe pain imaginable, how much is your most painful pain?' ?' You can ask questions such as In this case, 'when the most severe pain imaginable is 10' may correspond to the declaration statement. 'How much pain do you have when you are most sick?' and 'How much pain do you usually have?' may correspond to the confirmation phrases. If the patient answers '5' to the corresponding question, the analysis unit 150 may generate, store, and output '5' as the patient's pain intensity.

Pain frequency may represent the number of times of pain during a set period. Words related to the set period and the number of times may be included in the question. For example, 'How many days a week are you sick?' You can ask questions such as 'A week' may correspond to a word related to the set period, and 'a few days' may correspond to a word related to the number of days. If the patient answers '3 times' to the corresponding question, the analysis unit 150 may generate and output 'one week' and '3 times' as the patient's pain frequency. Pain frequency may include a set period related word included in the question, such as 'week'.

The pain period may refer to the onset of pain or the course of the pain. A word related to the onset of pain or a word related to the course of the pain may be included in the question. An example of a word related to the onset of pain is 'When did you start feeling sick?' You can ask questions such as Examples of words related to the course of pain include 'How long have you been sick?' You can ask questions such as If the patient answers 'a month ago' to the question, the analysis unit 150 may generate or output 'one month' as the patient's pain period.

The analysis unit 150 may include a detection unit 151 , a matching unit 153 , and a classification unit 155 .

The detection unit 151 may extract a characteristic word from among a plurality of words included in the patient's subjective answer.

For example, when the user answers a question related to the pain period with a sentence containing a plurality of words, such as 'a month ago', the detector 151 analyzes the sentence and extracts 'one month' as a characteristic word. can

The matching unit 153 may search for a matching word matching the characteristic word in a pre-registered questionnaire.

In order to grasp the meaning of the characteristic word, a paper-weight dictionary in which a plurality of words are registered may be prepared. The matching unit 153 may search for matching words such as 'one month', '1 month', '1 month', 'one month' that match the characteristic word 'one month'.

The classification unit 155 may generate pain information by using pain classification information in the questionnaire dictionary for the matching word. For example, in the questionnaire dictionary, 'one month' may correspond to a temporary pain classification. In this case, the classification unit 155 may classify the patient's pain as temporary, and include information indicating that the corresponding pain is temporary in the pain information.

The characteristic word may be a word representing the patient's pain, and a word matching the questionnaire among words used in daily life may be preferentially selected as the characteristic word.

Table 2 shows examples of characteristic words representing each pain information. The slot variable name may be a variable representing each pain information when the analysis unit 150 is implemented as software. Each variable appearing in Table 2 may be changed by the manager of the questionnaire or analysis unit 150 .

Slot variable name Meaning (kind, stage) Examples of stored data (characteristic words) pain_point pain area waist, shoulder pain_situation situation of pain when walking, when leaning pain_type pain expression vocabulary Stiffness (somatic pain) high_pain Peak Pain Intensity (1-10) 5 normal_pain Normal pain intensity (1-10) 3 pain_frequency Pain frequency (weeks) number 5 pain_term pain period one month

When the first answer is obtained through the first question for obtaining the first type of characteristic word, the analysis unit 150 detects not only the first type of characteristic word but also the second type of characteristic word from the first answer. The first type may indicate a sub-item of pain information, meaning, type, and stage of Table 2, and the like. For example, the first type may be a pain site. The first kind of characteristic word may be a waist, a shoulder, and the like. In order to obtain the corresponding characteristic word, the questionnaire unit 130 may ask the patient a first question such as 'Where does it hurt?'

In this case, as an answer to the first question, the patient may provide a first answer such as 'I have a sore back when walking'. In this case, the first answer relates to the first question, but answers to other second questions may be added due to the nature of the short-form answer. For example, in the first answer, not only the first type of characteristic word 'waist', but also the second type different from the first type, for example, 'when walking', a characteristic word of a pain occurrence situation, and a characteristic word of a pain expression vocabulary ' It may also include 'stiffness'. In this case, the analysis unit 150 may detect all types of characteristic words included in the first answer. In other words, even in the first question stage, all kinds of characteristic words 'waist', 'walking', and 'stiffness' may all be detected by the detection unit 151 of the analysis unit 150 .

