JP2007226388A - Command input device and program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To accurately and quickly perform the speech input of a command name for issuing the instruction command of specific processing relating to a medical image. <P>SOLUTION: When the speech data of a speech input from a microphone are output from an A/D converter, a CPU performs the speech recognition of the speech data, and converts the speech data into character information, and stores the character information in a storage part 24 as an input character string 240. Then, when a back mode is selected as a speech input mode, a back command name whose number of characters is shorter than that of a standard command name stored in a command table 242 and the input character string 240 are compared. The CPU executes the instruction command of the standard command name associated with the matched back command name among the back command names of the command table 242. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to a command input device and a program for inputting a command name for instructing execution of a specific process relating to a medical image.

  Conventionally, medical image generation apparatuses (hereinafter referred to as “modalities”) such as CT (Computed Tomography), CR (Computed Radiography), MRI (Magnetic Resonance Imaging), mammography apparatus, ultrasonic / endoscopic diagnosis apparatus, and the like. 2. Description of the Related Art There are known medical image transfer apparatuses that transfer medical images that have been taken and generated to an output device such as an imager or a viewer.

  This medical image transfer apparatus takes in a medical image data signal (for example, a video signal or digital data) output from a modality, for example, still image data or a moving image based on the DICOM (Digital Imaging and Communications in Medicine) standard. It is converted into image data (hereinafter, these data are collectively referred to as “image data”).

  As various functions related to the image data, a function of transferring and outputting to an external device such as an imager or a server, a function of temporarily storing the image data in an internal memory, a function of reproducing and outputting the image data, and the like are executed. A user such as a radiographer instructs execution of these functions by operating a key switch or a foot switch provided in the medical image transfer apparatus.

  For example, when the user observes an ultrasound image while placing the transducer of the ultrasound diagnostic apparatus on the patient's abdomen and prints out a still image of the ultrasound image, the user presses the key switch to the output device. Instruct to transfer. At this time, the user must operate both the modality and the medical image transfer apparatus. Further, since the line of sight may be removed from the modality for the operation of the medical image transfer apparatus, the diagnosis is hindered, and contact with various devices is also a problem in medical hygiene.

For this reason, a technique has been devised by voice input of a command name for instructing a specific process related to a medical image. For example, the following technique is known. That is, if the command name (voice data) input by voice from the microphone does not completely match the command name (voice data) registered in advance, a command similar to the command name input by voice is extracted. In addition, there is known a technique for confirming whether or not the extracted command name is a voice input by a user (see, for example, Patent Documents 1 and 2).
JP 2000-5158 A JP 2004-267634 A

  However, in the medical field, there is a possibility that the recognition rate of voice recognition may decrease due to the influence of noise from the external environment because there is a lot of traffic and many voices and broadcast sounds emitted from medical devices. . For this reason, in the techniques of Patent Documents 1 and 2, when speech recognition fails, the command name (speech data) extracted as similar does not include the command name uttered by the user, and an instruction command by voice input is issued. become unable.

  In this case, the user must repeat the utterance of the command name until the intended command name is extracted, and the instruction command to the medical image transfer apparatus takes time. In particular, it is desirable to capture a medical image accurately and quickly so as not to burden the patient. However, if it takes a long time to instruct the medical image transfer apparatus, it may impose a burden on the patient. It has become.

  The present invention has been made in view of the above-described problems, and an object of the present invention is to enable reliable and quick voice input of a command name for instructing a specific process related to a medical image. Is to do.

In order to solve the above problems, a command input device according to claim 1 is provided:
Voice input means;
Voice recognition means for performing voice recognition of the voice input by the voice input means and converting it into character information;
Storage means for storing a command name for performing an instruction to execute a specific process relating to a medical image and a shortened command name having a shorter number of characters than the command name;
Comparing means for comparing the character information converted by the voice recognition means with the shortened command name stored in the storage means;
Command execution means for executing an instruction instruction of the command name associated with the compared shortened command name when the comparison result by the comparison means matches;
It is characterized by having.

The invention according to claim 2 is the invention according to claim 1,
And further comprising selection means for selecting one of the first and second voice input modes,
The comparison means includes
When the first voice input mode is selected by the selection means, the character information converted by the voice recognition means is compared with the command name,
When the second voice input mode is selected, the character information converted by the voice recognition means is compared with the abbreviated command name.

The invention according to claim 3 is the invention according to claim 2,
The selection means includes
One of the first and second voice input modes is selected based on the character information converted by the voice recognition means.

The command input device according to claim 4,
Voice input means;
Voice recognition means for performing voice recognition of the voice input by the voice input means and converting it into character information;
Storage means for storing a plurality of command names for instructing execution of specific processing relating to medical images;
Extracting means for extracting a command name similar to the character information converted by the voice recognition means from the command names stored in the storage means;
A list display unit that displays a list of identification character information having a shorter number of characters than the command name in association with each command extracted by the extraction unit;
Comparison means for comparing the character information converted by the voice recognition means and the identification character information displayed in the list;
Command execution means for executing an instruction command of the command name displayed in association with the compared identification character information when the comparison result by the comparison means matches;
It is characterized by having.

