JP2007044239A - Medical image diagnostic apparatus, medical image processing device and medical image processing program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a technology for suitably erasing accompanying information embedded in a medical image. <P>SOLUTION: The medical image diagnostic apparatus 1 of the present invention comprises an image forming part 10 which forms a medical image based on the data reflecting inside configuration of a subject, an accompanying information extract section 15 for extracting accompanying information embedded in the image for output from the medical image, an accompanying information erasing part 17 for erasing extracted accompanying information from the image for output, a communication interface 18 for outputting the image for output from which accompanying information is erased to the outside database 30, and an image outputting interface 19 for outputting the image for output from which accompanying information is erased to the image outputting medium 40. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to a medical image diagnostic apparatus, a medical image processing apparatus, and a medical image processing program, and more specifically to a technique for erasing incidental information embedded in a medical image.

  Currently, various medical image diagnostic apparatuses such as an X-ray imaging apparatus, an X-ray CT apparatus, an MRI apparatus, an ultrasonic diagnostic apparatus, and a nuclear medicine diagnostic apparatus are widely used in the medical field. A medical image obtained by such a medical image diagnostic apparatus is displayed on a monitor together with various incidental information, and is used for interpretation work or the like.

  This “accompanying information” means information added later on the medical image, and includes character string information including a combination of characters written in any language (for example, kanji, hiragana, katakana, alphabet, numbers, etc.) It is formed by symbol information composed of various symbols (for example, punctuation marks, parentheses, colons, etc.). Note that the term “character” may be used including symbol information.

  Examples of character string information include patient personal information such as patient ID, date of birth, and sex, and examination information such as examination date, examination time, examination site, examination condition, and image number.

  When a medical image to which supplementary information is added is displayed on a monitor, generally, one of the first and second display modes described below is applied.

  In the first display mode, when the image data of the medical image acquired by the medical image diagnostic apparatus and the data of the incidental information are formed separately (see, for example, Patent Document 1), they are superimposed and output. This is a method of forming a production image and displaying the output image on a monitor. Examples of medical images displayed by this method include medical images obtained by collecting data reflecting the internal form of a subject and reconstructing the data, such as CT images and MRI images.

  On the other hand, the second display mode is a technique in which incidental information is embedded in a medical image to form an output image, which is displayed on a monitor. That is, in contrast to the first display mode in which two data (medical image and supplementary information) are overlapped at the time of image display, this second display mode is to convert the two data into a single data (output image) in advance. The difference is that it is processed and displayed on the monitor.

  An output image to which the second display mode is applied includes a captured image of a display screen when a medical image is displayed in real time. For example, on the display screen at the time of real-time diagnosis by the ultrasonic diagnostic apparatus, accompanying information such as a patient ID is simultaneously displayed together with sequentially obtained ultrasonic images. The captured image is stored as a single data in which incidental information is embedded in an ultrasonic image.

  By the way, when presenting medical images in academic societies, papers, etc., posting them on homepages, etc., or when sending medical images by mail or via the Internet, the medical images are used from the viewpoint of protecting patient privacy. It is necessary to delete the incidental information added to. Note that “deleting supplementary information” means concealing the supplementary information when displaying a medical image, that is, making the supplementary information invisible.

  For the medical image displayed in the first display mode, the image data of the medical image and the data of the incidental information are formed separately, so that only the image data is selectively subjected to display processing. Thus, it is possible to easily display the medical image from which the supplementary information has been deleted on the monitor.

  On the other hand, for the medical image displayed in the second display mode, the supplementary information is embedded in the medical image, and therefore it is not easy to selectively delete only the supplementary information from the medical image. At present, a method of manually deleting incidental information using image editing software or the like (for example, a method of painting an image area corresponding to incidental information) is applied.

  In addition, it is possible to apply a configuration in which incidental information is displayed in a specific image area in a medical image, and the incidental information is erased by automatically filling the specific image area.

JP 2005-81083 A

  However, when manually deleting the incidental information, no matter how careful the operator is, the possibility of an artificial deletion error cannot be wiped out. In particular, when ancillary deletion errors occur when incidental information is displayed overlaid on a patient's image, the image itself may be tampered with and cause misdiagnosis. Furthermore, such manual work is cumbersome and may take a long time.

  In addition, the following problems have been pointed out in the configuration in which the auxiliary information displayed in the specific image area is automatically deleted. In practice, the display position of the supplementary information may be changed, or the supplementary information to be displayed may be changed. In that case, it is necessary to designate the display area every time the auxiliary information is deleted. Therefore, although the process should have been automated, it is inconvenient because it involves manual work after all.

  In addition, there are medical images that contain both supplementary information to be deleted and supplementary information that does not need to be deleted. However, in the conventional method, only the former cannot be selectively automatically deleted, so these can be selected manually. I had to.

  Furthermore, it is better that the loss range of the medical image accompanying the deletion of the supplementary information is small, but in the conventional method, the loss range may be more than necessary. For example, in a configuration in which a specific image area is automatically filled, the entire image area is filled regardless of the actual display range of the supplementary information (for example, the length of the name).

  The present invention has been made in view of the circumstances as described above, and an object thereof is to provide a technique capable of suitably automatically erasing incidental information embedded in a medical image.

  In order to achieve the above object, the invention according to claim 1 is a medical image diagnostic apparatus having an image forming means for forming a medical image based on data reflecting an internal form of a subject, the medical image Extracting means for extracting the supplementary information embedded in the output image formed by embedding the supplementary information, erasing means for erasing the extracted supplementary information from the output image, and the supplementary information has been erased Output means for outputting an output image.

  The invention according to claim 2 is the medical image diagnostic apparatus according to claim 1, wherein the incidental information includes character string information, and the extraction unit analyzes the entire output image. The character string information embedded in the output image is extracted for each character.

  The invention according to claim 3 is the medical image diagnostic apparatus according to claim 2, further comprising contour detecting means for detecting a contour of the extracted character, wherein the erasing means is the detected Character string information is erased by changing the pixel value of the pixel in the inner region of the contour to a predetermined value.

The invention according to claim 4 is the medical image diagnostic apparatus according to claim 2, wherein the erasing unit divides the region including the extracted character and its peripheral region into a plurality of partial regions. A dividing unit; a pixel value acquiring unit that acquires a pixel value of each pixel in the partial peripheral region corresponding to the peripheral region for each partial region including the region corresponding to the character among the plurality of partial regions; Pixel value selection means for selecting a pixel value that occupies the maximum range in the partial peripheral region based on the acquired pixel value for the partial region, and the pixel value of each pixel in the partial region is selected. Erase the character by changing it to a pixel value
It is characterized by that.

  The invention according to claim 5 is the medical image diagnostic apparatus according to claim 1, wherein the output image from which the supplementary information is extracted by the extraction unit is analyzed to represent the internal form of the subject. The image processing apparatus further includes a superimposition region detecting unit that detects a superimposition region where the subject image and the extracted supplementary information overlap, and the erasing unit extracts a portion outside the detected superimposition region of the extracted supplementary information. It is characterized by selectively erasing.

  The invention according to claim 8 is the medical image diagnostic apparatus according to any one of claims 1 to 5, wherein the formed output image is a moving image including a plurality of still images. And the extraction unit extracts the supplementary information embedded in the plurality of still images of the moving image for each still image, and the erasing unit extracts the extracted information from each of the plurality of still images. The supplementary information is erased for each still image.

  The invention according to claim 11 is the medical image diagnostic apparatus according to any one of claims 1 to 10, wherein the incidental information includes one or more character string information and each character. Storage means for preliminarily storing erasure target specifying information for specifying the identification information of character string information to be erased among the one or more character string information, including identification information for identifying the type of column information; Erasure target determination means for determining whether the identification information extracted by the extraction means is specified by the erasure target specifying information, wherein the erasure means determines that the extracted identification information is specified. The character string information identified by the identification information is selectively deleted.

  According to a thirteenth aspect of the present invention, there is provided a medical image processing apparatus for processing a medical image formed based on data reflecting an internal form of a subject, and is formed by embedding incidental information in the medical image. Extraction means for extracting auxiliary information embedded in the output image, erasure means for erasing the extracted auxiliary information from the output image, and output means for outputting the output image from which the auxiliary information has been deleted And.

  The invention according to claim 14 is the medical image processing apparatus according to claim 13, wherein the supplementary information includes character string information, and the extraction means analyzes the entire output image. The character string information embedded in the output image is extracted for each character.

  The invention according to claim 15 is the medical image processing apparatus according to claim 14, further comprising contour detection means for detecting a contour of the extracted character, wherein the erasure means is detected. Character string information is erased by changing the pixel value of the pixel in the inner region of the contour to a predetermined value.

  The invention according to claim 16 is the medical image processing apparatus according to claim 14, wherein the erasing unit divides the region including the extracted character and its peripheral region into a plurality of partial regions. A dividing unit; a pixel value acquiring unit that acquires a pixel value of each pixel in the partial peripheral region corresponding to the peripheral region for each partial region including the region corresponding to the character among the plurality of partial regions; Pixel value selection means for selecting a pixel value that occupies the maximum range in the partial peripheral region based on the acquired pixel value for the partial region, and the pixel value of each pixel in the partial region is selected. The character is erased by changing to a pixel value.

  The invention according to claim 17 is the medical image processing apparatus according to claim 13, wherein the output image from which the supplementary information is extracted by the extraction means is analyzed to represent the internal form of the subject. The image processing apparatus further includes a superimposition region detecting unit that detects a superimposition region where the subject image and the extracted supplementary information overlap, and the erasing unit extracts a portion outside the detected superimposition region of the extracted supplementary information. It is characterized by selectively erasing.

  The invention according to claim 18 is the medical image processing apparatus according to any one of claims 13 to 17, wherein the formed output image is a moving image composed of a plurality of still images. And the extraction unit extracts the supplementary information embedded in the plurality of still images of the moving image for each still image, and the erasing unit extracts the extracted information from each of the plurality of still images. The supplementary information is erased for each still image.

  The invention according to claim 19 is the medical image processing apparatus according to any one of claims 13 to 18, wherein the auxiliary information includes at least one character string information and each character. Storage means for preliminarily storing erasure target specifying information for specifying the identification information of character string information to be erased among the one or more character string information, including identification information for identifying the type of column information; Erasure target determination means for determining whether the identification information extracted by the extraction means is specified by the erasure target specifying information, wherein the erasure means determines that the extracted identification information is specified. The character string information identified by the identification information is selectively deleted.

  The invention according to claim 20 is a receiving means for receiving an output image formed by embedding incidental information in a medical image formed on the basis of data reflecting the internal form of the subject; A medical image processing program that functions as an extraction unit that extracts auxiliary information embedded in a received output image and an erasing unit that deletes the extracted auxiliary information from the output image. It is.