When the second type of characteristic word is detected from the first answer, the analysis unit 150 may process the second question for obtaining the second type of characteristic as a duplicate question. According to the present embodiment, the question 'How does the pain get worse?' for acquiring the pain occurrence situation, 'How do you feel sick?' for acquiring the pain expression vocabulary etc. correspond to the second question item, and each may be treated as a duplicate question by the detection unit 151 . In this case, the information on the duplicate question may be provided to the questionnaire unit 130 .

The questionnaire unit 130 may exclude duplicate questions from among the questions about the patient. As a result, patients may not be asked duplicate questions such as 'How do I make the pain worse?'

Meanwhile, a plurality of different first questions for obtaining a first type of characteristic word necessary for generating pain information may be provided. For example, in Table 1, when the first type is a pain expression vocabulary, a plurality of first questions 'How are your symptoms?' and 'How do you feel sick?' may be prepared for obtaining the characteristic words of the first type. have.

In this case, the questionnaire unit 130 may ask the patient any one of the plurality of first questions. For example, the questionnaire unit 130 may ask a question set as a representative question among the plurality of first questions to the patient. Alternatively, the questionnaire unit 130 may ask a question randomly selected from among the plurality of first questions to the patient.

If the first type of characteristic word is not detected in the patient's answer, the questionnaire unit 130 may extract one of the remaining first questions and ask the patient again. For example, if a characteristic word is not detected in the answer to 'How are your symptoms?', the questionnaire unit 130 returns to the same question 'How are your symptoms?' Instead, you can ask the question in a different way, such as 'How are you?' If the same question continues, the patient is more likely to give the same answer. According to the present embodiment, a characteristic word that can be matched in the questionnaire dictionary can be easily derived from a patient by using a plurality of questions that have the same meaning and only the expression is changed.

When an intersecting characteristic word representing a plurality of types of pain is detected from the patient's answer, the questionnaire unit 130 may ask the patient a question about the induced question including the intersecting characteristic word.

As an example, when a pain expression vocabulary is used in muscle pain, the increase in pain may be partially recognized as shown in FIG. 3 .

The patient's answer may include sticking to it, suffocating it, and so on. Based on the existing experimental data, the questionnaire dictionary can classify 'coercion' and 'eaten up' as somatic pain. The analysis unit 150 may classify the user's symptoms as 'somatic pain' based on the questionnaire, and include 'somatic pain' corresponding to the classification content in the pain information.

However, when the patient answers 'it hurts', it is difficult to classify the patient's condition. This is because, experimentally, the expression 'sting' is commonly used (intersecting) in somatic pain, neuropathic pain, and visceral pain.

In this case, even if a plurality of questions with different expressions are asked, the user can continuously answer 'pick' corresponding to the intersecting characteristic word. In order to avoid the intersecting characteristic word, the questionnaire unit 130 may generate a guide question including the intersecting characteristic word. Guidance questions may include intersecting characteristic words such as 'Please express yourself differently except for stinging', and may include exclusive words such as 'except'. When asking a lead question that includes both an intersecting characteristic word and an exclusive word, the patient can express pain with other expressions instead of 'sting'.

The analysis unit 150 may analyze the patient's answer to determine pain information useful to the patient or the doctor.

A selection unit 157 for generating or selecting the following questions for a specific patient based on the identified pain information may be provided. The selection unit 157 may be formed in the paperweight unit 130 or the analysis unit 150 , or may be formed separately.

The selection unit 157 may transmit the following questions to the questionnaire unit 130 that is currently asking a question for a specific patient.

The questionnaire unit 130 may ask a specific patient the 'next question' received from the selection unit 157 , and provide an answer obtained from the specific patient again to the analysis unit 150 .

According to the present embodiment, the following questions for the patient may be changed according to the patient's current answer. According to this feedback-controlled embodiment, it is possible to adaptively ask a patient's current answer. For example, when the current answer includes an answer to a later question, the later question may be deleted from the list of questions to be asked. As an example, different questions may be selected depending on whether the patient's pain identified through the current answer continues. When it is determined that the persistent pain is persistent, the selection unit 157 may select a question suitable for the persistent pain. When it is determined that the pain is temporary pain or pain in a specific posture, the selection unit 157 may select a question suitable for temporary pain.