The invention according to claim 5 is the invention according to claim 4,
The extraction means includes
A command name matching the head of the converted character information is extracted from the command names stored in the storage means.

The invention according to claim 6 is the invention according to claim 4 or 5,
The list display means includes:
It is characterized by having a frequency order display means for displaying a list of command names extracted by the extraction means in descending order of user use frequency.

The invention according to claim 7 is the invention according to any one of claims 4 to 6,
The list display means includes:
Among the command names extracted by the extracting means, there is provided a related order list display means for displaying a list of command names related to the command name of the instruction command executed immediately before by the command execution means.

The invention according to claim 8 is the invention according to any one of claims 1 to 7,
The execution of the specific processing relating to the medical image includes at least one of transfer to an external device, writing to a storage unit, and display on a display unit.

The program according to claim 9 is a computer,
Speech recognition means for performing speech recognition of speech input by the speech input means and converting it into character information;
Storage means for storing a command name for performing an instruction to execute a specific process relating to a medical image and a shortened command name having a shorter number of characters than the command name;
Comparison means for comparing the character information converted by the voice recognition means with the shortened command name stored in the storage means;
Command execution means for executing an instruction instruction of the command name associated with the compared short command name when the result of comparison by the comparison means matches;
It is characterized by making it function as.

The program according to claim 10 is a computer,
Speech recognition means for performing speech recognition of speech input by the speech input means and converting it into character information;
Storage means for storing a plurality of command names for instructing execution of specific processing relating to medical images;
Extraction means for extracting a command name similar to the character information converted by the voice recognition means from the command names stored in the storage means;
List display means for displaying a list in association with identification character information for each command name extracted by the extraction means;
Comparison means for comparing the character information converted by the voice recognition means with the identification character information displayed in the list;
Command execution means for executing an instruction command of the command name displayed in association with the compared identification character information when the result of comparison by the comparison means matches;
It is characterized by making it function as.

  According to the first and ninth aspects of the present invention, when the character information converted by speech recognition and the abbreviated command name are compared and matched, a command name instruction command associated with the abbreviated command name is provided. Execute. In general, the accuracy of comparing character information is improved when the number of characters to be compared is shorter. For this reason, in comparison with character information, the accuracy of a shortened command name is higher than that of a command name. Therefore, the voice input of the command name for performing the instruction command for the specific process relating to the medical image can be more reliably performed by the shortened command name.

  According to the second aspect of the present invention, the same effect as that of the first aspect of the invention can be obtained. When the first voice input mode is selected, the character information, the command name, If the second voice input mode is selected, the character information is compared with the reverse command name. In general, in speech recognition, the recognition rate increases as the speech to be recognized becomes shorter. For this reason, when the user selects the second voice input mode in advance and utters the back command name, the command name instruction command can be executed more reliably.

  According to the third aspect of the invention, it is possible to obtain the same effect as the first aspect of the invention, and any one of the first and second voice input modes is converted by voice recognition. Select based on the character information. For this reason, the user can select the voice input mode without touching the command input device.

  According to invention of Claim 4 and 10, the command name similar to the character information converted by performing speech recognition is displayed in a list with the identification character information. When the character information converted by further voice recognition matches the displayed identification character information, an instruction command for the command name associated with the identification character information is executed. For this reason, even if voice recognition fails, a list of command names similar to voice input is displayed, so the user selects a desired command name from the command names and executes a command name instruction command Can be made.

  Further, since the identification character information has a shorter number of characters than the command name, when the user inputs the identification character information by voice, the recognition rate of voice recognition is increased and the accuracy of comparison with the identification character information is also improved. Therefore, the voice input of the command name for performing the instruction command for the specific process regarding the medical image can be performed more reliably and quickly by the voice input of the identification character information.

  According to the fifth aspect of the invention, it is possible to obtain the same effect as the fourth aspect of the invention, and extract and display the command name that matches the character information converted by voice recognition. . For this reason, for example, when the user inputs the head portion of a desired command name by voice, a list of command names starting with the head portion is displayed. Therefore, since the user can select a desired command name from the command names displayed in a list by voice input, the user can save time and effort to repeatedly input voice as in the case of voice recognition failure, and promptly input voice of a command name. Can be done.

  According to the invention described in claim 6 or 7, it is obvious that the same effect as that of the invention described in claim 4 or 5 can be obtained, and the extracted command names are displayed in a list in order of frequency of use by the user. Alternatively, a list of command names related to the command name of the instruction command executed immediately before may be displayed. As a result, the user can input a command name in accordance with the actual usage status according to the command names displayed in the list.

  According to the invention described in claim 8, it is possible to obtain the same effect as that of any one of the inventions described in claims 1-7, transfer of medical images to an external device, and storage Is written and displayed on the display unit according to the voice input of the command name.

  Hereinafter, a command input device of the present invention is applied to a medical image transfer device (hereinafter simply referred to as “transfer device”), and an embodiment of a medical image output system having the transfer device is described with reference to FIGS. Will be described in detail.