  The invention according to claim 21 is the medical image processing program according to claim 20, wherein the incidental information includes character string information, and the extraction means is configured to process the entire received output image. Analyzing and functioning to extract character string information embedded in the output image for each character.

  The invention according to claim 22 is the medical image processing program according to claim 21, wherein the computer is further caused to function as contour detection means for detecting a contour of the extracted character, and the erasing means. Is made to function to erase the character string information by changing the pixel value of the pixel in the inner region of the detected contour to a predetermined value.

  The invention according to claim 23 is the medical image processing program according to claim 21, wherein the erasing unit divides an area composed of the extracted character and its peripheral area into a plurality of partial areas. A dividing unit; a pixel value acquiring unit that acquires a pixel value of each pixel in the partial peripheral region corresponding to the peripheral region for each partial region including the region corresponding to the character among the plurality of partial regions; The partial area is caused to function as a pixel value selection unit that selects a pixel value that occupies the maximum range in the partial peripheral area based on the acquired pixel value, and the pixel value of each pixel in each partial area is A function of erasing the character by changing to the selected pixel value is provided.

  The invention according to claim 24 is the medical image processing program according to claim 20, wherein the computer analyzes the output image from which the additional information is extracted by the extracting means, and It further functions as a superimposition region detection unit that detects a superimposition region where the subject image representing the internal form and the extracted supplementary information overlap, and the erasure unit detects the superimposition detected from the extracted supplementary information. It is characterized by functioning to selectively erase a portion outside the region.

  The invention according to claim 25 is the medical image processing program according to any one of claims 20 to 24, wherein the formed output image is a moving image composed of a plurality of still images. And the extraction means functions to extract, for each still image, the supplementary information embedded in the plurality of still images of the moving image, and the erasing means is operated from each of the plurality of still images. The extracted supplementary information is made to function so as to be erased for each still image.

  The invention according to claim 26 is the medical image processing program according to any one of claims 20 to 25, wherein the supplementary information includes at least one character string information and each character. A storage for preliminarily storing erasure target specifying information for specifying the identification information of the character string information to be erased among the one or more character string information. Means for determining whether the identification information extracted by the extraction means is specified by the erasure target specifying information, and causing the erasure means to be the extracted identification information. The character string information identified by the identification information is selectively erased when it is determined that the character string is specified.

  According to the present invention, the supplementary information embedded in the output image is extracted, and the extracted supplementary information is erased from the output image and output. Therefore, the embedded supplementary information can be suitably automatically deleted before outputting the image. Thereby, there is no erasure error as in the case of manually erasing incidental information as in the prior art.

  According to the present invention described in claim 2, claim 14, or claim 21, the entire output image is analyzed, and character string information (additional information) embedded in the output image is determined for each character. Since it is configured to extract, not only incidental information embedded at a specific position of the output image, but also characters embedded at an arbitrary position can be automatically extracted and deleted.

  According to the third, fifteenth, or twenty-second aspect of the present invention, a contour of a character extracted from an output image is detected, and a pixel value of a pixel in an inner region of the detected contour is set to a predetermined value. The character string information is erased by changing. Therefore, the range to be erased is smaller than the conventional configuration in which the entire predetermined area is filled, and the image loss range can be minimized.

  In the present invention described in claim 4, claim 16, or claim 23, first, an area composed of characters extracted from an output image and its surrounding area is divided into a plurality of partial areas, and the plurality of partial areas are Among these, the pixel value of each pixel in the partial peripheral region of each partial region including the region corresponding to the character is acquired. Then, for each partial area, a pixel value that occupies the maximum range in the partial peripheral area is selected based on the acquired pixel value, and the pixel value of each pixel in each partial area is changed to the selected pixel value. Is configured to erase characters. Therefore, characters at an arbitrary position in the output image can be deleted, and the loss range of the image can be minimized. Further, if the size of each partial area is reduced by finely dividing the target area of each character, the character can be erased according to the situation of the image around the character.

  According to the present invention described in claim 5, claim 17, or claim 24, the output image is analyzed to detect a superimposed region where the subject image and the supplementary information overlap, and the superimposition of the extracted supplementary information is detected. Since the portion outside the region is selectively erased, the information on the subject image is not changed when the auxiliary information is erased. Therefore, incidental information can be erased while reducing the possibility of misdiagnosis.

  According to the present invention described in claim 8, claim 18 or claim 25, when processing an output image consisting of a moving image, the auxiliary information embedded in a plurality of still images forming the moving image is processed. Extraction is performed for each still image, and supplementary information extracted from each of the plurality of still images is erased for each still image, so that supplementary information in each still image forming a moving image is It can be erased reliably. Also, it is possible to smoothly process an output image composed of a moving image.

  The present invention according to claim 11, 19, or 26 determines whether the identification information extracted from the output image is specified by the erasure target specifying information, and when it is determined that it is specified The character string information identified by the identification information is erased. As a result, it is possible to selectively erase only the desired supplementary information from the supplementary information (character string information) embedded in the output image by specifying the desired identification information based on the erasure target identification information. it can.

  Exemplary embodiments of a medical image diagnostic apparatus, a medical image processing apparatus, and a medical image processing program according to the present invention will be described in detail with reference to the drawings as appropriate.

  Hereinafter, first to fourth embodiments according to the present invention will be described. In the first embodiment, the auxiliary information is automatically erased by painting the outline of the auxiliary information embedded in the medical image. In the second embodiment, additional information erasure processing different from that in the first embodiment is executed, and a region in which additional information is embedded in a medical image is divided and a filling process is performed for each divided region. It is. The third embodiment relates to an incidental information erasure process when the incidental information is embedded so as to overlap an image of a subject. In the fourth embodiment, desired information is selectively deleted from the embedded supplementary information.

[Definition of terms]
First, terms used in the following explanation are defined. First, a “medical image” is an image formed by a medical image diagnostic apparatus such as an X-ray CT apparatus, an MRI apparatus, an ultrasonic diagnostic apparatus, or a nuclear medicine diagnostic apparatus based on data reflecting the internal form of a subject. Means. This medical image is composed of a subject image representing the internal form of the subject and a background image thereof. For example, in an X-ray CT apparatus, a CT image reconstructed based on projection data collected by a gantry is a “medical image”. In an ultrasonic diagnostic apparatus, an echo signal obtained by an ultrasonic scan is used as an echo signal. The ultrasonic image formed based on this is a “medical image”.

  As described above, “accompanying information” means information added after the formation of a medical image, and is a character string including a combination of characters written in any language (for example, kanji, hiragana, katakana, alphabet, numbers, etc.) Information and symbol information including various symbols (for example, punctuation marks, parentheses, colons, etc.). Note that the term “character” (character string information) may be used to include symbol information. Furthermore, the “accompanying information” includes various marks such as a logo of a medical institution and various image information such as a QR code (registered trademark) (especially, refer to the fourth embodiment). In the third embodiment, various specific examples of incidental information will be described.

  The “output image” means an image obtained by embedding incidental information in a medical image. For example, in an X-ray CT apparatus, an image formed by embedding a patient ID or the like in a CT image corresponds to an “output image”, and in an ultrasonic diagnostic apparatus, a patient ID or the like is displayed on an ultrasonic image display screen. When “” is also displayed, the captured image on the display screen corresponds to the “output image”. Note that “embedding” means replacing part of data of a medical image with data of incidental information. Therefore, the output image is composed of single image data obtained by integrating the data of the supplementary information with the image data of the medical image.

<First Embodiment>
[Device configuration]
FIG. 1 illustrates an example of an external configuration of an X-ray CT apparatus corresponding to the medical image diagnostic apparatus according to the present embodiment. A medical image diagnostic apparatus 1 shown in FIG. 1 includes a gantry 2, a bed 3, a computer 4, a monitor device 5, and an input device 6. Since such a configuration is general, only a brief description will be given. The monitor device 5 and the input device 6 are used as a console of the medical image diagnostic apparatus 1.

  A subject (not shown) is placed on the top 3 </ b> A of the bed 3. A driving device for driving the top plate 3A in the direction of the arrow in the figure is built in the bed base 3B. The subject is moved in the opening 2A of the gantry 2 together with the top 3A. In the housing of the gantry 2, an X-ray tube that generates X-rays and an X-ray detector that detects X-rays are disposed at positions facing each other with the opening 2A interposed therebetween (not shown). . The X-ray tube and the X-ray detector are rotated around the opening 2A while maintaining the facing state.

  The computer 4 controls the movement of the top 3A and the rotation of the X-ray tube and the X-ray detector, respectively, and scans the subject with X-rays. Data detected by the X-ray detector is sequentially sent to the computer 4 and converted into projection data (pre-processing). The computer 4 reconstructs a series of collected projection data to form a medical image (CT image).

  The block diagram of FIG. 2 represents an example of the configuration of the control system of the medical image diagnostic apparatus 1. The medical image diagnostic apparatus 1 is roughly divided into a configuration for performing operation control of the device, a configuration for performing output image formation processing, a configuration for performing auxiliary information erasing processing, and a configuration for performing image output processing. is doing.

[Configuration to control the operation of the device]
The operation of each unit of the medical image diagnostic apparatus 1 is controlled by the control unit 21. The control unit 21 is configured by a microprocessor such as a CPU of the computer 4. The program storage unit 22 stores various computer programs for operation control in advance. The control unit 21 appropriately reads out the program stored in the program storage unit 22 and causes each unit described later to execute the program.

  The program storage unit 22 of this embodiment stores, in particular, an OCR (Optical Character Recognition) program 22a, a contour detection program 22b, and an information erasure program 22c.

  The OCR program 22a is a program for reading each character of the character string information in the accompanying information. As this OCR program 22a, known OCR software can be used (for example, refer to Japanese Patent Application Laid-Open No. 2004-265293).

  The contour detection program 22b is a program for detecting the contour of each character read based on the OCR program 22a. As this contour detection program 22b, a known contour detection program (for example, see Japanese Patent Laid-Open No. 10-305033 regarding an ultrasonic image) can be used, and a character is obtained using a reading result based on the OCR program 22a. It is also possible to use a program for detecting the contour of the. In the latter case, for example, a program for searching for and acquiring the coordinates of a pixel corresponding to the contour portion of the detected character can be applied.

  The information erasure program 22c is a program for erasing the character by painting the inside of the contour detected based on the contour detection program 22b. The information erasing program 22c erases the character by changing the pixel value of the pixel in the inner region of the character outline to a predetermined value.

  In order to erase the character, that is, to make the character invisible, it is necessary to make this “predetermined value” substantially equal to the pixel value of the pixel in the external vicinity region of the contour (the character). is there. For this purpose, the pixel value of the background image (described above) of the medical image can be used as the “predetermined value”, or the pixel value of the external vicinity region of the contour is acquired and the pixel value is set to the “predetermined value”. Can also be used.

[Configuration to perform output image formation processing]
The output image forming process is performed by the image forming unit 10, the auxiliary information embedding unit 11, and the image storage unit 12.