The questionnaire unit 130 may ask the patient a question related to muscle pain.

In the case of muscle pain, what is actually helpful to the doctor caring for the patient may include certain postures that cause muscle pain. Accordingly, the analysis unit 150 may identify a pain occurrence situation including a specific posture causing muscle pain through the patient's answer.

The classification unit 155 of the analysis unit 150 may classify whether the muscle pain is a continuous first pain or a second pain induced in a specific posture through analysis of the patient's answer.

When the muscle pain is classified as the first pain, the questionnaire unit 130 may ask the patient about the pain expression vocabulary, the usual pain intensity, and the pain period.

If the muscle pain is classified as the second pain, the questionnaire unit 130 may additionally ask the patient about the pain expression vocabulary, the usual pain intensity, and the pain period, as well as the maximum pain intensity and the pain frequency.

The analysis unit 150 may automatically write pain information on a medical sheet used by a doctor. The medical sheet may further include personal information of the patient, symptoms, and the like.

The pain information output from the analysis unit 150 may include a lookup table in which a 'meaning' item and a 'stored data example' item are described in Table 2, and the like. The pain information alone may not be enough for a doctor to judge the patient's symptoms, and a plan to compensate for this may be prepared.

In the present invention, a communication unit 170 that communicates with a doctor's terminal or a hospital server by wire or wireless may be provided. In this case, the communication unit 170 may transmit the recorded file in which the questions about the patient and the answers of the patient are recorded to the terminal of the doctor in charge of the patient or the hospital server. Pain information may also be transmitted to the terminal or hospital server by the communication unit 170 .

The analysis unit 150 may configure a script-based dialogue system that performs appropriate answers and questions in response to the patient's responses and does not perform the answered questions again when performing an interview to obtain information about the patient's muscle pain.

4 is a flow chart showing the paper-weighting method of the present invention. FIG. 4 may be described as an operation of the paperweight device shown in FIG. 1 .

In the questionnaire method of the present invention, a characteristic word related to pain may be extracted from an answer that the patient explained about pain in a sentence (performed by the detection unit 151). The questionnaire method of the present invention may extract a matching word matching the extracted characteristic word from a pre-registered questionnaire (performed by the matching unit 153). The questionnaire method of the present invention may generate pain information including medical information meaning a matching word in the questionnaire dictionary (performed by the classification unit 155 ). For example, the medical information may include somatic pain, neuropathic pain, visceral pain, and the like.

The questionnaire method of the present invention may transmit the patient's answer along with pain information to the doctor or hospital as it is.

Specifically, the questionnaire method of the present invention includes a questionnaire step (S 510), a detection step (S 520), a matching step (S 530), a selection step (S 540), an open question step (S 560), a confirmation step (S 570) ) may be included.

In the questionnaire step ( S510 ), a question related to pain may be asked to the patient and an answer from the patient may be obtained. The paperweight step may be performed by the paperweight unit 130 .

In the detection step ( S520 ), a characteristic word may be detected by analyzing the patient's answer. The detection step may be performed by the detection unit 151 . In the detection step ( S520 ), an operation of the conversion unit 135 for converting the patient's answer into text information or converting a text question into voice may be added.

In the matching step ( S530 ), a matching word matching the characteristic word may be extracted from the pre-registered questionnaire. The matching step may be performed by the matching unit 153 .

In the selection step ( S540 ), the next question may be selected using the matching word. This may be performed by the selection unit 157 . The selection step may be performed by the selection unit 157 .

In the open question step ( S560 ), free answers from the patient related to pain may be obtained. The open question step may be performed by the questionnaire unit 130 .

In the confirmation step (S570), the answer of the patient obtained in the questionnaire step may be reconfirmed. The confirmation step may be performed by the paperweight unit 130 .

A dialogue system in which the robot equipped with the questionnaire unit 130 first asks the patient a question, and processes the patient's response to acquire patient data may be implemented (interview step).

In the dialogue system flowchart of FIG. 4 , the robot's question performance may include a text to speech (TTS) process of converting a question sentence into a voice, and the patient response acquisition and format conversion part is STT ( Speech to Text) process may be included. In these two processes, Google API and Waton-based Korean API, Aibril API, can be used (detection stage).

