JP4873497B2 - Medical training device - Google Patents

Description

  The present invention relates to a medical training apparatus, and more particularly to a medical training apparatus that performs voice recognition processing.

  Calling a patient in medical practice is an important technique for medical professionals to make the patient feel safe and to gain trust from the patient. However, there has been no device that can perform objective recognition by performing voice recognition processing on voice calls and practice.

  In the conventional speech recognition processing, as shown in Patent Document 1, in order to increase the reliability of the long sentence recognition result, the input speech is divided into word sequences for each utterance, and the part-of-speech information, acoustic Information such as likelihood score, language likelihood score, word likelihood score and the like is assigned to each word. Then, the speech recognition processing is performed based on the recognition degree estimated by converting the determination value by the classification using the statistical value such as the average value and variance value in speech units and the part of speech information into the speech feature vector.

  Furthermore, in Patent Document 2, when a speaker speaks a specific word in a predetermined order, an optimum filtering pattern suitable for the person is automatically selected. The voice recognition process can be performed with a good recognition rate even when the voice state changes.

JP 2007-240589 A JP 2001-022382 A

  However, Patent Document 1 is excellent in analyzing long sentences, but does not utter so long sentences in actual medical training. Further, in the speech recognition process shown in Patent Document 1, since it may take time to analyze a long sentence, it is not suitable for calling in a medical practice that requires a real-time response.

  Further, in Patent Document 2, a filtering pattern is selected in accordance with the voice of the user. In an actual medical training scene, the main purpose is not to recognize the voice of the trainee, but it is easy for the patient to understand. It is asked whether or not it is a voice call. Therefore, as shown in Patent Document 2, it is not necessary to provide and change a filtering pattern suitable for each trainer.

  Therefore, an object of the present invention is to provide a medical training apparatus that determines whether or not a voice is easy for a patient to understand and improves medical technology for voice.

The medical training device according to claim 1 of the present invention is a result of the voice input means for inputting the voice from the practitioner, the voice processing means for processing the voice input from the voice input means, and the result of the voice processing means. and a control means for performing operation and facial expression control based pseudo patients having a driving means or changing facial expressions or operate on the basis of a signal from the control unit, a processing output means including, a In the medical training apparatus, the voice processing means recognizes the voice input from the voice input means as a language, and passes only the voice having a predetermined volume or higher with respect to the voice input from the voice input means. and volume filter which has a adjusting means for adjusting the predetermined volume to filter at the volume filter, a control unit, when the voice is less than a predetermined volume, the voice Operation of the pseudo patient body is the expression of the non-reactive and face controls the driving means so as not to change in response, voice, if it is determined that more than a predetermined volume, a pseudo patient body in response to the voice to be so that or alter the expression or operate to control the pseudo patient body.

A medical training device according to claim 2 of the present invention is the medical training device according to claim 1, wherein a vocabulary filter for determining whether or not the language recognized by the speech recognition means is a predetermined vocabulary. A medical training apparatus further comprising:

The medical training device according to claim 3 of the present invention is the medical training device according to claim 2 , wherein the predetermined vocabulary is medical technical terms.

A medical training device according to claim 4 of the present invention is the medical training device according to any one of claims 1 to 3, wherein the language recognized by the voice recognition means is accent, intonation, Alternatively, an utterance filter that determines whether or not an utterance standard regarding the utterance speed is satisfied is further provided .

A medical training device according to claim 5 of the present invention is the medical training device according to any one of claims 1 to 4 , wherein the language type recognized by the voice recognition means is changed. Can do.

A medical training device according to a sixth aspect of the present invention is the medical training device according to any one of the first to fifth aspects, wherein the simulated patient body is a voice based on a result of the voice processing means. Is further provided.

The medical training device according to claim 7 of the present invention is the medical training device according to any one of claims 1 to 6 , wherein the processing output means displays the result of the voice processing means. Display means are provided.

  In the medical training apparatus according to the first aspect of the present invention, the adjustment means adjusts the predetermined reference that is filtered in the filter, so that it provides optimal training for the voice call that is the medical technique of the trainee and is objective. The technical improvement can be aimed at by performing an effective evaluation.