  The image forming unit 10 corresponds to an example of the “image forming unit” of the present invention, and the gantry 2 for forming a medical image (CT image) in the manner described above, the bed 3, and preprocessing and reconstruction processing of the computer 4 It is comprised including the part related to.

  The auxiliary information embedding unit 11 embeds predetermined auxiliary information in the medical image formed by the image forming unit 10 to form an output image. The auxiliary information embedding unit 11 forms an output image by, for example, creating a capture image of a display screen when a medical image is displayed on a monitor together with auxiliary information. The supplementary information embedding unit 11 may perform image processing for embedding predetermined supplementary information on the formed medical image. The incidental information embedding unit 11 may be a microprocessor mounted on the computer 4 or an external computer.

  The image storage unit 12 stores the output image formed by the incidental information embedding unit 11. Each output image is given a unique file name. The image storage unit 12 is configured by a storage device such as a hard disk drive built in the computer 4, for example, but can also be configured by an external storage device.

[Configuration to delete incidental information]
The supplementary information erasing process is performed by the image reading unit 13, the image determining unit 14, the supplementary information extracting unit 15, the contour detecting unit 16, and the supplementary information erasing unit 17. Each of these units 13 to 17 is configured by a microprocessor mounted on the computer 4.

  The image reading unit 13 accesses the image storage unit 12 in accordance with an instruction from the control unit 21 and executes a process of reading the output image stored therein. The instruction (control signal) from the control unit 21 is attached with an image ID such as a file name of the output image to be read, and the image reading unit 13 stores the output image corresponding to the image ID as an image. Search from section 12 and read. The image reading unit 13 sends the read output image to the image determination unit 14. The image reading unit 13 corresponds to an example of “accepting unit” of the present invention.

  The image discriminating unit 14 corresponds to an example of the “image discriminating unit” of the present invention, operates in response to an instruction from the control unit 21, and the output image from the image reading unit 13 is a still image or a moving image. Determine if it is an image. This determination processing is performed based on, for example, the type of extension added to the file name of the output image. That is, various extension names indicating still images and various extension names indicating moving images are stored in advance, and the extension names given to the output images to be determined are stored as the stored extension names. By collating, it is determined whether the output image is a still image or a moving image.

  Note that information used to determine whether the output image is a still image or a moving image is not limited to the extension name. For example, arbitrary information (image type identification information) for image type identification is attached to the output image, and the image type identification information is used to determine whether the output image is a still image or a moving image. It is possible to constitute based on the above. Here, the image type identification information is the header of the data of the output image when the output image is transmitted from the image forming unit 10 to the image storage unit 12 or stored in the image storage unit 12. It comes to be attached to the part.

  The incidental information extraction unit 15 operates in accordance with the aforementioned OCR program 22a, analyzes the entire output image, and reads the embedded incidental information (particularly, each character of the character string information). Here, reading a character or detecting a character may be referred to as extracting a character. Here, when the image determination unit 14 determines that the output image is a moving image, the auxiliary information extraction unit 15 reads each character in the still image for each of the plurality of still images forming the moving image. Are executed individually. This incidental information extraction unit 15 corresponds to an example of “extraction means” of the present invention.

  The contour detection unit 16 corresponds to an example of the “contour detection unit” of the present invention, operates according to the above-described contour detection program 22b, and detects the contour of each character read by the supplementary information extraction unit 15. When the output image is a moving image, the contour detection unit 16 also individually processes the plurality of still images.

  The incidental information erasure unit 17 operates in accordance with the information erasure program 22c described above, and the pixel value of the pixel in the inner area of the contour detected by the contour detection unit 16 is changed to the pixel value (or a pixel substantially equal to the pixel value in the outer vicinity region of the contour). Value) and delete the character from the output image. When the output image is a moving image, the incidental information erasing unit 17 also individually processes the plurality of still images. The incidental information erasure unit 17 corresponds to an example of the “erasure unit” of the present invention.

[Configuration for image output processing]
The output image from which the supplementary information has been deleted is output to the monitor device 5, the image database 20, the external database 30, the image output medium 40, etc. of the medical image diagnostic apparatus 1.

  The image database (DB) 20 is a database built in or externally attached to the medical image diagnostic apparatus 1. In addition to an output image, a medical image formed by the image forming unit 10, projection data, patient information, and the like Various data are stored.

  The external database (DB) 30 is a database connected to the medical image diagnostic apparatus 1 via a network N such as a LAN or the Internet. The external database 30 is, for example, a DICOM (Digital Imaging and Communications in Medicine) standard database relating to medical digital images and communication.

  Output of the output image to the external database 30 is performed by a communication interface (I / F) 18 including a modem or the like corresponding to the network N. The communication interface 18 is a communication unit that communicates with an external device via a network, and corresponds to an example of an “output unit” according to the present invention.

  The image output medium 40 is various media on which output images are output, such as various media such as DVD-RAM and DVD-R, film media, and paper media. The output process to the image output medium 40 is performed by an image output interface (I / F) 19. The image output interface 19 includes a drive device compatible with the above media, a printer device that prints and outputs on a film medium or a paper medium, and the like. The image output interface 19 is an example of the “output unit” in the present invention.

  Although not shown, another computer is connected on the network N. The medical image diagnostic apparatus 1 can transmit an output image to the computer via the communication interface 18. This configuration is used when obtaining a second opinion for the output image or when the output image is used jointly by a plurality of doctors.

[Operation]
The operation of the medical image diagnostic apparatus 1 of the present embodiment having the above configuration will be described. The flowchart in FIG. 3 shows an example of the operation of the medical image diagnostic apparatus 1. FIG. 4 shows the transition of the image mode by the processing shown in FIG.

  First, the image forming unit 10 forms a medical image of the subject (S1). FIG. 4A shows an example of the form of a medical image to be formed.

  The incidental information embedding unit 11 embeds predetermined incidental information in this medical image to form an output image (S2).

  FIG. 4B shows an example of an output image obtained by embedding incidental information such as information on a subject and identification information for identifying the type in the medical image of FIG. 4A. . In addition, according to the above-mentioned definition, identification information is also a kind of incidental information.

  In the output image of the figure, as identification information, character string information “name” indicating a subject's name, character string information “ID” indicating subject ID information (patient ID), and a medical image at the time of forming a medical image. Character string information “kV” indicating the tube voltage of the X-ray tube and character string information “mA” indicating the tube current are respectively embedded. Further, as incidental information corresponding to each identification information, character string information “Toshiba Taro” corresponding to the subject name, character string information “1048” corresponding to the subject ID information, and a character string corresponding to the tube voltage Information “120” and character string information “50” corresponding to the tube current are respectively embedded. Note that a colon “:”, which is symbol information, is embedded between each piece of identification information and the incidental information corresponding to the identification information in order to clearly indicate the corresponding relationship.

  The medical image in which the supplementary information is embedded, that is, the output image is stored in the image storage unit 12 (S3).

  The image reading unit 13 accesses the image storage unit 12 in accordance with an instruction from the control unit 21 and acquires an output image (S4). The acquired output image is sent to the image discrimination unit 14. The image discriminating unit 14 discriminates whether the output image is a still image or a moving image (S5). The determination result is sent to the incidental information extraction unit 15 together with the output image.

(When the output image is a still image: S5; “still image”)
If the determination result is “still image”, the supplementary information extraction unit 15 analyzes the entire still image (= output image) and extracts each character embedded in the image (S6). For example, for the output image in FIG. 4B, “Mr.”, “Name”, “:”, “East”, “Shiba”, “Thick”, “Buro”, “I”, “D” , “:”, “1”, “0”, “4”, “8”, “k”, “V”, “:”, “1”, “2”, “0”, “m”, “ The characters “A”, “:”, “5”, and “0” are extracted.

  Subsequently, the contour detection unit 16 detects the contour of each character extracted by the supplementary information extraction unit 15 (S7). The detection result is acquired as the coordinates of the pixel located in the outline portion of the character.

  Next, the supplementary information erasure unit 17 changes the pixel value of each pixel in the inner region of the contour detected by the contour detection unit 16 to the pixel value of the outer neighboring region of the contour. Thereby, each character in the output image is deleted, and as a result, the incidental information is deleted (S8).

  FIG. 4C shows a state in which the auxiliary information in the output image of FIG. 4B is erased (in FIG. 4C, the erase range is clearly indicated by diagonal lines). The hatched area in the output image indicates the inner area of the outline of each character detected by the outline detection unit 16. The pixel value of the pixels in the shaded area is actually changed to a value equivalent to the external neighborhood area (for example, background image) of the contour, and the shaded area (that is, the incidental information) is not visible. .

  This completes the incidental information erasure process when the output image is a still image. The output image from which the supplementary information has been deleted is stored in the image database 20, displayed on the monitor device 5 as necessary, transmitted to the external database 30, or output to the image output medium 40.

(When the output image is a moving image: S5; “moving image”)
When it is determined that the output image is a moving image, the supplementary information extraction unit 15 first divides the output image into a plurality of still images (frames) (S11). The incidental information extraction unit 15 sequentially analyzes the plurality of frames for each frame, and extracts each embedded character (accompanying information) (S12). The frames for which the auxiliary information extraction processing has been completed are sequentially sent to the contour detection unit 16 together with the extracted characters.

  The contour extracting unit 16 sequentially detects the contours of the extracted characters for the frames sequentially sent from the supplementary information extracting unit 15 (S13). The frames for which the contour detection processing has been completed are sequentially sent to the incidental information erasure unit 17 together with the detection information.

  The supplementary information erasure unit 17 sequentially erases the supplementary information by changing the pixel value of each pixel in the inner region of the detected contour to the pixel value of the outer neighboring region for the frames sequentially sent from the contour detection unit 16. (S14).

  This completes the incidental information erasure processing when the output image is a moving image. The output image from which the supplementary information has been deleted is stored in the image database 20, displayed on the monitor device 5 as necessary, transmitted to the external database 30, or output to the image output medium 40.

[Function and effect]
As described above, the medical image diagnostic apparatus 1 according to the present embodiment automatically extracts the supplementary information (character string information) embedded in the medical image for each character and automatically detects the outline of each character. Then, the incidental information is deleted by automatically changing the pixel value of the pixel in the inner region of the outline to a value (substantially) the same as the pixel value in the vicinity thereof. Accordingly, there is no erasure error as in the case of manual operation as in the prior art.

  In the character extraction process, by extracting the characters by analyzing the entire output image, not only the supplementary information embedded at a specific position in the image but also the character embedded at an arbitrary position is automatically Can be extracted and erased.

  As described above, according to the present embodiment, the incidental information in the output image can be suitably automatically deleted. Therefore, particularly before the image for output is transferred over the network, output to various image output media, or stored in the database of the apparatus itself, additional information (character string) such as subject name and subject ID information Information) can be easily deleted, and personal information in secondary use of the output image can be protected easily and reliably.