The patient response (patient's response) and vocabulary matching algorithm (matching step) may be a process of opening a corresponding category of a questionnaire corresponding to a question and finding whether there is a word included in the keyword through matching between the keyword in the category and text data. In this process, keywords with various conjugations such as verbs and adjectives (in this system, pain expression vocabulary) are detected using Aibril's conversation API, and in the case of other keywords such as nouns, numbers, and negative words, it can be detected from the internal questionnaire. can

The question selection algorithm (selection step) is performed differently depending on the success of word matching in the response vocabulary matching algorithm. If the word matching succeeds, the next extracted data is saved and the next question is selected. Pick one question and do it.

Performing an additional open question depending on whether or not the final question is asked (S 550) is an open question in case the user has not been able to talk about symptoms only by answering the previous question when all the data to be obtained from the user is obtained. It is performed in the form (S 560), and this answer may be recorded and stored without being processed immediately.

The collected information performed in the pain dialogue system and a list of questions corresponding thereto may be shown in Table 1.

As for the flow of questions, different questions are asked in the future by classifying whether the pain is continuous pain or pain that intensifies in a specific posture or situation according to the vocabulary collected in the expression collection part of the pain occurrence situation, and this process is shown in FIG. 5 .

5 is a flowchart illustrating a process in which a specific question is selected by the selection unit 157 of the present invention.

If it is determined that there is a pain occurrence situation, all collected information is interviewed in turn. If it is determined that the pain occurrence situation does not exist, it is regarded as a case of persistent chronic pain. can be performed. Specifically, if there is a pain occurrence situation (S541), a question about the pain expression vocabulary (S542), a question about the highest pain intensity (S543), a question about the usual pain intensity (S544), and a question about the pain frequency A question about (S545) and a question about the pain period (S546) may be made in order. If the pain occurrence situation does not exist (S 541), the question about the pain expression vocabulary (S 542'), the question about the usual pain intensity (S 544'), and the question about the pain period (S 546') are made in order. possible (performed by the selection unit 157).

The form in which the information collected in the system is stored is shown in Table 2, and the pain expression vocabulary is divided into somatic pain, neuropathic pain, and visceral pain according to the vocabulary. can Since it is sometimes ambiguous which pain group the word belongs to, an intersection classification criterion such as 'somatic pain or neuropathic pain' may be additionally prepared. In addition, items such as other pain for which there is no record of which pain group in the literature it belongs to and 'sick' for which it is impossible to accurately classify pain were set as exception classification criteria. There is (performed by the classification unit 155).

After performing the final question, the additional open question execution part is a device that allows the user to freely speak if he or she has a will to describe symptoms other than answering the robot's question. It can be recorded and saved without proceeding.

The final collection data confirmation sentence output unit provided in the questionnaire unit 130 summarizes the user's response, and the user can confirm whether the robot correctly recognized the user's words. In case of persistent chronic pain, the confirmation sentence can be output in the form of "You started having symptoms such as 'cracking' in your 'shoulder' from '1 month' ago, and you are in pain with a strength of '5'." If the confirmation sentence is general pain, "You started having symptoms like 'stiffness' in 'waist' from '1 week' before, and you have had pain for '4 days' in the past week. Also, 'when bending down' '7' You are sick at the intensity of '3', and you usually get sick with the intensity of '3'." It can be output in the same format as

6 is a diagram illustrating a computing device according to an embodiment of the present invention. The computing device TN100 of FIG. 6 may be a device (eg, a paperweight device, etc.) described herein.

In the embodiment of FIG. 6 , the computing device TN100 may include at least one processor TN110 , a transceiver device TN120 , and a memory TN130 . Also, the computing device TN100 may further include a storage device TN140 , an input interface device TN150 , an output interface device TN160 , and the like. Components included in the computing device TN100 may be connected by a bus TN170 to communicate with each other.

The processor TN110 may execute a program command stored in at least one of the memory TN130 and the storage device TN140. The processor TN110 may mean a central processing unit (CPU), a graphics processing unit (GPU), or a dedicated processor on which methods according to an embodiment of the present invention are performed. The processor TN110 may be configured to implement procedures, functions, methods, and the like described in connection with an embodiment of the present invention. The processor TN110 may control each component of the computing device TN100 .