Further, in the medical training apparatus according to claim 1 , since the voice recognition processing is performed for the sound having a predetermined volume or higher by using the volume filter, the optimal voice training practice is provided for the patient and the apparatus situation. In addition, since the objective evaluation can be performed, the technology can be improved.
In addition, because it is equipped with a simulated patient body, the result of processing the voice of the practitioner by the voice processing means can be fed back to the practitioner via the pseudo patient body, and it is possible to practice to further improve the technique of voice call Become.

The medical training device according to claim 2 of the present invention includes a vocabulary filter for determining whether or not the language recognized by the voice recognition means is a predetermined vocabulary, so that the voice made by the practitioner is better understood by the patient. You can practice to make it easy to do.

In the medical training apparatus according to claim 3 of the present invention, the vocabulary determined by the vocabulary filter is used as the medical technical term, so that it is possible to perform practical training for avoiding the medical technical term difficult for the patient to understand.

In the medical training device according to the fourth aspect of the present invention, since the language that can be recognized by the speech filter is checked, the training performed by the trainee can be practiced so that the patient can speak more easily. In addition, since the speech standard is language accent, intonation, or speech rate, it is possible to practice so that the voice spoken by the practitioner becomes an accent, intonation, or speech rate that is easier for the patient to hear.

In the medical training device according to claim 6 of the present invention, it is possible to use in various countries by changing the language system recognized by the voice recognition means, and it is also possible to train international medical workers become.

In the medical training device according to the seventh aspect of the present invention, since the display means is provided, a result of processing the voice of the trainee by the voice processing means can be output via the display means.

(Embodiment 1)
FIG. 1 shows a schematic diagram of a medical training apparatus including voice processing according to the present embodiment. The medical training device shown in FIG. 1 is a comprehensive medical training device capable of training such as cutting in addition to calling. A medical training apparatus M shown in FIG. 1 includes an instrument table 1 provided with medical instruments 11a to 11e, a simulated patient body 2 imitating a patient, and a clinical table 3 for performing treatment by placing the simulated patient body 2 thereon. And an information processing device 4 that displays various information as a GUI and receives various commands to the simulated patient body 2.

  The instrument table 1 includes an instrument holder 1b on the front side of a table 1a that is rotatably attached to an examination table 3 via an arm, and there is a cutting tool such as an air turbine handpiece or a micromotor handpiece, a scaler, a three Medical instruments 11a to 11e made of a way syringe, a vacuum syringe or the like are detachably attached. In addition, the instrument table 1 is provided with a display unit 5 above the table 1a so that the patient's chart can be called and displayed, and the operation contents of the clinical instrument being practiced can be monitored. In addition to the medical instruments 11a to 11e shown in FIG. 1, the medical instrument is provided with an oral camera and a photopolymerization irradiator (none of which are shown). A foot controller 12a is provided to control these medical instruments. In addition, a medical instrument circuit is provided for detecting the operating state of these medical instruments (not shown).

  The medical device detects and drives each of the medical devices 11a to 11e picked up from the device holder 1b, detects and drives the operation of the foot controller 12a, and operates the operation means provided on the medical device itself. It may be detected and driven. As the instrument holder 1b, in addition to what is provided on the instrument table 1 as shown in FIG. 1, those that can be detachably attached to other diagnostic instruments 11a to 11e such as those provided on the examination table 3 are also targeted. In addition, the medical instrument circuit has a voltage value, a current value corresponding to the number of rotations or the number of rotations of the medical instruments 11a to 11e, or an air pressure, an air flow rate, a frequency, a vibration frequency, or the medical instruments 11a to 11e. Detects the suppression pressure when contacting the simulated patient body 2 and the operation signal of the foot controller 12a connected to the medical table 3. The medical instruments 11a to 11e are connected to a water supply source, an air supply source, and an air suction unit. However, these are well-known, and detailed description thereof is omitted.