  Further, according to the present embodiment, only the area inside the outline of each character in the output image is erased, so the erasure range is small compared to the conventional configuration in which the entire predetermined area is filled, and the necessary minimum level. That's it. As a result, there is an advantage that the loss range of the image accompanying the erasure of the auxiliary information can be minimized. In particular, when the incidental information overlaps the subject image, the loss range of the subject image is smaller than before, so that the influence on interpretation is reduced and the possibility of occurrence of a diagnostic error can be reduced. .

  In addition, it automatically determines whether the output image is a still image or a moving image, and in the case of a moving image, the same processing as the still image is executed for each frame. Whether the image is a still image or a moving image can be processed smoothly. Furthermore, since the supplementary information is erased for each frame, even when there is a frame in which supplementary information different from other frames is embedded, or there is a frame in which the supplementary information is embedded in a different position The incidental information of each frame can be erased accurately. Therefore, personal information of the subject can be reliably protected.

[Modification]
Various modifications of the first embodiment as described above will be described below.

[Modification 1]
In the above embodiment, an outline is detected for each extracted character and the inside of the outline is filled. However, a character string composed of two or more adjacent characters is extracted from a plurality of extracted characters. It can also be configured to erase as a unit.

  That is, the incidental information extraction unit 15 extracts each character in the image in the same manner as in the above embodiment, and the contour detection unit 16 performs a character string (for example, a character string including two or more adjacent characters among the plurality of extracted characters). , “Toshiba Taro”) in FIG. 4 is detected.

  At that time, the contour detection unit 16 recognizes the character string as one object and detects the contour. As a technique for recognizing a character string as a single object, for example, in the example of FIG. 4, a plurality of characters following the symbol “:” (that is, at the right side of the symbol “:”) are recognized as a single object. be able to.

  The incidental information erasure unit 17 changes the pixel value of each pixel in the inner region of the outline of the character string recognized as a single object to a predetermined value (pixel value in the outer neighboring region of the contour).

  According to this modification, compared with the conventional configuration in which the specific range in the output image is uniformly filled, the loss range of the image accompanying the erasure of the accompanying information can be reduced. For example, if an area of 8 full-width characters is secured as a display area for displaying the name of the subject, in the case of “Toshiba Taro” in FIG. 4, the entire display area is filled in the conventional configuration. According to the modified example, only an area of 4 full-width characters corresponding to “Toshiba Taro” is filled.

  Note that the number of characters in the character string recognized as a single object and the recognition method thereof are not limited to the above-described contents.

  For example, in the example of FIG. 4, it is possible to obtain the number of characters (4 characters) of “Toshiba Taro” and fill an area (rectangular area) for 4 characters out of a display area for 8 characters.

  In addition, when there is a space in one type of character string information, for example, when the subject name “Toshiba Taro” is embedded, a plurality of partial character strings separated by the space (“Toshiba”, “Taro”) can be recognized as single objects, and each can be deleted.

[Modification 2]
In the above embodiment, all the incidental information in the output image is deleted, but this is not always necessary. For example, in the example of FIG. 4, it is possible to apply a configuration in which only “Toshiba Taro” is selectively deleted from the character string information “Name: Taro Toshiba”. That is, since the character string “name” represents the type of information that follows, there is no problem even if it is disclosed, but the character string “Toshiba Taro” is the subject of the subject of the medical image. Since it is a name (personal information), it must be concealed when disclosed to others.

  Such selective erasure can be realized by the following method, for example. First, when embedding supplementary information, a specific symbol or character (separator information) such as a symbol “:” between a character string indicating the type of information and a character string of actual information corresponding to the type Set to place. As this delimiter information, information not used for other purposes is applied.

  When the accompanying information extraction unit 15 extracts the delimiter information, the contour detection unit 16 detects the contour of the character following the delimiter information. At this time, it is not necessary to detect the outline of the character before the delimiter information. The supplementary information erasing unit 17 fills the inner area of the character outline following the delimiter information. Thereby, the characters following the delimiter information are erased, while the information before the delimiter information is embedded without being painted.

[Modification 3]
When the output image is a moving image, the same incidental information is often embedded in each frame. In that case, it is desirable to delete the same area of each frame when deleting the incidental information of each of the plurality of frames.

  For example, with respect to the character string information “name: Toshiba Taro” in the example of FIG. 4, the entire “name: Toshiba Taro” is deleted in a certain frame, and only “Toshiba Taro” is deleted in another frame. When the moving image thus obtained is reproduced and displayed, the “name:” portion may or may not be displayed, making it difficult to see the image.

  To avoid such inconvenience, when the same incidental information (“Name: Toshiba Taro”) is embedded in each frame, the same area (eg “Toshiba Taro”) of each frame is deleted. To do.

  In the third embodiment to be described later, a configuration is disclosed in which the erase regions of the respective frames are matched when the auxiliary information is embedded so as to overlap the subject image.

[Modification 4]
The medical image diagnostic apparatus 1 according to the above embodiment can transfer an output image to the external database 30 or an external computer over the network, or output it to various image output media 40. Such processing should be executed after deleting the auxiliary information in the output image. Therefore, in order to avoid an output error of an output image whose auxiliary information has not yet been deleted, the following configuration can be applied.

(Configuration example 1)
The control unit 21 sets a flag for the output image from which the supplementary information is erased by the supplementary information erasing unit 17 and stores the flag in the image database 20. When the operator operates the input device 6 or the like to specify a desired output image and requests transfer to the external database 30 or output to the image output medium 40, the control unit 21 specifies the specified output image. It is determined whether or not a flag is set on the image.

  When the flag is set, the control unit 21 controls the communication interface 18 or the image output interface 19 to transfer or output the output image in accordance with an operator request.

  On the other hand, when the flag is not set, the control unit 21 displays a confirmation message such as “Personal information has not been deleted but will it be transferred (output)?” And soft keys “OK” and “Cancel”. Is displayed on the monitor device 5. When “OK” is selected (clicked), the control unit 21 transfers or outputs the output image. Conversely, if “cancel” is selected, the control unit 21 cancels the original request.

  As another processing mode when the flag is not set, for example, a confirmation message such as “Do you want to transfer (output) after deleting personal information?”, Soft keys “OK” and “Cancel” Can be displayed. If “OK” is clicked, the personal information is deleted and then the output image is transferred (output). If “Cancel” is clicked, the original request is canceled. Is possible.

(Configuration example 2)
The control unit 21 sets a flag in the same manner as in (Configuration Example 1). When the operator requests transfer or output of the output image, the control unit 21 determines whether or not there is a flag for the output image.

  When there is a flag, the control unit 21 transfers or outputs the output image according to the request. On the other hand, if there is no flag, the request is rejected and a dialog box with a message indicating that is displayed on the monitor device 5.

  Each of the above-described modified examples 1 to 4 can be arbitrarily applied to second to fourth embodiments described later as necessary.

<Second Embodiment>
The present embodiment discloses an incidental information erasing process different from the first embodiment.

[Device configuration]
The medical image diagnostic apparatus of this embodiment has the same external configuration as that of the first embodiment (see FIG. 1). The control system is substantially the same as that of the first embodiment (see FIG. 2), but the present embodiment is different in that the contour detection unit 16 is unnecessary and the configuration of the incidental information erasure unit 17 is different. This is different from the first embodiment.

  FIG. 5 shows an example of the internal configuration of the incidental information erasure unit 17 of the present embodiment. The supplementary information erasing unit 17 includes an area dividing unit 17a, a pixel value acquiring unit 17b, a pixel value selecting unit 17c, and a pixel value changing unit 17d. Further, data (display image, supplementary information) is directly input from the supplementary information extracting unit 15 to the supplementary information erasing unit 17.

(Region division)
The area dividing unit 17a corresponds to an example of the “dividing unit” of the present invention, and for each character extracted by the supplementary information extracting unit 15, a plurality of areas (referred to as target areas) composed of a character and its surrounding area are provided. It works to divide into partial areas.

  FIG. 6A shows a target area R composed of the letter A and its surrounding area. The target region R includes a color 1 region R1 and a color 2 region R2 corresponding to the peripheral region of the character A. Color 1 and color 2 are different colors. In other words, the pixels in the region R1 and the pixels in the region R2 have different pixel values. The character A is displayed in a color different from the colors 1 and 2 (that is, the pixels in the character A have different pixel values from the regions R1 and R2 of the colors 1 and 2). Hereinafter, the pixel value of the letter A is pa, the pixel value of the color R region R1 is p1, and the pixel value of the color R region R2 is p2.

  The target area R has, for example, a size obtained by adding a predetermined number of pixels in the vertical and horizontal directions of the peripheral area to the top, bottom, left, and right of the extracted character A, respectively. That is, assuming that the number of vertical pixels of character A is M, the number of horizontal pixels is N, the number of vertical pixels in the peripheral area is m, and the number of horizontal pixels is n, the vertical direction of the target area R The size is (M + 2m) pixels, and the horizontal size is (N + 2n) pixels. Here, the size of the smallest rectangle circumscribing the character A is defined as the size of the character A.

  The setting of the size of the target area and the definition of the character size are not limited to the above. For example, a character display space (such as a preset full-width character display space) can be set as the target region. Further, as the character size, the size of the outer periphery of the character can be applied.

  The area dividing unit 17a divides the target area R into a plurality of partial areas r1 to rk (where k = kv × kh; kv = the number of divisions in the vertical direction, kh = the number of divisions in the horizontal direction). Each partial area has a size of about several pixels vertically and horizontally. In the present embodiment, each of the partial regions r1 to rk is rectangular, but may be any shape.

  FIG. 6B shows an aspect of a partial region group R ′ composed of four partial regions rα, rβ, rγ, rδ that are a part of the partial regions r1 to rk. The partial region rα includes a region Aα corresponding to a part of the character A and a partial peripheral region R1α corresponding to a part of the color 1 region R1. The partial region rβ includes a partial peripheral region R1β corresponding to a part of the color 1 region R1 and a partial peripheral region R2β corresponding to a part of the color 2 region R2. The partial region rγ includes a region Aγ corresponding to a part of the character A, a partial peripheral region R1γ corresponding to a part of the color 1 region R1, and a partial peripheral region R2γ corresponding to a part of the color 2 region R2. And are included. The partial region rδ includes a region Aδ corresponding to a part of the character A and a partial peripheral region R2δ corresponding to a part of the color 2 region R2.

  Note that, in a partial area including a plurality of partial peripheral areas having different colors, such as the partial areas rβ and rγ, the whole of the plurality of partial peripheral areas may be referred to as a “partial peripheral area”.

(Pixel value acquisition unit)
The pixel value acquisition unit 17b corresponds to an example of the “pixel value acquisition unit” of the present invention. For each of the plurality of partial areas, each pixel in the partial peripheral area (that is, an area other than a part of the character in the partial area). Get the pixel value of.