〔System configuration〕
First, the system configuration of the medical image output system S will be described. FIG. 1 is a diagram illustrating an example of a system configuration of the medical image output system S. As illustrated in FIG. According to FIG. 1, a medical image output system S includes a transfer device 1 connected to each of a plurality of modalities M, an imager 3 as an output device 9, a server 5, and a color printer 7 connected via a communication network N. Has been configured.

  The modality M is an X-ray imaging apparatus, an ultrasonic diagnostic apparatus, an endoscopic diagnostic apparatus, a CT, or the like, and outputs a data signal of a medical image captured and generated to the transfer apparatus 1. This modality M includes one that converts a captured medical image into digital data according to the DICOM standard and outputs it, and another that outputs the medical image as a video signal or a data signal of digital data.

  The imager 3 forms a medical image on a thermosensitive film based on the image data that conforms to the DICOM standard transferred from the transfer device 1 and outputs the medical image. The heat-sensitive film is a film in which a photosensitive layer is formed by coating an emulsion containing a photosensitive and heat-sensitive photosensitive material on a support such as PET (polyethylene terephthalate).

  The server 5 is configured by a general computer including a display device, a large-capacity storage device, and the like, and stores and stores image data transferred from the transfer device 1 in the storage device or based on the image data. It functions as a viewer that displays and outputs medical images on a display device. The color printer 7 is constituted by a laser printer or the like, and forms a medical image on a recording sheet based on the image data transferred from the transfer device 1 and outputs / discharges the medical image.

  The transfer device 1 converts the data signal of the medical image output from the modality M into image data in a data format corresponding to the output device 9 selected according to the user's instruction command, and the image data is transmitted via the communication network N. Forward. Further, the transfer device 1 temporarily stores the converted image data in an internal memory in accordance with a user instruction command, and causes the modality M to reproduce and display the image data. These various functions relating to image data are referred to as specific processing relating to medical images.

  Instruction commands relating to various functions such as transfer of these medical images to the output device 9, temporary storage in the internal memory of the transfer device 1, and playback display are displayed on the display 150 of the transfer device 1 as command names.

  The user selects a desired command name from among the command names displayed on the display 150 by depressing the key switch 120, a foot switch (not shown), or the like, and corresponds the command name to the transfer apparatus 1. Directs instructions. In the present embodiment, by issuing a command name to the earset microphone (hereinafter simply referred to as “microphone”) 15, an instruction command corresponding to the command name can be issued to the transfer apparatus 1.

[Configuration of transfer device]
Next, the functional configuration of the transfer apparatus 1 will be described with reference to FIG. FIG. 2 is a block diagram illustrating an example of a functional configuration of the transfer apparatus 1. According to FIG. 2, the transfer device 1 is amplified by a CPU (Central Processing Unit) 10, an audio input unit 12, an operation input unit 16, a display I / F 18 that controls the display unit 19, and a video amplifier 21. A decoder 20 that decodes the data signal, a communication unit 22, a program memory 24, an image memory 26, and a storage unit 28 are connected to a system bus 30.

  The CPU 10 is a control unit that comprehensively manages and controls the transfer device 1 by controlling the operation of each functional unit and controlling input / output of data between the functional units. Specifically, a program stored in the program memory 24 is read according to an operation signal input from the operation input unit 16, and processing according to the program is executed. Then, based on the processing result, the display content of the display unit 19 is updated, image data is transferred, stored in the storage unit 28, and the like.

  Further, the CPU 10 has a voice recognition function 11. The voice recognition function 11 is a function that performs voice recognition processing on voice data input from the voice input unit 12 and converts the voice data into a character string, and is a known technique such as HHM (Hidden Markov model). Can be adopted as appropriate.

  Since the operation principle of the voice recognition function 11 is a known technique, a detailed description thereof will be omitted, but a brief description will be as follows. First, a voice feature analysis is performed on the input digital voice data by MFCC (Mel Frequency Cepstral Coefficients) or the like to detect a voice section. Based on the detection result, pattern recognition is performed for each section of the voice data in comparison with the recognition dictionary, and the voice data is converted into a character string.

  The CPU 10 performs voice recognition on the voice data input via the A / D converter 13 and converts the voice data into a character string. If the character string matches the command name, the CPU 10 generates an instruction command corresponding to the command name. Follow the process. For this reason, instead of pressing the operation input unit 16, the user can issue an instruction command to the transfer device 1 by operating the transfer device 1 by uttering a command name and inputting the voice from the microphone 15. become able to.

  The audio input unit 12 includes a microphone 15 that can be attached to and detached from the microphone terminal T1, an amplifier 14, and an A / D converter 13. The amplifier 14 amplifies the audio signal input from the microphone 15 connected to the microphone terminal T 1 and outputs the amplified audio signal to the A / D converter 13. The A / D converter 13 performs A / D conversion on the audio signal amplified by the amplifier 14 and outputs it to the CPU 10 as audio data.

  The operation input unit 16 includes a key switch 120 such as a cursor key or a numeric keypad, a foot switch, and the like, and outputs an operation signal of the pressed key switch to the CPU 10.