  The simulated patient body 2 is composed of a head model 2a, a torso model 2b, left and right arm models 2c, and left and right leg models 2d, in which the posture and expression of the simulated patient body 2 are changed. A simulated patient body drive circuit is provided for allowing the patient to detect medical conditions for the simulated patient body 2 (see the functional block diagram of FIG. 3). Moreover, in order to make the pseudo patient body 2 look very similar to a human body, a wig serving as a hair is put on and the artificial skin is put on. Such a simulated patient body 2 is not a conventional type in which a skeleton is formed of mechanical parts, but is configured as a so-called android-type robot that is covered with artificial skin or artificial hair and is very similar to a human body. During the training, it can be placed on the clinic table 3 like a person, and various treatments can be performed. Mechanical, electrical, and fluid energy (working medium) is used to change posture and facial expression. Is connected to a drive source for supplying The simulated patient body 2 may be integrated or interlocked with the treatment table 3 or may be configured to operate independently. The configuration and driving of the simulated patient body 2 are described in detail in the patent document (International Publication No. 2008/023464 pamphlet).

  The medical table 3 includes a seat portion sheet 3a placed on the base 30 so as to be movable up and down, a tiltable back plate sheet 3b connected to the rear of the seat portion seat 3a, and an upper end of the back plate sheet 3b. And a tiltable headrest 3c connected to each other. The medical table 3 is provided with seat seat lifting / lowering means, back plate seat tilting means, and headrest tilting means (all not shown) for controlling to an optimum position according to the medical situation, and operation control is performed by the foot controller 12a. Is done. A conventional hydraulic cylinder, an electric motor, or the like can be used for such a seat seat raising / lowering means or a back plate seat tilting means, and an electric motor or the like can be used for the headrest tilting means.

  In addition, the treatment stand 3 is provided with a treatment stand pole 6 and is provided with arms 61 and 62 that branch off in the middle of the stand pole and project in a rotatable manner. The arm 61 is provided with a surgical light 63, and the arm 62 is provided with an imaging camera 64 (for imaging the treatment and movement of the medical instruments 11 a to 11 e of the practitioner, the posture, movement and facial expression of the simulated patient body 2). An imaging means) and a microphone 65 are installed to collect sound during treatment. In addition, in the vicinity of the treatment table 3, a water tap for supplying water when rinsing the oral cavity and a spiton 3d including a saliva bowl are provided.

  The information processing apparatus 4 is composed of, for example, a workstation, and is composed of a PC main body 41, display means 42 such as a liquid crystal monitor, and operation means 43 such as a keyboard and a mouse. Various kinds of information are displayed on the display means 42 as a GUI, and various instructions are received by the operation means 43 such as a keyboard and a mouse. The display means 42 can display the consultation status of the simulated patient body 2 at the time of training and can call and display necessary information as needed. By embedding a specific program in the PC main body 41 and operating it, the simulated patient body 2 can be operated. Therefore, the training instructor can operate the facial expression of the simulated patient body 2 while grasping the situation of the trainee. Moreover, various sensors built in each part of the simulated patient body 2 can be called to display the detection signal on a monitor.

  A system stop operation switch 7 is provided on the shoulder of the back plate sheet 3b. The system stop operation switch 7 is an emergency stop switch that stops the entire system (the simulated patient body 2, the clinical table 3, the medical instruments 11a to 11e, etc.) when at least one of the simulated patient body 2 or the medical table 3 malfunctions. is there. The installation position of the system stop operation switch 7 is preferably the shoulder portion of the back plate seat 3b because it is easy for the trainee to operate it, but considering the accessibility by the trainee such as the vicinity of the instrument table 1 and the headrest 3c. Are selected and set as appropriate. Moreover, although the whole system is desirable as a stop target, safety can be maintained if at least the operations of the simulated patient body 2 and the treatment table 3 are stopped.

  In addition, a pseudo patient body weakening operation switch 8 for driving the pseudo patient body weakening means is provided at the neck of the pseudo patient body 2. The simulated patient body weakening operation switch 8 extracts all or a part of the operating pressure (air pressure, fluid pressure, etc.) serving as a working medium (air, water, oil, etc.) in order to drive the simulated patient body driving means. Switch. An opening valve (not shown) is provided in the middle of the drive system (pipe line system) of these working media. By operating the pseudo patient body weakening operation switch 8, the opening valve is opened and the working media is discharged out of the system. Is done. The pseudo patient body weakness operation switch 8 is preferably provided at the neck of the pseudo patient body 2 in terms of accessibility of the trainee, but the side of the pseudo patient body 2, the vicinity of the instrument table 1 or the headrest 3c, etc. In the same manner as described above, it is appropriately selected and set in consideration of accessibility by the trainee. In addition, the simulated patient body 2 may be weakened after the system is stopped by using the system stop operation switch 7 as well.