  At this time, referring to the extraction result by the supplementary information extraction unit 15, it is desirable that a partial region that does not include any pixel corresponding to a character is not a target of pixel value acquisition processing (and subsequent processing). Further, the pixel value selection unit 17c corresponds to a partial region in which all pixels correspond to characters (described later).

  In the case of FIG. 6B, the pixel value acquisition unit 17b performs the following processing on the partial regions rα, rβ, rγ, rδ. The partial region rβ is excluded from the processing target because there is no pixel corresponding to the character.

  First, the pixel value acquisition unit 17b acquires the pixel value of each pixel for the partial region rα. In particular, the pixel value of the area excluding the area Aα corresponding to the character A is acquired. As a result, the pixel value p1 of each pixel corresponding to the partial peripheral region R1α is obtained.

  Similarly, pixel values of regions excluding the regions Aγ and Aδ corresponding to the character A are acquired for the partial regions rγ and rδ, respectively. As a result, for the partial region rγ, the pixel value p1 in each pixel in the partial peripheral region R1γ and the pixel value p2 in each pixel in the partial peripheral region R2γ are acquired. For the partial region rδ, the pixel value p2 at each pixel in the partial peripheral region R2δ is acquired.

(Pixel value selection part)
The pixel value selection unit 17c corresponds to an example of the “pixel value selection unit” of the present invention. For each partial region, the pixel value that occupies the maximum range in the partial peripheral region among the pixel values acquired by the pixel value acquisition unit 17b. The process of selecting is performed.

  (1) For this purpose, the pixel value selection unit 17c first executes a classification process of the acquired pixel values. This classification process is performed by extracting different values from the acquired pixel values. 6B, only the pixel value p1 is extracted, the pixel values p1 and p2 are extracted for the partial region rγ, and only the pixel value p2 is extracted for the partial region rδ (partial region rβ). Is excluded from processing.)

  (2) Next, for each pixel value extracted by the classification process, the pixel value selection unit 17c obtains the number of pixels in the partial area having the pixel value. At this time, it is sufficient to perform the process of obtaining the number of pixels only for the partial area from which a plurality of pixel values are extracted in the classification process.

  For the partial regions rα and rδ in FIG. 6B, only one type of pixel value is extracted, so that the processing is not performed. For the partial region rγ, since the extracted pixel values are plural (p1, p2), the number of pixels having the pixel value p1 and the number of pixels having the pixel value p2 are respectively obtained.

  Here, in the partial region rγ, the numbers of pixels having pixel values p1 and p2 are P1γ and P2γ, respectively (P1γ <P2γ).

  In consideration of the proportional relationship between the number of pixels and the range (area) of the image, the number of pixels having the pixel value p1 in the partial region rγ corresponds to the range of the partial peripheral region R1γ, and the pixel value p2 is The number of pixels included corresponds to the range of the partial peripheral region R2γ. P1γ <P2γ means that the range of the partial peripheral region R2γ is larger than the range of the partial peripheral region R1γ.

  (3) Further, the pixel value selection unit 17c selects a pixel value having the maximum number of pixels for each partial region.

  Specifically, for a partial region from which only one pixel value has been extracted in the classification process, the single pixel value is selected. In addition, for a partial region from which a plurality of pixel values are extracted, a pixel value having the maximum number of pixels having that value is selected.

  In addition, when there are two or more pixel values with the maximum number of pixels (that is, when the number of pixels of two or more pixel values is equal and maximum), all of those pixel values, or of them Select one. In the latter case, for example, the pixel value for the partial area is selected with reference to the selection result for the adjacent partial area.

  For a partial area that does not include a partial peripheral area (that is, a partial area in which the entire area corresponds to a character), for example, a pixel value having the maximum number of pixels in the adjacent partial area is selected.

  In the case of FIG. 6B, the pixel value p1 is selected for the partial region rα, the pixel value p2 is selected for the partial region rγ (P1γ <P2γ), and the pixel value p2 is selected for the partial region rδ. . The partial region rβ is not subject to processing.

(Pixel value change part)
For each partial region, the pixel value changing unit 17d changes the pixel value of each pixel in the region corresponding to the character to the pixel value selected by the pixel value selecting unit 17c.

  In the case of FIG. 6B, the pixel value of each pixel in the partial region rα is changed to the pixel value p1, the pixel value of each pixel in the partial region rγ is changed to the pixel value p2, and each pixel in the partial region rδ is changed. The pixel value is changed to the pixel value p2. The partial region rβ remains as it is.

  As a result, as shown in FIG. 6C, the partial area group R ′ is changed to the partial area R1α ′ of the color 1 having the pixel value p1, and the partial area rβ remains unchanged. A partial peripheral region R1β of color 1 and a partial peripheral region R2β of color 2, the entire region of the partial region rγ is changed to the partial region R2γ ′ of color 2 of the pixel value p2, and the entire region of the partial region rδ is changed to the pixel value p2. Color 2 partial region R2δ ′.

[Operation]
The supplementary information erasing process executed by the medical image diagnostic apparatus of the present embodiment as described above will be described with reference to the flowchart of FIG. In the figure, the same steps as those in the first embodiment are denoted by the same reference numerals.

  First, as in the first embodiment, the image forming unit 10 forms a medical image of the subject (S1), embeds predetermined incidental information in the medical image to form an output image (S2), The output image is stored in the image storage unit 12 (S3). When the image reading unit 13 reads the output image (S4), it is determined whether the image is a still image or a moving image (S5).

(When the output image is a still image)
In the case of a still image, first, as in the first embodiment, each character embedded in the output image is extracted (S6). Each extracted character is sent to the incidental information erasure unit 17 together with the output image.

  The area dividing unit 17a of the incidental information erasing unit 17 sequentially divides the target area composed of the character and its surrounding area into a plurality of partial areas for each extracted character (S21). Each partial region is sequentially sent to the pixel value acquisition unit 17b.

  The pixel value acquisition unit 17b acquires the pixel value of each pixel in the partial peripheral region (region other than a part of the character) in each partial region (S22). The pixel value acquisition unit 17b sequentially performs this process for each character extracted in step S6. The acquired pixel value data is sequentially sent to the pixel value selection unit 17c for each partial region.

  For each partial region, the pixel value selection unit 17c selects a pixel value that occupies the maximum range in the partial peripheral region from among the pixel values acquired by the pixel value acquisition unit 17b (S23). The pixel value selection unit 17 sequentially performs this process for each extracted character. The data of the selected pixel value is sequentially sent to the pixel value changing unit 17d for each partial area.

  The pixel value changing unit 17d changes the pixel value of each pixel in each partial region to the pixel value selected by the pixel value selecting unit 17c (S24). This process is sequentially executed for each extracted character.

  Thereby, the pixel value of the pixel in each partial area of the output image is changed to the pixel value that occupies the widest range in the partial peripheral area of the partial area. In particular, the pixel value of the region corresponding to the character in each partial region is changed to the pixel value that occupies the widest range in the peripheral region of the character, and as a result, the character is erased.

(When the output image is a movie image)
On the other hand, when the output image is a moving image, the output image is divided into a plurality of frames (S11), and each embedded character is extracted for each frame (S12). The frames for which the supplementary information extraction processing has been completed are sequentially sent to the supplementary information erasure unit 17 together with the extracted characters.

  The area dividing unit 17a, the pixel value acquiring unit 17b, the pixel value selecting unit 17c, and the pixel value changing unit 17d of the incidental information erasing unit 17 execute the processes of steps S21 to S24 for each frame (S31 to S34).

  With the above processing, for each frame forming the output image, the pixel value of the pixel in each partial area (particularly, the area corresponding to the character) is changed to the pixel value that occupies the widest range in the partial peripheral area. The As a result, the characters in each frame are deleted, and as a result, the characters in the output image (moving image) are deleted.

[Function and effect]
As described above, the medical image diagnostic apparatus according to the present embodiment automatically extracts incidental information (character string information) embedded in a medical image for each character, and sets a target region including the character as a plurality of partial regions. The auxiliary information is erased by painting each partial area with a color that occupies the maximum range in the area other than characters. Accordingly, there is no erasure error as in the case of manual operation as in the prior art, and characters at arbitrary positions in the image can be erased. Further, since only the character area in the output image is erased, the erasure range of the minimum necessary is sufficient.

  Further, even when the output image is a moving image, the incidental information can be erased smoothly and reliably. When the output image is a moving image, it is desirable to delete the same portion of the incidental information in each frame in consideration of easy viewing of the processed moving image.

  Further, as an effect peculiar to the present embodiment, if the division of the target area of each character is made fine, the size of each partial area is reduced, so that the area filled with characters is suitably adapted to the situation of the surrounding image. be able to.

  However, for characters embedded in a position that overlaps the subject image (especially the region of interest), the color of the region corresponding to the character is changed according to the surrounding color, causing the subject image to be falsified and misdiagnosis Care must be taken because it may lead to Note that how to deal with characters embedded in a position overlapping the subject image will be described in a third embodiment to be described later.

[Modification]
In the above embodiment, the pixel value changing unit 17d changes the pixel value of the entire partial area. However, only the pixel value of the area corresponding to the character may be changed.

  In addition, instead of setting the target area for each extracted character, it recognizes a character string consisting of two or more adjacent characters as a single object, sets the target area including its peripheral area, and sets the target area. It is possible to divide the area into a plurality of partial areas for processing.

<Third Embodiment>
In the present embodiment, a configuration for suitably erasing incidental information embedded at a position overlapping a subject image is disclosed.

[About the superimposed state of the subject image and incidental information]
An output image in which the subject image and the accompanying information overlap will be described. 8 and 9 show examples of such output images (FIG. 9 is an enlarged image of the image of FIG. 8). Various incidental information is embedded in the output image shown in both figures. First, the supplementary information will be briefly described.

  The following supplementary information is embedded in the upper left position of the output images in both figures: patient ID (subject ID information) “226673-1”; imaging region and display region diameter “M: (320.00)” Study ID “69: 3: 1”; incidental information change mark “*”; bed angle “−10.00 mm”; tilt angle “+ 0.0D”. The study ID is identification information of “study” that represents an examination unit in the DICOM hierarchical structure.

  In the upper left position, the following supplementary information is embedded: patient name (name of subject) “TOSHIBA TARO”; imaging date and imaging time “2004.07.12 17: 50: 30.549”; Tube voltage and tube current time product “120 kV / 150 mAs”; scan time, slice thickness and imaging slice thickness “0.50 s / 1.0 mm / 1.0 × 16”; helical pitch “HP 15.0”.

  In the lower left position, the following auxiliary information is embedded: window level “WL = 0”; window width “WW = 300”; contrast agent name, concentration, injection rate (amount 1, amount 2) “omnipaque1” / 100/150 (300) / 500 (250); contrast agent name “omnipaque2”.