  The display I / F 18 is configured by an RGB interface, an NTSC interface, or the like, and controls ON / OFF of the display element of the display unit 19 based on the control of the CPU 10. The display unit 19 corresponds to the display 150 shown in FIG. 1 and is configured by a CRT (Cathode-ray Tube), an LCD (Liquid Crystal Display), or the like. The display unit 19 displays a display screen based on the control of the CPU 10 input via the display I / F 18 and reproduces and displays image data.

  The image data may be reproduced and displayed on the display unit of the modality M. In this case, an encoder that converts the image data into, for example, an NTSC format video signal, and the video signal to the modality M are displayed. This can be realized by providing a video output terminal for output.

  The video amplifier 21 is a modality M connected to a video terminal T3 capable of inputting a NTSC (National Television Standards Committee) / PAL (Phase Alternation by Line) composite video signal, a Y / C component video signal, an RGB separate video signal, or the like. The video signal output from is amplified and output to the decoder 20. The decoder 20 decodes the video signal amplified by the video amplifier 21 into a predetermined data format under the control of the CPU 10 to generate image data. The data format to be decoded includes DICOM format, JPEG format, MPEG format, and the like, and is selected according to the output device 9 to be transferred. In the case of the modality M that can directly input digital data in the RS422 format without going through the video amplifier 21 or the decoder 20, an RS422 receiver may be provided instead of the video amplifier 21 or the decoder 20.

  The communication unit 22 is configured by a LAN interface or the like, connected to the communication network N via the network terminal T5, and performs data communication with the imager 3, the server 5, and the color printer 7 of the output device 9 via the communication network N. It is a functional part.

  The program memory 24 is configured by a ROM (Read Only Memory), a flash ROM, or the like, and is a memory area that stores various programs such as an initial program and application programs executed by the CPU 10. The image memory 26 is configured by a volatile memory such as a VRAM (Video RAM), and is a memory area that temporarily stores image data to be transferred to the output device 9.

  The storage unit 28 includes an HDD (Hard Disk Drive), a semiconductor memory, and the like, and is a non-volatile storage area that stores data related to a program executed by the CPU 10, image data converted by the decoder 20, and the like. .

[First Embodiment]
Next, a first embodiment of the transfer device 1 will be described with reference to FIGS. The transfer apparatus 1 according to the first embodiment is provided with a standard mode and a back mode as command name voice input modes.

  The standard mode is a mode in which an instruction command for the transfer apparatus 1 is input by voice with a standard command name represented by one or a plurality of words. Standard command names include, for example, “Kiroku” as an instruction command for storing image data in the storage unit 28, “Saisei” as an instruction command for reproduction and display on the display unit 19, and an instruction command for transfer to the output device 9. There is "Tenso".

  The reverse mode is a mode in which an instruction command to the transfer apparatus 1 is input by voice using a character string having a shorter number of characters or a numeric reverse command name than the standard command name. Examples of the back command name include “K” corresponding to the standard command name “Kiroku”, “S” corresponding to “Saisei”, “Tee” corresponding to “Tenso”, and the like.

  The user selects either the standard mode or the back mode by pressing the key switch 120 or the foot switch of the operation input unit 16, and utters a command name corresponding to the selected voice input mode. An instruction command is given to the transfer device 1.

  FIG. 3A is a diagram illustrating an example of a data configuration of the storage unit 28 in the first embodiment. According to FIG. 3A, the storage unit 28 stores an input character string 240 and a command table 242. The input character string 240 is a character string acquired by the CPU 10 performing voice recognition processing on the voice data.

  As shown in FIG. 3B, the command table 242 is a data table that stores standard command names and back command names in association with each other in the order of number n. For example, at the head (n = 1) of the command table, the standard command name “KIROK” and the reverse command name “K” are stored in association with each other.

  The CPU 10 compares the input character string 240 with the standard command name when the standard mode is selected as the voice input mode, and compares with the back command name when the back mode is selected. Then, as a result of the comparison, it is determined that the instruction command for the matched standard command name or back command name is inputted by voice, and the corresponding processing is executed.

  In general, the recognition rate increases as the speech data to be speech-recognized is shorter. For this reason, when the reverse mode is selected as the voice input mode in advance and the user inputs the reverse command name by voice, the recognition rate of the voice recognition can be improved. Further, the accuracy of comparing character information increases as the length of the character string to be compared becomes shorter. For this reason, the comparison with the input character string 240 is enhanced by performing the comparison with the back command name shorter than the standard command name.

  Next, a specific operation of the transfer apparatus 1 will be described with reference to the flowchart of FIG. 4 and the display screen example of FIG. First, the CPU 10 determines the voice input mode selected by the user based on the operation signal output from the operation input unit 16 (step A1). If it is determined that the standard mode has been selected (step A1; standard mode), the processes of steps A3 to A21 are performed. If it is determined that the reverse mode has been selected (step A1; back mode), steps A23 to Process A41 is performed.