Each of the memory TN130 and the storage device TN140 may store various information related to the operation of the processor TN110 . Each of the memory TN130 and the storage device TN140 may be configured as at least one of a volatile storage medium and a non-volatile storage medium. For example, the memory TN130 may include at least one of a read only memory (ROM) and a random access memory (RAM).

The transceiver TN120 may transmit or receive a wired signal or a wireless signal. The transceiver TN120 may be connected to a network to perform communication.

On the other hand, the embodiment of the present invention is not implemented only through the apparatus and/or method described so far, and a program for realizing a function corresponding to the configuration of the embodiment of the present invention or a recording medium in which the program is recorded may be implemented. And, such an implementation can be easily implemented by those skilled in the art from the description of the above-described embodiments.

Although the embodiment of the present invention has been described in detail above, the scope of the present invention is not limited thereto, and various modifications and improvements by those skilled in the art using the basic concept of the present invention as defined in the following claims are also presented. It belongs to the scope of the invention.

10...Patient 110...Sensing unit
130... Paperweight Unit 131... Question Department
133...acquisition unit 135...conversion unit
150...analysis unit 151...detection unit
153...Matching Division 155...Classifying Division
157...selection unit 170...communication unit

Claims (7)

and an analysis unit that outputs the patient's pain information when the patient's answer is input;
The pain information includes at least one of a pain site, a pain expression vocabulary, a pain occurrence situation, pain intensity, pain frequency, and pain period,
a questionnaire unit is provided that visually or aurally asks a specific patient a question, obtains the answer to the question and provides it to the analysis unit,
The analysis unit analyzes the answer to determine the pain information,
A selection unit is provided for generating or selecting the next question for the specific patient based on the identified pain information,
The selection unit delivers the next question to the questionnaire unit currently being questioned for the specific patient,
the questionnaire unit asks the specific patient the next question, and provides the next answer obtained from the specific patient back to the analysis unit;
The questionnaire unit asks the patient a plurality of questions to obtain a plurality of types of characteristic words necessary for generating the pain information,
When a first answer is obtained through a first question for obtaining a first type of characteristic word, the analysis unit detects not only the first type of characteristic word but also a second type of characteristic word from the first answer, ,
When the second type of characteristic word is detected in the first answer, the analysis unit processes a second question for obtaining the second type of characteristic as a duplicate question,
The questionnaire unit excludes the duplicate question from among the questions about the patient.
According to claim 1,
The questionnaire unit induces a subjective answer directly spoken by the patient,
The analysis unit is provided with a detection unit, a matching unit, and a classification unit,
The detection unit extracts a characteristic word from among a plurality of words included in the short-form answer,
The matching unit searches for a matching word matching the characteristic word in a pre-registered questionnaire,
The classification unit generates the pain information by using the pain classification information in the questionnaire dictionary for the matching word.
delete and an analysis unit that outputs the patient's pain information when the patient's answer is input;
The pain information includes at least one of a pain site, a pain expression vocabulary, a pain occurrence situation, pain intensity, pain frequency, and pain period,
A questionnaire unit is provided for asking the patient a question to obtain a characteristic word necessary for generating the pain information,
When an intersecting characteristic word representing a plurality of types of pain is detected in the answer, the questionnaire unit is configured to allow the patient to express pain with other expressions except for the intersecting characteristic word and exclusive of the intersecting characteristic word Questionnaire that asks the patient a guided question with words.
delete delete In the paper-weighting method performed by the paper-weighting device,
Questionnaire step of asking a patient a question related to pain and obtaining an answer from the patient;
a detection step of analyzing the patient's answer to detect a characteristic word;
a matching step of extracting a matching word matching the characteristic word from a pre-registered questionnaire;
a selection step of selecting a next question item using the matching word;
an open question step of obtaining a free answer from the patient related to the pain;
a confirmation step of reconfirming the patient's answer obtained in the questionnaire step;
When a first answer is obtained through the first question for obtaining a first type of characteristic word, not only the first type of characteristic word but also a second type of characteristic word is detected from the first answer,
If the second type of characteristic word is detected in the first answer, processing a second question item for obtaining the second type of characteristic word as a duplicate question;
Questionnaire method for excluding the duplicate question from among the questions about the patient.

Images