  In the lower right position, the following supplementary information is embedded: patient comments “1-5A”, [comment2], “comment3”, “comment4”; collection comments “Test1”, “Test2”, “Test3” "Test4"; age and gender "30Y / M"; patient position, patient insertion direction and image viewing direction "SU / HF / VFF"; helical interpolation type, reconstruction function, filter name and APMC "INTERP-5 / FC60 / ORG / S / ". Here, APMC indicates body motion correction processing.

  Also, a body position mark “R” is embedded in the left center position, and a body position mark “P” is embedded in the lower center position. Furthermore, a 5 cm scale having a scale of 5 cm unit is embedded in the background of the patient comment in the lower right position.

  In FIG. 8, patient comments “1-5A”, [comment2], “comment3”, and “comment4” in the lower right position are embedded at positions overlapping the subject image. Further, in FIG. 9 which is an enlarged image of the subject image, all incidental information is embedded at a position overlapping the subject image.

[Device configuration]
The medical image diagnostic apparatus of this embodiment has the same external configuration as that of the first embodiment (see FIG. 1). The control system is substantially the same as that of the first embodiment (see FIG. 2), but the present embodiment is provided with a configuration for dealing with a region (superimposed region) where the subject image and the auxiliary information overlap. The form is different from the first embodiment.

  FIG. 10 is a block diagram illustrating an example of a control system of the present embodiment. The medical image diagnostic apparatus according to the present embodiment includes a superimposed region detection unit 23 and a superimposed region selection unit 24. The processing by the incidental information erasure unit 17 will be described in the section [Operation].

(Superimposed area detector)
The superimposition region detection unit 23 corresponds to an example of the “superimposition region detection unit” of the present invention, analyzes the output image from which the supplementary information is extracted by the supplementary information extraction unit 15, and extracts the subject image and the subject image. A process of detecting a superimposition area with the accompanying information is performed. For this purpose, the overlapping area detection unit 23 executes, for example, the following process.

  First, with reference to the coordinates of each pixel corresponding to each extracted supplementary information, a range of each supplementary information (that is, a distribution range of pixels for displaying each supplementary information) is acquired.

  Next, a range of the subject image (that is, a distribution range of pixels for displaying the subject image) is detected based on the pixel value of each pixel of the output image. As an example of this detection process, the pixels of the output image are sequentially referenced from the image edge, for example, in the horizontal direction, and the coordinates of the pixels that change from the pixel value of the background image to other pixel values are detected. The pixel arrangement obtained thereby corresponds to the contour of the subject image.

  Further, the pixel distribution range of each supplementary information acquired first and the pixel distribution range of the detected subject image are then referred to, and the overlapping range is obtained. Thereby, for each incidental information, a distribution range of pixels in the overlapping region with the subject image is obtained.

  For example, since the patient name “TOSHIBA TARO” in FIG. 8 does not overlap with the subject image, the overlapping range is “none”. For the patient comment “comment4”, the leading “com” included in the contour of the subject image and the subsequent “m” are detected as overlapping ranges.

  Note that, when the output image is a moving image, the superimposed region detection unit 23 performs the above-described processing for each frame forming the moving image.

(Superimposed area selection part)
The control unit 21 operates the superimposed region selection unit 24 only when the output image to be processed is a moving image. When the output image is a still image, the image data and the like pass through the superposition region selection unit 24 and are sent to the accompanying information erasing unit 17 on the downstream side.

  The superimposition area selection unit 24 corresponds to an example of the “superimposition area selection unit” of the present invention, and analyzes the superimposition area detected by the superimposition area detection unit 23 for each frame for each incidental information of the output image composed of a moving image. Then, the maximum overlapping region corresponding to each supplementary information is selected. For this purpose, for example, for each incidental information, the number of pixels in the overlapping region in each frame is obtained and compared, and the overlapping region having the largest number of pixels is selected.

[Operation]
The supplementary information erasing process executed by the medical image diagnostic apparatus of the present embodiment as described above will be described with reference to the flowchart of FIG. In the figure, the same steps as those in the first embodiment are denoted by the same reference numerals.

  First, as in the first embodiment, the image forming unit 10 forms a medical image of the subject (S1), embeds predetermined incidental information in the medical image to form an output image (S2), The output image is stored in the image storage unit 12 (S3). When the image reading unit 13 reads the output image (S4), it is determined whether the image is a still image or a moving image (S5).

(When the output image is a still image)
In the case of a still image, first, as in the first embodiment, each character embedded in the output image is extracted (S6). Thereby, each incidental information (character string information) in the output image is extracted. Each of the extracted supplementary information is sent to the overlapping area detection unit 23 together with the output image.

  The superimposition area detection unit 23 detects a superimposition area with the subject image for each extracted supplementary information (S41). The detected overlapping region (data of the pixel value) of each supplementary information is sent to the supplementary information erasing unit 17 together with the output image.

  The supplementary information erasure unit 17 selectively erases only the portion that is not included in the overlapping region for each supplementary information (S42). This erasing process is executed, for example, by changing the pixel value of the pixel outside the superimposition area in the range of the auxiliary information to the pixel value of the background image. As this erasing process, any method such as the method of the first embodiment or the second embodiment can be applied.

  As a result, for each supplementary information embedded in the output image, only the portion that does not overlap the subject image is deleted, and the portion that overlaps the subject image remains.

(When the output image is a movie image)
On the other hand, when the output image is a moving image, the output image is divided into a plurality of frames (S11), and each embedded character is extracted for each frame (S12). Thereby, incidental information (character string information) in each frame is extracted. The frames for which the extraction process has been completed are sequentially sent to the overlapping area detection unit 23 together with the extracted auxiliary information.

  The superimposed region detection unit 23 detects a superimposed region where the extracted supplementary information overlaps the subject image for each frame (S51). The frames for which this detection process has been completed are sequentially sent to the overlapping area selection unit 24 together with the detected overlapping area.

  Upon completion of the superimposition area detection process for all frames of the output image, the superimposition area selection unit 24 has the maximum range of superimposition areas in each frame (maximum superimposition area and Is selected) (S52). The selected maximum overlapping area of each supplementary information is sent to the supplementary information erasing unit 17 together with the output image.

  The supplementary information erasing unit 17 selectively erases the portion of each supplementary information in each frame except for the portion corresponding to the maximum overlapping region selected for the supplementary information (S53). This erasure process is executed by an arbitrary method such as the method of the first embodiment or the second embodiment, similarly to step S42.

  As a result of the above processing, all incidental information in each frame of the output image is deleted leaving a portion corresponding to each maximum overlapped area. In other words, the output image obtained by the above processing is an image in which the same portion (maximum superimposition region) of each supplementary information is left for all frames.

[Function and effect]
As described above, the medical image diagnostic apparatus according to the present embodiment selects only the portion excluding the overlapping region overlapping the subject image in the output image when deleting the supplementary information embedded in the output image. The portion overlapping the subject image is left as it is. Thereby, since the information on the subject image is not changed when the incidental information is erased, it is possible to reduce the possibility of misdiagnosis while erasing unnecessary incidental information.

  In addition, when the output image is a moving image, the same portion of all frames is deleted for each auxiliary information, so that the visibility of the image at the time of moving image reproduction is improved (Modification 3 of the first embodiment is used). reference).

  Furthermore, since each supplementary information is erased so as to leave a portion corresponding to the maximum superimposition area, the unnecessary supplementary information is erased to the maximum while ensuring the visibility at the time of video playback, and personal information is protected. It is possible to plan. Such a process is effectively applied when erasing incidental information embedded in a moving image in which the size of the subject image changes, such as a moving image representing the motion state of the heart.

[Modification]
As for the incidental information including symbol information that is not character string information as in the 5 cm scale in FIGS. 8 and 9, the incidental information is extracted by associating a predetermined pixel value when embedding the incidental information. It can be configured to detect the overlap region and erase the other part.

  In addition, it is desirable that all the incidental information related to personal information such as the patient name is deleted, including the overlapping area with the subject image. Therefore, as in the first embodiment, a character string indicating the type of information is embedded before (or after) the actual personal information, and the information corresponding to the character string is erased entirely. It can be configured as follows.

  Further, the erasure range may be changed according to the use of the output image. For example, when the subject image is used for diagnosis, such as when obtaining a second opinion, the supplementary information is deleted leaving only the superimposed area, while the output image is posted on a homepage, etc. When the subject image is not used for diagnosis, the entire incidental information is erased. Note that the input operation for the use of the output image is performed by, for example, providing a plurality of soft keys for selecting an erasing range in the dialog box and clicking a desired soft key.

<Fourth Embodiment>
In the present embodiment, a configuration is disclosed in which only desired supplementary information can be selectively deleted from the supplementary information embedded in the output image.

[Device configuration]
The medical image diagnostic apparatus of this embodiment has the same external configuration as that of the first embodiment (see FIG. 1). The control system is substantially the same as that of the first embodiment (see FIG. 2), but is characterized by the configuration of the auxiliary information embedding unit 11 for embedding the auxiliary information and the configuration for selecting the embedded auxiliary information. have.

  FIG. 12 is a block diagram illustrating an example of a control system of the present embodiment. In the figure, the same program storage unit 22 as in FIG. 2 is omitted.

(Attached information embedding part)
The supplementary information embedding unit 11 of this embodiment embeds identification information for identifying the type of the information as supplementary information together with actual information such as personal information of the patient.

  The identification information is the character string keywords “name”, “ID”, “kV”, and “mA” in the example of the first embodiment (see FIG. 4). Here, the information identified by the identification information “name” is the character string information “Toshiba Taro”, the information identified by the identification information “ID” is the character string information “1048”, and is identified by the identification information “kV”. The information identified is character string information “120”, and the information identified by the identification information “mA” is character string information “50”.

  Hereinafter, information identified by the identification information is referred to as “identification target information”. The identification information and the identification target information are embedded in association with each other using, for example, delimiter information such as “:” in Modification 2 of the first embodiment.

  In addition to the identification information that allows the person who sees them to know the meaning empirically, such as these character string keywords, it is specific to each identification target information such as barcode data such as QR code (registered trademark). Any associated image or symbol can be used as identification information. However, this identification information is preferably composed of data that can be extracted by the incidental information extraction unit 15 (OCR software).

(Data storage unit)
The data storage unit 25 corresponds to an example of the “storage unit” of the present invention that stores the erasure target specifying information 25a. The data storage unit 25 can be configured by an arbitrary storage device, but is preferably a storage device such as a hard disk drive that can rewrite data.

  The erasure target specifying information 25a is information for specifying the identification information of the identification target information to be erased among the identification target information embedded in the output image. The erasure target specifying information 25a includes identification information corresponding to various types of identification target information embedded in the output image, for example, in a list format. Then, identification information of identification target information to be erased is specified by a flag or the like.