  When determining that the standard mode has been selected, the CPU 10 displays a command input screen 190 as shown in FIG. 5A on the display unit 19, and receives a voice input from the microphone 15 from the A / D converter 13. Wait for input of audio data. Then, voice recognition processing is performed on the voice data input from the A / D converter 13 (step A3), and the voice data is converted into a character string and stored in the storage unit 28 as the input character string 240 (step A5). .

  Next, the CPU 10 selects the standard command name at the head (n = 1) of the command table 242 as a command candidate (step A7), and compares the command candidate with the input character string 240 (step A11). As a result of this comparison, if the CPU 10 determines that they match (step A11; Yes), the CPU 10 executes processing corresponding to the command candidate instruction command (command) (step A17).

  If it is determined that the command candidate and the input character string 240 do not match (step A11; No), the standard command name selected as the command candidate is not the tail of the command table 242 (step A13; No). The next (n = n + 1) standard command names are sequentially selected as command candidates (step A15). Then, the process proceeds to step A19, and the input character string 240 is compared with the command candidate.

  For example, when voice data “Kiroku” is inputted by voice on the command input screen 190 in FIG. 5A, “Kirok” as the input character string 240 is compared with the standard command name in the command table 242. Go. In the command table of FIG. 3B, it is determined that the leading standard command name matches the input character string 240, and storage (recording) of the image data acquired from the modality M into the storage unit 28 is started. The message screen 192 shown in FIG. 5B is displayed.

  If the CPU 10 determines in step A13 that the standard command name selected as the command candidate is the tail of the command table 242 (step A13; Yes), the CPU 10 requests a voice input again (for example, “one more time” Enter command ") is displayed on the display unit 19 (step A19), and the process proceeds to step A21.

  For example, when the CPU 10 determines that the end key of the key switch 120 has been pressed after the processing of steps A17 and A19 (step A21; Yes), the CPU 10 ends the voice input in the standard mode and presses the end key. If it is determined that the process is to be continued (step A21; No), the process proceeds to step A3, and the processes of steps A3 to A19 are repeated.

  On the other hand, if the CPU 10 determines in step A1 that the reverse mode has been selected, the CPU 10 displays the command input screen 194 in FIG. 5C and recognizes the voice data output from the A / D converter 13 as voice recognition. Processing is performed (step A23), and the input character string 240 is stored in the storage unit 28 from the recognition result (step A25). The CPU 10 selects the back command name at the head (n = 1) of the command table 242 as a command candidate (step A27), and compares the command candidate with the input character string 240 (step A31).

  If it is determined as a result of comparison in step A31 (step A31; Yes), processing corresponding to the instruction command for the reverse command name selected as the command candidate is executed (step A37). When the CPU 10 determines that the end key has been pressed after the command execution (step A41; Yes), the CPU 10 ends the voice input in the back mode and continues (step A41; No). Then, the process proceeds to step A23, and the processes of steps A23 to A39 are repeated.

  The processing (steps A33, A35, A39) performed when the back command name (command candidate) selected in step A31 does not match the input character string 240 is the same as steps A13, A15, A19 in the standard mode. Therefore, the description thereof is omitted.

  For example, when the input character string 240 “K” is acquired by performing voice recognition, the input character string 240 “K” is compared with the reverse command name of the command table 242. In the command table of FIG. 3B, it is determined that the leading back command name and the input character string 240 match. Then, the instruction command of “Kiroku” of the standard command name associated with the back command name, that is, the storage (recording) of the image data acquired from the modality M to the storage unit 28 is started, and FIG. Message screen 192 is displayed.

  Thus, in the reverse mode, the transfer device 1 can be instructed to store the image data in the storage unit 28 by a voice input of “K”, which has a smaller number of characters than the command name “KIROK” in the standard mode.

  As described above, according to the first embodiment, the back command name having a shorter number of characters than the standard command name is stored in association with the standard command name, and the voice input in the back mode matches the back command name. Executes the instruction command of the standard command name corresponding to the reverse command name. As a result, the transfer apparatus 1 can be operated with a voice input shorter than the standard command name.

  For this reason, it can be made hard to receive the influence of the environmental noise of a medical field, a user's voice color change, etc., and the voice input of the instruction command with respect to the transfer apparatus 1 can be performed more reliably. In addition, since the transfer device 1 can be operated by voice input, an operation for capturing a medical image is facilitated, and a user's contact with the transfer device 1 is reduced, so that hygiene is improved.

  In the first embodiment described above, the selection between the standard mode and the back mode is performed by operating the operation input unit 16, but the selection of the voice input mode may be performed by voice input. More specifically, a keyword for selecting the standard mode (for example, “Hyojun”) and a keyword for selecting the back mode (for example, “Ura”) are set in advance, and these keywords are set. It is also possible to switch between the standard mode and the back mode when voice is input. Thereby, the user can select the voice input mode without touching the operation input unit 16 of the transfer apparatus 1.

  Further, although it has been described that the command name is instructed to be executed after voice recognition of the command name, the command name may be once confirmed with the user and executed. For example, when the standard mode is selected and it is determined in step A11 that the command candidate matches the input character string 240, the command confirmation screen 196 of FIG. When voice data “high” is inputted by voice, a command name instruction command is executed. When voice data “yes” is inputted by voice, the process proceeds to step A13 and the command table 242 is executed. Select command candidates from and compare them.

  When the back mode is selected, the command confirmation screen 198 shown in FIG. Then, when voice data “wai (yes)” is inputted by voice, a command name instruction command is executed, and when voice data “n (no)” having the number of characters is inputted by voice compared to the standard mode. The process proceeds to step A33.

  In this way, by confirming the command name that has been voice-recognized on the command confirmation screens 196 and 198, execution of an erroneous instruction command can be prevented, and in the back mode, the recognition rate of voice input at the time of the confirmation Can be increased.

  In addition, the reverse command name is set to the initial letter when the standard command name is expressed in alphabets. For example, a command input screen 200 as shown in FIG. ”,“ D (2) ”,“ Sun (3) ”,... May be set as the back command name. Also in this case, since the reverse command name can be set to a shorter number of characters than the standard command name, it is needless to say that the same effect as in the first embodiment can be obtained.

[Second Embodiment]
Next, a second embodiment of the transfer device 1 will be described with reference to FIGS. Note that the same components as those of the transfer device 1 in the first embodiment are denoted by the same reference numerals, and detailed description thereof is omitted as appropriate.

  The transfer apparatus 1 in the second embodiment is configured to have a standard mode and a selection mode as voice input modes. In the selection mode, when there is no standard command name that matches the input character string 240 in the standard mode, a list of command names similar to the input character string 240 (hereinafter referred to as “similar command names”) is displayed. This mode enables selection from the similar command names.

  FIG. 6A shows an example of the data configuration of the storage unit 28 in the second embodiment. According to FIG. 6A, the storage unit 28 stores an input character string 240, a command table 244, and a similar command table 246. The command table 244 in the second embodiment is a data table that stores standard command names in the order of number (n) as shown in FIG.

  The similar command table 246 is a data table in which command names similar to the input character string 240 are extracted from the command table 244 and stored accumulatively, and similar command names are in order of number (m) as shown in FIG. Remember.

  When the standard command name that matches the input character string 240 is not stored in the command table 244 in the standard mode, the CPU 10 switches the voice input mode to the selection mode. Then, the first character of the input character string 240 is acquired, the standard command name starting from the first character is extracted from the command table 244 as a similar command name, and stored in the similar command table 246. The user can select a desired similar command name by voice input of the number (m) in the similar command table 246 and cause the transfer apparatus 1 to execute an instruction command of the command name.

  Next, a specific operation of the transfer apparatus 1 in the second embodiment will be described with reference to the flowchart of FIG. In addition, the same processing number as the flowchart of FIG. 4 of 1st Embodiment attaches | subjects the same step number, and the description is abbreviate | omitted.

  First, the CPU 10 sets the voice input mode as the standard mode, and performs the same processes of steps A3 to A21 as in the first embodiment. However, if it is determined in step A13 that the command candidate is the tail of the command table 244, That is, if the standard command name that matches the input character string 240 is not stored in the command table 244 (step A13; Yes), the voice input mode is switched from the standard mode to the selection mode, and the following processing is performed.

  Specifically, the first (n = 1) standard command name in the command table 244 is selected as a command candidate (step B23), and the first character of the input character string 240 matches the first character of the command candidate. Whether or not is compared (step B27). If the command candidate matches the first character of the input character string 240 as a result of the comparison (step S27; Yes), the standard command name is additionally stored in the command table 244 as a similar command name (step B29).

  If the command candidate does not match the first character of the input character string 240 as a result of the comparison in step B25, the CPU 10 determines whether the standard command name selected as the command candidate is the tail of the command table 244. Is determined (step B31). If the CPU 10 determines that it is not the end (step B31; No), it selects the next (n = n + 1) standard command name from the command table 244 (step B33), and proceeds to the processing of step B25. To do.

  For example, it is assumed that an input character string 240 “SEIGA” is acquired as a result of voice recognition. At this time, since the standard command name “SEIGA” is not stored in the command table 244 of FIG. 6B, the standard command names “SEISHIGA” and “SETTING” starting with “SE” are extracted from the command table 244. And stored in the similar command table 246 having the data structure shown in FIG. In this way, a similar command table 246 storing a standard command name starting with the first character of the input character string 240, that is, a similar standard command name is created.

  If the CPU 10 determines that the standard command name selected in step B31 is the tail of the command table 244 (step B31; Yes), the CPU 10 reads the similar command table 246 from the storage unit 28, and FIG. As in the similar command selection screen 202, the number m and the similar command names are displayed as a list on the display unit 19 (step B35).

  Then, for example, the display unit 19 displays that the mode has been switched to the selection mode by a display message “Please select the corresponding command number in numbers” (step B37). The CPU 10 waits for input of voice data from the A / D converter 13, performs voice recognition processing on the output voice data (step B 39), and stores a character string obtained from the recognition result as an input character string 240. The information is stored in the unit 28 (step B41).

  Then, the head number (m = 1) of the similar command table 246 is selected, and this number is compared with the input character string 240 (step B43). As a result of the comparison in step B43, when it is determined that they match (step B45; Yes), the CPU 10 shifts the processing to step A17, and executes the instruction command of the similar command name corresponding to the number m (step A17). .

  On the other hand, if it is determined in step B45 that they do not match (step B45; No), the next number (m = m + 1) is sequentially selected (step B49), and the input character string 240 is compared with the number m. The CPU 10 repeats the selection and comparison of the number m, and when it is determined that the selected number m is the end of the similar command table 246 (step B47), the display unit 19 displays that the voice input is requested again. (Step B51), the process proceeds to Step B23.

  For example, after the display of the similar command selection screen 202 in FIG. 8A, when the input character string 240 “I” is acquired as a result of speech recognition, the CPU 10 corresponds to the number m = 1 in the similar command table 246. An instruction command with a similar command name “Seishiga” is executed. Specifically, the mode shifts to a still image mode for acquiring still image data from the modality M, displays a command input screen 204 as shown in FIG. 8B, and waits for input of a command name in the still image mode. To do.

  Then, when “ki” is acquired as the input character string 240 next, since the standard command name “ki” is not stored in the command table 244, the standard command name starting with “ki” is extracted, and FIG. The similar command selection screen 206 in (c) is displayed. The user selects a desired command name number from the displayed command names by voice, and starts recording of still image data as shown in FIG. 8D.

  As described above, according to the second embodiment, when the input character string 240 acquired by voice recognition in the standard mode does not match the standard command name, the similar command names are extracted from the command table 244 and listed. indicate. When the number displayed together with the similar command name is inputted by voice by the user, the instruction command for the similar command name corresponding to the number is executed.

  Thereby, even when the recognition rate of voice recognition is lowered, a command name can be selected by voice input of a short number of characters called a number from command names similar to the user's voice input. For this reason, the instruction command for the transfer device 1 can be more reliably performed, and voice input of the instruction command for the transfer device 1 can be performed more promptly even in a poor external environment.

  In the second embodiment described above, the number is displayed together with the similar command name. For example, character information such as alphabet such as “A”, “B”, “C” is displayed together with the similar command name, It goes without saying that the same effect can be obtained even if similar command names can be selected by the alphabet.

  Further, the similar command name is extracted depending on whether or not it matches the first character of the input character string 240. For example, the first and last characters of the input character string 240 and the first and last characters of the command name are extracted. A command name that matches each of the characters may be extracted as a similar command name, or a command name that matches the vowel array of the input character string 240 may be extracted as a similar command name. As described above, by changing the condition setting for extracting the similar command name, it is possible to extract a command name that approximates the voice input of the user.

  The display order of similar command names may be changed according to the frequency of use of the standard command name by the user. In this case, the standard command name in the command table 244 stores the number of times the standard command name has been selected by the user. When the extracted similar command names are displayed as a list, they are displayed in the descending order of the associated number of times. As a result, the user can easily select and input a command name that is frequently used.

  In addition, if the similar command name stored as the similar command table 246 has one similar command name of the instruction command that can be executed according to the processing state of the transfer apparatus 1, the instruction command is executed as it is. Also good. Specifically, it is assumed that a voice input is made when the command input screen 208 of FIG. 8D is displayed, and an input character string 240 of “te” is acquired. At this time, “Tensou” and “Toshi” are extracted as similar command names.

  If the command that can be executed during the recording of the still image data is “taste”, the recording of the image data is stopped as it is, and the command input screen 210 as shown in FIG. 8E is displayed. . As a result, an instruction command for the transfer apparatus 1 can be performed more quickly.

  Further, when waiting for the user's voice input, a list of command names highly relevant to the command executed immediately before by the CPU 10 may be displayed. FIG. 9 shows an example of a tree structure in which the relevance of command names is shown in a hierarchy.

  For example, as command names that are highly related to the command name “DOGA (video)”, “KIROKU (record)”, “Saisei (playback)”, “TENSO (transfer)”, “Show (erase)”, and “Cancel” Among them, “Kishi (start)” and “Cancel” are command names highly relevant to “Kirok”. These relationships are hierarchized in advance as shown in FIG.

  In the initial state of the transfer apparatus 1, the CPU 10 displays the uppermost command name as in the command input screen 212 of FIG. When “set” is acquired as the input character string 240, command names beginning with “set” among the command names in the uppermost layer, that is, “seisiga (still image)” and “set (set)”. Extracted and displayed as shown in FIG.

  For example, when the input character string 240 “Ichi” is acquired, the input character string 240 is selected based on the input character string 240 so as to shift to the still image mode as shown in FIG. Execute the command corresponding to the number. At this time, the command names below the selected command name “SEISHIGA” are “KIROK (record)”, “SEISI (play)”, “Tenso (transfer)”, “Show (erase)” and “Cancel”. Display and wait for the next voice input.

  Next, when “ki” is acquired as the input character string 240, a command name starting with “ki” is displayed as in the command input screen 218 in FIG. When the user selects “Kirok (record)” from the command names by voice input, the command names “Kai (start)” and “Cancel” are displayed as shown in FIG. 10E.

  When the CPU 10 obtains “K” as the input character string 240, the CPU 10 starts recording an instruction command with a command name starting with “K”, that is, image data of a still image, and displays a “taste (stop)” in the lower layer. 10 (f) is displayed. At this time, there is no command name in the lower layer of “taste (stop)”. In this case, when the input character string 240 “te” is acquired, the CPU 10 stops the recording of the image data, and at the same time the command names “start (stop)” and “cancel” of “taste (stop)”. "Is displayed as shown in FIG.

  As described above, since the command name selected by the user and the command name having a high related example are displayed, the user can sequentially select the command name displayed on the display unit 19 so that the transfer device 1 can be actually used. You can select command names in the order in which they are used.

The block diagram showing an example of the system configuration of a medical image output system. The block diagram which shows an example of a function structure of a transfer apparatus. (A) in a 1st embodiment is a storage part, (b) is a figure showing an example of each data composition of a command table. 6 is a flowchart for explaining a specific operation of the transfer apparatus according to the first embodiment. An example of the display screen of the transfer apparatus in 1st Embodiment. (A) is a memory | storage part in 2nd Embodiment, (b) is a command table, (c) is a figure which shows an example of each data structure of a similar command table. 9 is a flowchart for explaining specific operations of the transfer apparatus according to the second embodiment. An example of the display screen of the transfer apparatus in 2nd Embodiment. An example of a tree structure showing the relationship of command names in a hierarchy. An example of the display screen of the transfer apparatus in a modification.

Explanation of symbols

S Medical Image Output System 1 Transfer Device 9 Output Device 11 Speech Recognition Function 12 Voice Input Unit 13 A / D Converter 14 Amplifier 15 Microphone 16 Operation Input Unit 19 Display Unit 22 Communication Unit 24 Program Memory 26 Image Memory 28 Storage Unit 240 Input Character string 242 Command table 244 Command table 246 Similar command table

Claims (10)

Voice input means;
Voice recognition means for performing voice recognition of the voice input by the voice input means and converting it into character information;
Storage means for storing a command name for performing an instruction to execute a specific process relating to a medical image and a shortened command name having a shorter number of characters than the command name;
Comparing means for comparing the character information converted by the voice recognition means with the shortened command name stored in the storage means;
Command execution means for executing an instruction instruction of the command name associated with the compared shortened command name when the comparison result by the comparison means matches;
A command input device comprising: And further comprising selection means for selecting one of the first and second voice input modes,
The comparison means includes
When the first voice input mode is selected by the selection means, the character information converted by the voice recognition means is compared with the command name,
2. The command input device according to claim 1, wherein when the second voice input mode is selected, the character information converted by the voice recognition means is compared with the abbreviated command name. The selection means includes
3. The command input device according to claim 2, wherein one of the first and second voice input modes is selected based on the character information converted by the voice recognition means. Voice input means;
Voice recognition means for performing voice recognition of the voice input by the voice input means and converting it into character information;
Storage means for storing a plurality of command names for instructing execution of specific processing relating to medical images;
Extracting means for extracting a command name similar to the character information converted by the voice recognition means from the command names stored in the storage means;
A list display unit that displays a list of identification character information having a shorter number of characters than the command name in association with each command extracted by the extraction unit;
Comparison means for comparing the character information converted by the voice recognition means and the identification character information displayed in the list;
Command execution means for executing an instruction command of the command name displayed in association with the compared identification character information when the comparison result by the comparison means matches;
A command input device comprising: The extraction means includes
5. The command input device according to claim 4, wherein a command name that matches the head of the converted character information is extracted from command names stored in the storage means. The list display means includes:
6. The command input device according to claim 4, further comprising a frequency order display means for displaying a list of command names extracted by the extraction means in descending order of use frequency of the user. The list display means includes:
The apparatus further comprises a related order list display means for displaying a list of command names related to the command name of the instruction command executed immediately before by the command execution means among the command names extracted by the extraction means. Item 7. The command input device according to any one of Items 4 to 6.   The execution of the specific processing relating to the medical image includes at least one of transfer to an external device, writing to a storage unit, and display on a display unit. Command input device. Computer
Speech recognition means for performing speech recognition of speech input by the speech input means and converting it into character information;
Storage means for storing a command name for performing an instruction to execute a specific process relating to a medical image and an abbreviated command name having fewer characters than the command name;
Comparison means for comparing the character information converted by the voice recognition means with the shortened command name stored in the storage means;
Command execution means for executing an instruction instruction of the command name associated with the compared short command name when the result of comparison by the comparison means matches;
Program to function as. Computer
Speech recognition means for performing speech recognition of speech input by the speech input means and converting it into character information;
Storage means for storing a plurality of command names for instructing execution of specific processing relating to medical images;
Extraction means for extracting a command name similar to the character information converted by the voice recognition means from the command names stored in the storage means;
List display means for displaying a list in association with identification character information for each command name extracted by the extraction means;
Comparison means for comparing the character information converted by the voice recognition means with the identification character information displayed in the list;
Command execution means for executing an instruction command of the command name displayed in association with the compared identification character information when the result of comparison by the comparison means matches;
Program to function as.

Images