  FIG. 2 shows a flowchart for explaining the operation of the medical training apparatus according to the present embodiment. FIG. 3 shows a functional block diagram of voice processing in the medical training apparatus according to the present embodiment. First, the CPU 101 shown in FIG. 3 is realized by the PC main body 41 and executes voice recognition processing and the like. The voice input unit 102 shown in FIG. 3 is realized by the microphone 65, and collects voice generated by the trainee's voice. The position of the microphone 65 is an example, and the present invention is not limited to this. The microphone 65 may be attached to the ear or chest of the simulated patient body 2, the chest of the trainee, or the like. The display unit 103 shown in FIG. 3 can be realized by the display unit 5 or the display unit 42, and displays the result of the voice processing executed by the CPU 101. The voice output means 104 shown in FIG. 3 can be realized by a speaker provided in the simulated patient body 2 or the information processing apparatus 4 and outputs a reply or the like due to the voice of the trainee.

  The simulated patient body drive circuit 105 shown in FIG. 3 is a control circuit for driving the simulated patient body 2, and controls the reaction of the simulated patient body 2 when the trainee speaks. The display unit 103, the audio output unit 104, and the simulated patient body drive circuit 105 correspond to a process output unit that outputs a result of the audio process. A memory 106 shown in FIG. 3 is a part for storing a voice processing program executed by the CPU 101, a result of the voice processing, and the like, and can be realized by a storage unit provided in the PC main body 41, an external storage unit, or the like. The voice processing unit 107 shown in FIG. 3 is a process executed by the CPU 101, and processes the voice of the trainee input by the voice input unit 102. The audio processing unit 107 according to the present embodiment is described as software executed on the CPU 101. However, the audio processing unit 107 according to the present invention is not limited to this, and is hardware that has functions equivalent to the software. It may be configured.

  The voice processing unit 107 includes an adjustment mechanism 107a, a filter 107b, and a voice recognition unit 107c. The filter 107b is means for filtering the volume, vocabulary, utterance, and the like according to a predetermined standard. For example, the volume filter has a function of passing a voice over a predetermined volume, and the vocabulary filter has a function of determining whether or not the language recognized by the voice recognition unit 107c is a predetermined vocabulary. The utterance filter has a function of determining whether or not the language recognized by the voice recognition means satisfies a predetermined utterance standard.

  The voice recognition unit 107c is a unit that recognizes the voice input from the voice input unit 102 as a language. For example, when the trainee utters the words “Is it all right” as a voice call, the speech recognition means 107c converts the speech “Is it all right” into data “Is it all right”? The language system that can be recognized by the speech recognition unit 107c is not limited to Japanese, but may be another language system such as English. Furthermore, if the language system can be arbitrarily changed in the medical training device according to the present embodiment, the device can be used in many countries and international medical workers can be trained. .

  The adjustment mechanism 107a is a means for changing the reference for filtering in the filter 107b. For example, when the adjustment mechanism 107a is applied to the volume filter, the adjustment mechanism 107a changes a predetermined volume. As a result, it is possible to adjust the voice passing to the voice recognition means 107c. That is, if the predetermined volume is increased, the voice recognition means 107c is not processed unless the volume that the trainee calls is increased. In addition, since the volume that can be heard varies depending on the age and sex of the patient, it is possible to practice by assuming various patients using the adjustment mechanism 107a. Further, by increasing the predetermined volume in stages, it is possible to train the trainees in a step-by-step manner. The predetermined volume needs to be changed depending on the position of the microphone that collects the voice of the trainee.

  For example, when the adjustment mechanism 107a is applied to the vocabulary filter, the adjustment mechanism 107a changes a predetermined vocabulary. Accordingly, it is possible to adjust the reference whether or not the language recognized by the voice recognition unit 107c can be understood by the patient. In other words, if the vocabulary that the patient cannot understand is adjusted so as to be registered as a predetermined vocabulary, the vocabulary other than the predetermined vocabulary must be used when the apprentice speaks. Moreover, since the vocabulary that can be understood varies depending on the age, sex, etc. of the patient, it is possible to practice by assuming various patients using the adjustment mechanism 107a. In addition, there is a medical technical term as a predetermined vocabulary considered as a medical training device.

  For example, when the adjustment mechanism 107a is applied to the utterance filter, the adjustment mechanism 107a changes a predetermined utterance standard. Thereby, it is possible to adjust whether or not the language recognized by the voice recognition means 107c is an utterance that can be understood by the patient. That is, the voice recognition means 107c can be recognized but cannot be understood if the voice is difficult to hear for the patient. Therefore, when the trainee speaks, the voice must be easy to hear by the patient. In addition, since the utterance that is easy to hear differs depending on the age and sex of the patient, it is possible to practice by assuming various patients using the adjustment mechanism 107a. In addition, as the predetermined utterance standard, language accent, intonation, or utterance speed can be mainly considered.

  Next, the operation of the medical training apparatus according to the present embodiment will be described using the flowchart shown in FIG. First, in step S <b> 1, the apprentice speaks to the simulated patient body 2 (for example, “please open your mouth”), and the voice of the voice is input from the microphone 65. In step S2, the sound input in step S1 is filtered by the volume filter of the filter 107b to determine whether or not it is above a predetermined volume. The predetermined volume is adjusted in advance by the adjusting mechanism 107a in step S3. If it is determined in step S2 that the volume is lower than the predetermined volume, the process proceeds to step S4. In step S4, the simulated patient body drive circuit 105 controls the simulated patient body 2 so that the motion / expression of the simulated patient body 2 is unresponsive, assuming that no voice is heard.

  On the other hand, if it is determined in step S2 that the volume is higher than the predetermined volume, the process proceeds to step S5. In step S5, the speech recognition means 107c is processed for the speech that has passed step S2 and recognized as a language. In step S6, it is determined whether or not the language recognized in step S5 does not correspond to a predetermined vocabulary by the vocabulary filter of the filter 107b. If the language recognized in step S5 does not correspond to a predetermined vocabulary (for example, “oral cavity” when saying “please open the mouth”), the process proceeds to step S7, and if it corresponds, the process proceeds to step S8c. The predetermined vocabulary is adjusted in advance by the adjusting mechanism 107c in step S3.

  Next, in step S7, it is determined whether or not the utterance filter of the filter 107b satisfies a predetermined utterance criterion. In the utterance filter, it is determined whether or not the speech corresponding to the language recognized in step S5 matches the accent, intonation or utterance speed of the language that is a predetermined utterance standard. If it does not correspond to the predetermined utterance standard, the process proceeds to step S8a, and if it corresponds, the process proceeds to step S8b. Note that the predetermined utterance reference is adjusted in advance by the adjustment mechanism 107c in step S3.

  In step S8a, since it is determined that the language recognized in step S5 does not satisfy the predetermined utterance standard, the patient can hear the practitioner but cannot understand what he is talking about. ing. Therefore, in step S8a, the simulated patient body drive circuit 105 controls the simulated patient body 2 so that the motion / expression represents that the motion / expression of the simulated patient body 2 cannot be heard. In step S9a, a voice indicating that it cannot be heard from the voice output means 104 (for example, “I don't hear well” or “Please ask again”) is output (voice output A).

  In step S8b, since it is determined that the language recognized in step S5 satisfies a predetermined utterance standard, this corresponds to a situation where the voice of the trainee can be heard and understood as a patient. Therefore, in step S8b, the simulated patient body drive circuit 105 controls the simulated patient body 2 so that the motion / expression of the simulated patient body 2 becomes an operation / expression corresponding to voice call (for example, an operation of opening the mouth). In step S9b, the voice output unit 104 outputs a voice (for example, “Yes”) corresponding to the voice call (voice output B).

  In step S8c, since it is determined that the language recognized in step S5 corresponds to the predetermined vocabulary, this corresponds to a situation in which the content of the voice of the apprentice cannot be understood as a patient. Therefore, in step S8c, the simulated patient body drive circuit 105 controls the simulated patient body 2 so that the motion / expression of the simulated patient body 2 becomes an operation / expression representing that the contents of the voice call cannot be understood. In step S9c, the voice output means 104 outputs a voice (for example, “I don't know well”) indicating that the content of the voice call cannot be understood (voice output C).

  Note that the information output in steps S8a to c and steps S9a to c may be displayed on the display unit 103 or stored in the memory 106. The configuration of the medical training device according to the present embodiment is an exemplification, and the present invention is not limited to this.

(Embodiment 2)
In the medical training device according to the first embodiment, the voice processing means provided in the device that performs comprehensive training using the simulated patient body 2 has been described, but the present invention is not limited to this.

  Therefore, in the medical training apparatus according to the present embodiment, the microphone that is the voice input unit 102, the PC that includes the CPU 101 that executes the voice processing unit 107, and the display unit 103 are used on a commonly used medical table or medical instrument. A configuration in which a speaker as a monitor and audio output means 104 is attached is conceivable. That is, the medical training apparatus according to the present embodiment has a configuration in which the simulated patient body 2, the system stop operation switch 7, the imaging camera 64, and the like are not provided in the schematic diagram illustrated in FIG. Also in the block diagram shown in FIG. 3, the medical training apparatus according to the present embodiment is the same as the first embodiment except that the simulated patient body drive circuit 105 is not provided.

  Since the medical training apparatus according to the present embodiment does not obtain the reaction of the simulated patient body 2 as in steps S8a to c shown in FIG. 2 because the configuration as described above is employed, steps S9a to c Only the sound like this is output. Moreover, if it is a structure like the medical training apparatus which concerns on this Embodiment, it can install not only as a training apparatus with respect to a student but also as a training apparatus with respect to a dentist by installing in the existing medical desk etc. .

It is the schematic of the medical training apparatus which concerns on Embodiment 1 of this invention. It is a flowchart for demonstrating operation | movement of the medical training apparatus which concerns on Embodiment 1 of this invention. It is a block diagram of the medical training apparatus which concerns on Embodiment 1 of this invention.

Explanation of symbols

  1 instrument table, 1a table, 1b instrument holder, 2 simulated patient body, 2a head model, 2b torso model, 2c left and right arm model, 2d left and right leg model, 3 clinical table, 3a seat seat, 3b back plate Seat, 3c Headrest, 4 Information processing device, 5 Display unit, 6 Treatment stand pole, 7 System stop operation switch, 8 Simulated patient body weakness operation switch 11a to 11e Medical instrument, 12a Foot controller, 30 base, 41 PC body 42 display means, 43 operation means, 61, 62 arm, 63 surgical light, 64 imaging camera, 65 microphone, 101 CPU, 102 voice input means, 103 display means, 104 voice output means, 105 simulated patient body drive circuit, 106 memory, 107 sound processing means.

Claims (7)

Voice input means for inputting voice from the practitioner,
Voice processing means for processing the voice input from the voice input means;
A pseudo patient body having control means for controlling motion and facial expression based on a result of the voice processing means, and drive means for operating or changing facial expression based on a signal from the control means; Processing output means including:
A medical training device comprising:
The voice processing means is
Voice recognition means for recognizing the voice input from the voice input means as a language;
For the voice input from the voice input means,
A volume filter that passes only the sound above a predetermined volume ;
Adjusting means for adjusting the predetermined volume for filtering by the volume filter ;
Have
The controller is
When the sound is lower than the predetermined volume, the driving means is controlled so that the operation of the pseudo patient body does not react and the facial expression does not change corresponding to the sound,
The voice, when it is determined that the predetermined volume or more, and controls the pseudo patient body to so that or in response to voice changing facial expressions or operate the pseudo patient body Medical training device. The medical training device according to claim 1,
A medical training apparatus , further comprising a vocabulary filter for determining whether or not the language recognized by the voice recognition means is a predetermined vocabulary. The medical training device according to claim 2,
The medical training apparatus characterized in that the predetermined vocabulary is medical technical terms. A medical training device according to any one of claims 1 to 3,
A medical training apparatus , further comprising: an utterance filter that determines whether or not the language recognized by the voice recognition unit satisfies an utterance standard related to accent, intonation, or utterance speed .   A medical training device according to any one of claims 1 to 4,
  A medical training apparatus characterized in that the language type recognized by the voice recognition means can be changed.   A medical training device according to any one of claims 1 to 5,
  The simulated patient body is:
  A medical training apparatus, further comprising voice output means for outputting voice based on the result of the voice processing means.   A medical training device according to any one of claims 1 to 6,
  The processing output means includes
  A medical training apparatus comprising display means for displaying a result of the voice processing means.

Images