  For example, the erasure target specifying information 25a corresponding to the example of FIG. 4 includes identification information “name”, “ID”, “kV”, and “mA” in a list format. When the subject name and the subject ID information are to be erased, the identification information “name” and “ID” are specified by flags.

(Erase Target Determination Unit)
The erasure target determination unit 26 corresponds to an example of the “erasure target determination unit” of the present invention, and for each identification information extracted by the supplementary information extraction unit 15, is the identification information specified by the erasure target specification information 25 a? to decide.

  More specifically, the erasure target determination unit 26, for each piece of extracted identification information, the character string forming the identification information matches one of the character strings of the identification information identified by the erasure target identification information 25a. Judge whether to do. If they match, it is determined that the extracted identification information is “specified by the erasure target specifying information 25a”. On the other hand, if they do not match, it is determined that the extracted identification information is “not specified by the erasure target specifying information 25a”.

(Attached information deletion part)
The incidental information erasure unit 17 of the present embodiment erases the identification target information identified by the identification information when the erasure target determination unit 26 determines that the extracted identification information is “specified”. To work. In other words, the incidental information erasure unit 17 is identified by the identification information determined as “identified by the erasure target identification information 25a” by the erasure target determination unit 26 among the identification target information embedded in the output image. Only the character string information is selectively deleted.

[Operation]
The supplementary information erasing process executed by the medical image diagnostic apparatus of the present embodiment as described above will be described with reference to the flowchart of FIG. In the figure, the same steps as those in the first embodiment are denoted by the same reference numerals.

  First, as in the first embodiment, the image forming unit 10 forms a medical image of the subject (S1). The auxiliary information embedding unit 11 embeds predetermined identification target information (character string information) in the medical image together with the identification information to form an output image (S60). The formed output image is stored in the image storage unit 12 (S3). When the image reading unit 13 reads the output image (S4), it is determined whether the image is a still image or a moving image (S5).

(When the output image is a still image)
In the case of a still image, first, as in the first embodiment, each character embedded in the output image is extracted (S6). Thereby, each incidental information (character string information) in the output image is extracted. The extracted incidental information includes, in particular, one or more pieces of identification target information and identification information for identifying each piece of identification target information. The extracted incidental information is sent to the erasure target determination unit 26 together with the output image.

  When barcode information such as QR code (registered trademark) is embedded as identification information, the barcode information is extracted and sent to the erasure target determination unit 26 together with the identification target information and the output image. It is done.

  The erasure target determination unit 26 determines, for each extracted identification information, whether the identification information is specified by the erasure target specifying information 25a (S61). The determination result is sent to the incidental information erasure unit 17 together with the output image.

  The supplementary information erasure unit 17 erases only the identification target information identified by the identification information determined to be “specified” in step S61 from the identification target information embedded in the output image. (S62). This erasing process is performed by an arbitrary method such as the method of the first embodiment or the second embodiment.

  As a result, only one or more identification target information embedded in the output image is selectively erased as identified by the identification information identified by the erasure target identification information 25a.

(When the output image is a movie image)
On the other hand, when the output image is a moving image, the output image is divided into a plurality of frames (S11), and each embedded character is extracted for each frame (S12). Thereby, incidental information (character string information) in each frame is extracted. The frames for which the extraction process has been completed are sequentially sent to the overlapping area detection unit 23 together with the extracted auxiliary information.

  The erasure target determination unit 26 determines whether the identification information is identified by the erasure target identification information 25a for each identification information extracted from each frame. (S71). This determination process is performed for each frame, and the frames for which the process has been completed are sequentially sent to the auxiliary information erasure unit 17 together with the determination result.

  The incidental information erasure unit 17 erases only the identification target information identified by the identification information determined to be “specified” in step S71 from the identification target information embedded in each frame. (S72). This erasure process is performed for each frame by an arbitrary method such as the method of the first embodiment or the second embodiment.

  Through the above processing, for each frame of the output image, only one identified by the identification information specified by the deletion target specifying information 25a among the one or more embedded identification target information is selectively deleted. It becomes.

[Function and effect]
As described above, the medical image diagnostic apparatus according to the present embodiment has the erasure target specifying information 25a for specifying whether or not each identification target information embedded in the output image is an erasure target in advance. . When the output image is read from the image storage unit 12, the identification target information embedded in the output image and the identification information are extracted, and the extracted identification information is erased by the erasure target specifying information 25a. Determine if the subject is specified. Then, only the erasure target information corresponding to the identification information specified as the erasure target is selectively erased from the output image.

  Thereby, by specifying the desired identification information in the erasure target specifying information 25a, it is possible to selectively delete only the desired identification target information. In particular, it is desirable to always specify information related to the personal information of the subject such as the name of the subject and the subject ID information by the deletion target specifying information 25a.

[Modification]
It is desirable to provide means for arbitrarily changing the identification information (erase target) specified by the erase target specifying information 25a. The following forms can be applied as the erasure target changing means.

  First, as a first form, a dialog box for designating an erasure target is displayed on the monitor device 5. In this dialog box, for example, identification information of each identification target information embedded in the output image is listed, and a check box for switching designation / cancellation is formed beside each identification information. The user operates an input device (for example, a mouse) to display a check mark in a check box of desired identification information, and designates the identification information as an erasure target. Each identification information specified in this way is set as an erasure target in the erasure target specifying information 25a by the control unit 21.

  As a second mode, a configuration in which the erasure target is automatically changed according to the use of the output image can be applied. For this purpose, identification information to be erased is set in advance for each use of the output image. When a dialog box for use designation is displayed on the monitor device 5 and the use is designated by the input device, the control unit 21 sets the identification information of the erase target corresponding to the designated use as the erase target specifying information 25a. Set.

<Medical image processing apparatus and medical image processing program>
In the above description, various embodiments of the medical image diagnostic apparatus according to the present invention have been described in detail. Hereinafter, a medical image processing apparatus and a medical image processing program according to the present invention will be described.

  The medical image processing apparatus of the present invention is the computer 4 (see FIG. 1; may include the monitor device 5 and the input device 6) of each of the above embodiments, or any computer used for processing medical images. In particular, it is configured to include an accompanying information extracting unit 15 and an accompanying information erasing unit 17 (see FIG. 2 and the like). Computers used for medical image processing include computers exclusively used by doctors and servers that collectively manage medical images from various medical image diagnostic apparatuses.

  More specifically, in the first embodiment, the medical image processing apparatus includes at least the supplementary information extraction unit 15, the contour detection unit 16, the supplementary information erasing unit 17, the control unit 21, and the program storage unit in FIG. 22 is comprised. Furthermore, the communication interface 18 and / or the image output interface 19, the auxiliary information embedding unit 11, the image storage unit 12, the image database 20, and the like may be included (the same applies hereinafter). The image discriminating unit 14 is provided when handling both still images and moving images (the same applies hereinafter).

  In addition, the medical image processing apparatus corresponding to the second embodiment includes at least the supplementary information extraction unit 15, the supplementary information erasing unit 17 (region dividing unit 17a to pixel value changing unit 17d) in FIG. 1 and FIG. The unit 21 and the program storage unit 22 are included.

  The medical image processing apparatus corresponding to the third embodiment includes at least the supplementary information extraction unit 15, the superimposition region detection unit 23, the superimposition region selection unit 24, the supplementary information erasure unit 17, and the control in FIGS. The unit 21 and the program storage unit 22 are included.

  The medical image processing apparatus corresponding to the fourth embodiment includes at least the supplementary information extraction unit 15, the data storage unit 25, the erasure target determination unit 26, the supplementary information erasure unit 17, and the control unit in FIGS. 21 and a program storage unit 22.

  According to such a medical image processing apparatus, the auxiliary information embedded in the medical image can be suitably automatically deleted as in the embodiments of the medical image diagnostic apparatus described above.

  The medical image processing program according to the present invention is a computer that causes an arbitrary computer to execute the processes described in the first to fourth embodiments (see the flowcharts in FIGS. 3, 7, 11, and 13). It is a program. The medical image processing program may be installed in the computer itself (stored in the program storage unit 22 or the like in FIG. 2), or the program is installed in a server on a network, and the computer is connected to the network. You may make it read and use via. Thereby, the computer functions as the accompanying information extracting unit 15 and the accompanying information erasing unit 17 in particular.

  The medical image processing program corresponding to the first embodiment causes the computer to function as the auxiliary information extracting unit 15, the contour detecting unit 16, the auxiliary information erasing unit 17, and the control unit 21, and executes the processing shown in FIG. Let Further, it may function as the auxiliary information embedding unit 11 and the image determination unit 14 (the same applies hereinafter).

  The medical image processing program corresponding to the second embodiment causes the computer to function as the incidental information extraction unit 15, the incidental information erasure unit 17 (region division unit 17a to pixel value change unit 17d), and the control unit 21. Then, the process shown in FIG. 7 is executed.

  In addition, the medical image processing program corresponding to the third embodiment causes a computer to function as the incidental information extraction unit 15, the overlapping region detection unit 23, the overlapping region selection unit 24, the incidental information deletion unit 17, and the control unit 21. Then, the processing shown in FIG. 11 is executed.

  The medical image processing program corresponding to the fourth embodiment causes the computer to function as the incidental information extraction unit 15, the data storage unit 25, the erasure target determination unit 26, the incidental information erasure unit 17, and the control unit 21, The process shown in FIG. 13 is executed.

  Such a medical image processing program includes a semiconductor memory such as a ROM, a RAM, a flash memory, a memory device such as an integrated circuit, an optical disk, a magneto-optical disk (CD-ROM / DVD-RAM / DVD-ROM / MO). Etc.) and a magnetic recording medium (hard disk / floppy (registered trademark) disk / ZIP, etc.).

  The configuration described in detail above merely discloses one specific example for suitably carrying out the present invention, and arbitrary modifications can be appropriately made within the scope of the gist of the present invention. is there.

It is the schematic showing an example of the external appearance structure of 1st Embodiment of the medical image diagnostic apparatus which concerns on this invention. It is a schematic block diagram showing an example of the structure of the control system of 1st Embodiment of the medical image diagnostic apparatus which concerns on this invention. It is a flowchart showing an example of operation | movement of 1st Embodiment of the medical image diagnostic apparatus which concerns on this invention. It is a schematic explanatory drawing showing an example of the incidental information deletion process by 1st Embodiment of the medical image diagnostic apparatus which concerns on this invention. FIG. 4A shows a schematic form of a medical image before embedding incidental information. FIG. 4B shows a schematic form of an output image formed by embedding incidental information. FIG. 4C shows a schematic form of an output image from which incidental information has been deleted. It is a schematic block diagram showing an example of the partial structure of the control system of 2nd Embodiment of the medical image diagnostic apparatus which concerns on this invention. It is a schematic explanatory drawing showing an example of the incidental information deletion process by 2nd Embodiment of the medical image diagnostic apparatus which concerns on this invention. FIG. 6A shows a schematic form of the output image before the auxiliary information is deleted. FIG. 6B shows a partially enlarged view of FIG. FIG. 6C shows a schematic form of the portion shown in FIG. 6B after the incidental information is erased. It is a flowchart showing an example of operation | movement of 2nd Embodiment of the medical image diagnostic apparatus which concerns on this invention. It is explanatory drawing for demonstrating the state with which the to-be-examined image and incidental information overlapped in 3rd Embodiment of the medical image diagnostic apparatus which concerns on this invention. It is explanatory drawing for demonstrating the state with which the to-be-examined image and incidental information overlapped in 3rd Embodiment of the medical image diagnostic apparatus which concerns on this invention. It is a schematic block diagram showing an example of the partial structure of the control system of 3rd Embodiment of the medical image diagnostic apparatus which concerns on this invention. It is a flowchart showing an example of operation | movement of 3rd Embodiment of the medical image diagnostic apparatus which concerns on this invention. It is a schematic block diagram showing an example of the partial structure of the control system of 4th Embodiment of the medical image diagnostic apparatus which concerns on this invention. It is a flowchart showing an example of operation | movement of 4th Embodiment of the medical image diagnostic apparatus which concerns on this invention.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Medical diagnostic imaging apparatus 2 Gantry 3 Sleeper 4 Computer (medical image processing apparatus)
DESCRIPTION OF SYMBOLS 5 Monitor apparatus 6 Input device 10 Image formation part 11 Additional information embedding part 12 Image storage part 13 Image reading part 14 Image discrimination | determination part 15 Additional information extraction part 16 Contour detection part 17 Additional information deletion part 17a Area division part 17b Acquisition of pixel value Unit 17c Pixel value selection unit 17d Pixel value change unit 18 Communication interface 19 Image output interface 20 Image database 21 Control unit 22 Program storage unit 23 Superimposition region detection unit 24 Superimposition region selection unit 25 Data storage unit 25a Erase target specific information 26 Erase target Determination unit 30 External database 40 Image output medium

Claims (26)

A medical image diagnostic apparatus having an image forming means for forming a medical image based on data reflecting an internal form of a subject,
Extracting means for extracting auxiliary information embedded in an output image formed by embedding auxiliary information in the medical image;
Erasing means for erasing the extracted incidental information from the output image;
Output means for outputting an output image from which the incidental information has been deleted;
A medical image diagnostic apparatus comprising: The incidental information includes character string information,
The extraction means analyzes the entire output image and extracts character string information embedded in the output image for each character.
The medical image diagnostic apparatus according to claim 1. A contour detecting means for detecting the contour of the extracted character;
The erasing means erases character string information by changing a pixel value of a pixel in an inner region of the detected contour to a predetermined value.
The medical image diagnostic apparatus according to claim 2. The erasing means includes
A dividing unit that divides the extracted character and its peripheral region into a plurality of partial regions;
Pixel value acquisition means for acquiring a pixel value of each pixel in the partial peripheral region corresponding to the peripheral region for each partial region including the region corresponding to the character among the plurality of partial regions;
Pixel value selection means for selecting a pixel value that occupies a maximum range in the partial peripheral region based on the acquired pixel value for each partial region;
With
Erasing the character by changing the pixel value of each pixel of each subregion to the selected pixel value;
The medical image diagnostic apparatus according to claim 2. The image processing apparatus further includes a superimposition region detection unit that analyzes the output image from which the supplementary information is extracted by the extraction unit and detects a superimposition region where the subject image representing the internal form of the subject and the extracted supplementary information overlap. ,
The erasing means selectively erases a portion outside the detected overlapping area from the extracted incidental information.
The medical image diagnostic apparatus according to claim 1. The formed output image is a moving image composed of a plurality of still images,
The extraction means extracts the auxiliary information embedded in the plurality of still images of the moving image for each still image,
The superimposing region detection means detects the superimposing region of the extracted incidental information for each still image,
A superposition area selection means for selecting a superposition area of the maximum range among the superposition areas detected for each still image;
The erasing means selectively erases a portion corresponding to the outside of the selected overlapped area of the incidental information of each of the plurality of still images;
The medical image diagnostic apparatus according to claim 5.   The superimposing area detecting means detects a range of the subject image based on a pixel value of each pixel of the output image, and based on the detected range of the subject image and the range of the extracted auxiliary information. The medical image diagnostic apparatus according to claim 5 or 6, wherein the superimposed region is detected. The formed output image is a moving image composed of a plurality of still images,
The extraction means extracts the auxiliary information embedded in the plurality of still images of the moving image for each still image,
The erasing means erases the extracted supplementary information from each of the plurality of still images for each still image;
The medical image diagnostic apparatus according to claim 1, wherein the medical image diagnostic apparatus is a medical image diagnostic apparatus. The formed output image is a moving image composed of a still image or a plurality of still images,
Further comprising image discriminating means for discriminating whether the output image is a still image or a moving image;
The extraction unit extracts, for each still image, supplementary information embedded in a plurality of still images forming the moving image when the output image is determined to be a moving image.
The erasing means erases the extracted supplementary information from each of the plurality of still images for each still image;
The medical image diagnostic apparatus according to claim 1, wherein the medical image diagnostic apparatus is a medical image diagnostic apparatus.   The medical image diagnostic apparatus according to claim 8 or 9, wherein the erasing unit erases the same area of the plurality of still images when the auxiliary information is erased for each still image. . The accompanying information includes one or more pieces of character string information and identification information for identifying the type of each character string information,
Storage means for preliminarily storing erasure target specifying information for specifying the identification information of character string information to be deleted among the one or more character string information;
Erasure target determination means for determining whether the identification information extracted by the extraction means is specified by the erasure target specification information;
Further comprising
The erasing means selectively erases the character string information identified by the identification information when it is determined that the extracted identification information is specified;
The medical image diagnostic apparatus according to claim 1, wherein the medical image diagnostic apparatus is a medical image diagnostic apparatus.   The medical image diagnosis apparatus according to claim 11, wherein the identification information includes a character string representing a type of character string information to be identified or an image associated with the character string information to be identified. A medical image processing apparatus for processing a medical image formed based on data reflecting an internal form of a subject,
Extracting means for extracting auxiliary information embedded in an output image formed by embedding auxiliary information in the medical image;
Erasing means for erasing the extracted incidental information from the output image;
Output means for outputting an output image from which the incidental information has been deleted;
A medical image processing apparatus comprising: The incidental information includes character string information,
The extraction means analyzes the entire output image and extracts character string information embedded in the output image for each character.
The medical image processing apparatus according to claim 13. A contour detecting means for detecting the contour of the extracted character;
The erasing means erases character string information by changing a pixel value of a pixel in an inner region of the detected contour to a predetermined value.
The medical image processing apparatus according to claim 14. The erasing means includes
A dividing unit that divides the extracted character and its peripheral region into a plurality of partial regions;
Pixel value acquisition means for acquiring a pixel value of each pixel in the partial peripheral region corresponding to the peripheral region for each partial region including the region corresponding to the character among the plurality of partial regions;
Pixel value selection means for selecting a pixel value that occupies a maximum range in the partial peripheral region based on the acquired pixel value for each partial region;
With
Erasing the character by changing the pixel value of each pixel of each subregion to the selected pixel value;
The medical image processing apparatus according to claim 14. The image processing apparatus further includes a superimposition region detection unit that analyzes the output image from which the supplementary information is extracted by the extraction unit, and detects a superimposition region where the subject image representing the internal form of the subject overlaps the extracted supplementary information. ,
The erasing means selectively erases a portion outside the detected overlapping area from the extracted incidental information.
The medical image processing apparatus according to claim 13. The formed output image is a moving image composed of a plurality of still images,
The extraction means extracts the auxiliary information embedded in the plurality of still images of the moving image for each still image,
The erasing means erases the extracted supplementary information from each of the plurality of still images for each still image;
The medical image processing apparatus according to claim 13, wherein the medical image processing apparatus is a medical image processing apparatus. The accompanying information includes one or more pieces of character string information and identification information for identifying the type of each character string information,
Storage means for preliminarily storing erasure target specifying information for specifying the identification information of character string information to be deleted among the one or more character string information;
Erasure target determination means for determining whether the identification information extracted by the extraction means is specified by the erasure target specification information;
Further comprising
The erasing means selectively erases the character string information identified by the identification information when it is determined that the extracted identification information is specified;
The medical image processing apparatus according to any one of claims 13 to 18, wherein the medical image processing apparatus according to any one of claims 13 to 18 is characterized. Computer
Receiving means for receiving an output image formed by embedding additional information in a medical image formed based on data reflecting the internal form of the subject;
Extraction means for extracting the supplementary information embedded in the received output image;
Erasing means for erasing the extracted incidental information from the output image;
A medical image processing program characterized by being made to function. The incidental information includes character string information,
The extraction means functions to analyze the entire received output image and extract character string information embedded in the output image for each character.
21. The medical image processing program according to claim 20, wherein: Allowing the computer to further function as contour detection means for detecting the contour of the extracted character;
The erasing means functions to erase character string information by changing a pixel value of a pixel in the inner region of the detected contour to a predetermined value.
The medical image processing program according to claim 21, wherein: The erasing means,
A dividing unit that divides the extracted character and its peripheral region into a plurality of partial regions;
Pixel value acquisition means for acquiring a pixel value of each pixel in the partial peripheral region corresponding to the peripheral region for each partial region including the region corresponding to the character among the plurality of partial regions;
Pixel value selection means for selecting a pixel value that occupies a maximum range in the partial peripheral region based on the acquired pixel value for each partial region;
As well as function
Functioning to erase the character by changing the pixel value of each pixel of each of the partial areas to the selected pixel value;
The medical image processing program according to claim 21, wherein: The computer analyzes the output image from which the supplementary information is extracted by the extraction unit, and detects a superimposition region detection in which a subject image representing an internal form of the subject and the extracted supplementary information overlap each other To further function as a means,
Causing the erasing means to function to selectively erase a portion outside the detected overlapped area of the extracted incidental information;
21. The medical image processing program according to claim 20, wherein: The formed output image is a moving image composed of a plurality of still images,
The extraction means functions to extract the auxiliary information embedded in the plurality of still images of the moving image for each still image;
Causing the erasing means to function to erase the extracted incidental information from each of the plurality of still images for each still image;
25. The medical image processing program according to any one of claims 20 to 24, wherein: The accompanying information includes one or more pieces of character string information and identification information for identifying the type of each character string information,
The computer comprises storage means for preliminarily storing erasure target specifying information for specifying the identification information of the character string information to be erased among the one or more character string information,
Causing the computer to further function as an erasure target determination means for determining whether the identification information extracted by the extraction means is specified by the erasure target specifying information;
Causing the erasing means to function to selectively erase character string information identified by the identification information when it is determined that the extracted identification information is specified;
The medical image processing program according to any one of claims 20 to 25, wherein: