KR20150107515A - medical image processor and method thereof for medical diagnosis - Google Patents

Abstract

Disclosed are a medical image processor for medical diagnosis and a method thereof. According to an embodiment of the present invention, the medical image processor for medical diagnosis comprises: a front-end processor which acquires channel data from ultrasonic signals reflected from an object; an image generating part which generates ultrasonic signals by manipulating image parameters of channel data acquired through the front-end processor; an image analyzing part which detects lesions by analyzing ultrasonic signals generated through the image generating part; and a display part which provides the results of analysis of the image analyzing part.

Description

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a medical image processing apparatus and a method thereof,

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a medical image processing technique, and more particularly, to a technique for generating and providing various medical images for medical diagnosis.

When performing medical diagnosis and treatment using medical images at a medical site, it is necessary to quantitatively measure the size, range, and function of a lesion. Since the medical image may give different meanings depending on the user's eyes, the user may not recognize the lesion existing in the image by mistake. In addition, when a lesion is detected using a medical image, the lesion may not be recognized from a large-capacity medical image according to the skill or fatigue of the user.

Computer aided diagnosis (CAD) technology is a technology that analyzes the medical images by intelligent high-performance image analysis system computer and provides lesion status. However, even if a large amount of medical image data exists in the computer-assisted diagnosis technology, it can not be efficiently used by a large amount of data in a medical field. Accordingly, there is a need for a medical image processing technology capable of automatically generating and providing a medical image that can substantially assist a user.

According to one embodiment, a medical image processing apparatus and a method thereof that substantially assist medical diagnosis are proposed.

A medical image processing apparatus according to an embodiment of the present invention includes a front end processor for acquiring channel data from an ultrasound signal reflected from a target object, an ultrasound image generating unit for generating ultrasound images by manipulating image parameters of the channel data acquired through the front end processor, And an image analyzing unit for analyzing the ultrasound image generated through the image generating unit to detect a lesion.

The channel data according to an exemplary embodiment is raw data that does not include signal processing including beamforming in the image generating unit. At this time, the channel data may be a digital RF signal obtained by digitizing an analog RF electric signal converted from an ultrasonic wave reflection signal through a transducer. Alternatively, the channel data may be IQ data demodulated into a baseband using an analog RF electric signal converted from an ultrasonic wave reflected signal through a transducer or a digital RF signal obtained by digitizing an analog RF electric signal.

The front-end processor according to an exemplary embodiment includes an analog-to-digital converter for converting an analog RF electric signal converted from an ultrasonic wave reflected signal through a transducer into a digital RF electric signal. Further, the front-end processor may further include a front-end processor for generating IQ data by demodulating the analog RF electric signal or the digital RF signal to the baseband by being located at the front end or the rear end of the analog-digital converter.

The image generation unit performs software processing of channel data signal processing and ultrasound image generation through manipulation of image parameters through a computer.

The front-end processor according to an exemplary embodiment stores the obtained channel data in the memory of the host computer. The image generator reads the channel data stored in the memory of the host computer, manipulates the image parameters of the read channel data, Thereby generating a new ultrasound image.

The image generating unit according to an embodiment generates an ultrasound image from channel data according to an image parameter set by a user. The image generating unit may generate the ultrasound image from the channel data according to the image parameter corresponding to the lesion-detected ultrasound image through the image analyzing unit.

The image analysis unit according to an exemplary embodiment is a computer aided diagnosis (CAD) system that receives ultrasonic images generated through an image generation unit via a software interface or through a network and analyzes received ultrasound images.

The image analysis unit may include a preprocessor for correcting the ultrasound image generated through the image generator, an image feature extraction unit for extracting image features from the ultrasound image corrected through the preprocessor, And determining whether the lesion is present in the ultrasound image.

The image analyzing unit analyzes an ultrasound image by reflecting additional information received from a user with respect to the ultrasound image. The judging unit according to an embodiment judges whether a lesion is present using a fuzzy neural network.

The image analysis unit may further include a post-processing unit for providing a determination result through the determination unit and providing feedback information received from the user who has confirmed the determination result to the determination unit for later determination. The post-processing unit according to an exemplary embodiment provides an ultrasound image having a boundary outline in a lesion area to emphasize a lesion area detected when a lesion is detected.

The medical image processing apparatus according to an embodiment further includes a controller for requesting an image analysis unit to analyze an image, receiving an image analysis result, and generating an alarm signal according to the received analysis result and providing the generated alarm signal through a display unit.

The control unit according to an embodiment identifies the image and the manipulated image parameter through the identifier of the image detected by the image analyzing unit, and provides the alarm signal together with the detected image through the display unit.

The controller may provide an alarm signal through a display unit when a lesion is detected by the image analyzer during the provision of the ultrasound image through the display unit, and when the user's alarm check command for the provided alarm signal is input, Mode, the user acquires the ultrasound image including the ultrasound image and the lesion detected, and provides the ultrasound image through the display unit in non-real time so that the lesion can be precisely judged through comparison between the ultrasound images. And provides a multi-ultrasound image acquired through signal processing and re-scan while changing parameters. In addition, the controller may switch to a real-time mode while providing a multi-ultrasound image in a non-real-time mode in a non-real-time mode, and may provide a multi-ultrasound image including an ultrasound image and a lesion detected through the display unit in real time.

The medical image processing apparatus according to an embodiment further includes a first memory for storing preset image parameters and usage information for each image parameter and providing the stored image parameters to the image generator.

The medical image processing apparatus according to an exemplary embodiment of the present invention includes a channel data acquisition unit that acquires channel data obtained through a front end processor and an ultrasound image generated from the acquired channel data, To the image generating unit.

The display unit according to an exemplary embodiment may provide the ultrasound image analyzed through the image analysis unit as a real-time image, a still image, a review image, or a combination thereof in the form of a multi-ultrasound image.

According to another aspect of the present invention, there is provided a medical image processing method including the steps of acquiring channel data from an ultrasound signal reflected from a target object, generating an ultrasound image by manipulating image parameters of the obtained channel data, To detect a lesion and to provide a detection result.

According to an exemplary embodiment, a meaningful ultrasound image is automatically generated using channel data and various image parameters obtained from an ultrasound reflection signal, and the ultrasound image is used for medical diagnosis, thereby improving the accuracy of medical diagnosis. In particular, it can be used in a computer aided diagnosis (CAD) system to provide medical images with high accuracy of lesion detection.

Further, instead of using the beamformed data for generating the ultrasound image, a new ultrasound image is generated based on the channel data, which is original data before the beamforming process, so that the image acquisition is easy, Various ultrasound images can be obtained.

In addition, it is possible to generate various ultrasound images by applying various image parameters to the channel data, so that it is possible to easily detect the presence of abnormal lesions and the location of lesions from various ultrasound images, thereby improving the accuracy of medical diagnosis. In particular, the analyzed ultrasound images can be provided in real time.

Furthermore, by processing the process for generating an ultrasound image by software rather than hardware such as an FPGA, various image generation functions can be provided without limitation of the hardware, and various ultrasound images with high accuracy can be generated.

1 is a configuration diagram of a medical image processing apparatus according to an embodiment of the present invention;
2 is a detailed configuration diagram of the front-end processor of FIG. 1 according to an embodiment of the present invention;
FIG. 3 is a detailed configuration diagram of the image generating unit of FIG. 1 according to an embodiment of the present invention,
FIG. 4 is a detailed configuration diagram of the image analysis unit of FIG. 1 according to an embodiment of the present invention;
5 is a configuration diagram of a medical image processing apparatus according to another embodiment of the present invention.
FIG. 6 is a view for explaining the principle of a fuzzy neural network used for lesion judgment according to an embodiment of the present invention;
7 is a block diagram illustrating a multi-image providing screen and user input means for explaining a multi-live technique according to an exemplary embodiment of the present invention;
8 is a flowchart illustrating a medical image processing method for medical diagnosis according to an embodiment of the present invention.

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings. In the following description of the present invention, a detailed description of known functions and configurations incorporated herein will be omitted when it may make the subject matter of the present invention rather unclear. In addition, the terms described below are defined in consideration of the functions of the present invention, which may vary depending on the intention of the user, the operator, or the custom. Therefore, the definition should be based on the contents throughout this specification.

1 is a configuration diagram of a medical image processing apparatus 1a according to an embodiment of the present invention.

1, a medical image processing apparatus 1a includes a front end processor 10, an image generating unit 11, an image analyzing unit 12, and a display unit 13, 17).

A back end processor including the image generating unit 11, the image analyzing unit 12 and the control unit 17 may be located in a host PC. The front end processor 10 may be located in the host computer, but may be located outside the host computer and connected to the host computer via a data bus or the like.

Only the components related to the present embodiment are shown in the medical image processing apparatus 1a shown in Fig. Therefore, it will be understood by those skilled in the art that other general-purpose components other than the components shown in FIG. 1 may be further included.

The front-end processor 10 obtains channel data from an ultrasonic reflection signal reflected from a target object to be diagnosed. To this end, when a transducer transmits an ultrasonic signal to the interior of the object, receives the ultrasonic signal reflected from the object, and converts the ultrasonic reflection signal into an electric signal, the front-end processor 10 transmits the converted electric signal to the trans- It can receive from the ducer. The channel data is data that has undergone minimal signal processing based on raw data or original data that has not been subjected to signal processing including beamforming in the image generating unit 11. For example, the channel data is a digital RF signal obtained by digitizing an analog RF signal converted from an ultrasonic wave reflected signal through a transducer. Or channel data is I data and Q data which are demodulated into a baseband for an analog RF electric signal converted from an ultrasonic wave reflection signal through a transducer or a digital RF signal obtained by digitizing an analog RF electric signal. The detailed configuration of the front end processor 10 according to one embodiment will be described later with reference to FIG.

The image generating unit 11 processes and scans the ultrasound reflection signal received by the transducer to generate an ultrasound image. Then, the image generating unit 11 applies image parameters to the channel data obtained from the ultrasound reflection signal through the front-end processor 10, performs signal processing and re-scans the image data, and automatically generates a new ultrasound image . At this time, the image generating unit 11 may generate various ultrasound images by applying various image parameters to the channel data. The presence of abnormal lesions and the location of the lesion can be easily found from various ultrasound images, which can improve the accuracy of medical diagnosis.

The image parameter is a parameter capable of changing the resolution of the ultrasound image. For example, the image parameters may be frequency, gain, low frequency filtering value (LPF), extraction rate (DR), beamforming related parameters, and the like. The image parameter may be a combination of two or more images. The image parameters are set to predefined default values and can be changed by the user. Image parameters affect the resolution of the ultrasound image, for example the gain can affect the brightness of the ultrasound image. Various ultrasonic images can be obtained by operating various image parameters and applying each to the channel signal. The obtained various ultrasound images enable accurate medical diagnosis.

A new ultrasound image generating process using the image parameters and the channel data of the image generating unit 11 will be described. The channel data matching the preset image parameters are extracted from the channel data obtained through the front-end processor 10, And performs signal processing and scan conversion on the channel data to generate a new image. For example, only the harmonic frequency signal of interest is extracted from the channel data, and the extracted harmonic frequency signal is subjected to signal processing and scan conversion to generate a new image. The detailed configuration of the image generating unit 11 will be described later with reference to Fig.

The image analyzer 12 analyzes the ultrasound image generated through the image generator 11 to detect lesions. Then, the analysis result is provided through the display unit 13. The image analysis unit 12 may be configured to directly call a method from the image generation unit 11 through a software interface. Alternatively, the image analyzing unit 12 may be in the form of a server receiving the ultrasound image through the network from the image generating unit 11.

The image analyzing unit 12 analyzes an ultrasound image using a computer aided diagnosis (CAD) technique and provides analysis results. Computer-assisted diagnostic technology means analyzing ultrasound images by intelligent high-performance image analysis system computer to judge the presence or absence of lesion, change of lesion or the like, and provide judgment result. Such a computer assisted diagnostic device can automatically detect the presence of a lesion in the ultrasound image and can locate and display the location of the lesion. Furthermore, by automatically segmenting the lesion and calculating its size, it is possible to overcome the limit of accuracy by manual measurement.

Examples of medical diagnosis using computer assisted diagnostic technology are as follows. The image generating unit 11 adjusts the image parameters from the channel data by the user and generates an ultrasound image through re-scan, and provides the ultrasound image to the image analyzing unit 12 through the network. Then, the image analyzing unit 12 receives the ultrasound image from the image generating unit 11, determines whether the lesion is present in the ultrasound image by using the computer-assisted diagnostic technology, and outputs the determination result to the user through the display unit 13 . The detailed configuration of the image analysis unit 12 according to one embodiment will be described later with reference to FIG.

The control unit 17 requests the image analysis unit 12 to analyze the image and receives the analysis result from the image analysis unit 12. Then, an alarm signal is generated according to the analysis result, and an alarm signal generated by the alarm signal is provided through the display unit 13. The image analyzing unit 12 according to an embodiment provides an image of a lesion detected in the controller 17 and an identifier of the ultrasound image displayed on the screen through the display unit 13 when the lesion is detected. The control unit 17 identifies the image and the manipulated image parameter through the identifier of the image in which the lesion is detected by the image analysis unit 12. [ In addition, an alarm signal can be provided through the display unit 13 together with the image in which the lesion is detected. At this time, if a user's alarm confirmation command for the alarm signal is inputted, the controller 17 may acquire the multi-ultrasound image for the determination of the lesion through comparison between the ultrasound images of the user and provide the acquired ultrasound image through the display unit 13.

2 is a detailed configuration diagram of the front end processor 10 of FIG. 1 according to an embodiment of the present invention.

1 and 2, the front-end processor 10 includes an analog-to-digital converter (ADC) 102, and may further include a front-end processor 104.

The ADC 102 receives the analog RF signal from the transducer and samples it at a predetermined rate to convert it into a digital RF signal. At this time, the ADC 102 is provided for the number of conversion elements of the transducer so that each conversion element can correspond to the ADC 102 one to one. The converted digital RF signal can be stored in the memory of the host computer. For example, it may be located on the mainboard or graphics card of the host computer, but the memory location is not limited thereto.

The front-end processing unit 104 may be located at the front end or the rear end of the ADC 102. [ When it is located at the front end of the ADC 102, it demodulates the analog RF signal into baseband, generates IQ data, and transmits it to the ADC 102. On the other hand, when it is positioned at the rear end of the ADC 102, the digital RF signal converted through the ADC 102 is demodulated into baseband to generate IQ data. The demodulated IQ data may be stored in the memory of the host computer. For example, it may be stored on the main board or graphics card of the host computer, but the memory location is not limited thereto. In FIG. 2, the front-end processing unit 104 is located at the rear end of the ADC 102, but may be located at the front end of the ADC 102 as described above.

The present invention generates a new ultrasound image by manipulating image parameters on channel data, which is original data before beamforming processing, instead of using a beamformed signal to generate various ultrasound images. The channel data may be RF data or IQ data demodulating RF data. When manipulating image parameters from channel data, it is easy to acquire new ultrasound images and various ultrasound images can be obtained according to the manipulated image parameters.

FIG. 3 is a detailed configuration diagram of the image generating unit 11 of FIG. 1 according to an embodiment of the present invention.

1 and 3, the image generating unit 11 automatically generates various ultrasound images using channel data and various image parameters. The image generating unit 11 includes a beamformer 110, a digital signal processor (DSP) 112, and a digital scan converter (DSC)

The beam former 110 receives and stores channel data output from the front end processor 10. At this time, the beam former 110 may generate a reception focusing signal by adding an appropriate delay to the channel data in consideration of the time required to reach each conversion element of the transducer from the object. The beam former 110 may receive parameters related to beamforming.

The signal processing unit 112 processes the receive focusing signal from the beam former 110 to generate image data. The signal processing unit 112 according to an embodiment converts the reception focusing signal into a baseband signal and generates image data. The scan conversion unit 114 performs scan conversion to display image data input from the signal processing unit 112. At this time, when the signal processing unit 112 receives the channel data via the beam former 110 and obtains the preset image parameters, the signal processing unit 112 and the scan conversion unit 114 convert the image parameters So that a new image can be generated by performing signal processing and scan conversion.

The image generating unit 11 according to an exemplary embodiment of the present invention may perform a series of processes for generating ultrasound images such as the beamformer 110, the signal processing unit 112, and the scan conversion unit 114, . At this time, the image generating unit 11 performs a process according to the algorithm programmed through the host computer. At this time, high-speed operation processing can be performed through the GPU of the host computer.

Generally, a process for generating an ultrasound image is performed through hardware such as an FPGA. However, in the case of hardware, there are restrictions on resources such as internal memory, and the development and production cost are high, and it is difficult to upgrade the system. In addition, the physical size of the diagnostic system increases due to the hardware volume. Accordingly, the present invention can perform a software-based signal processing process to provide various image generation functions without being limited by the hardware volume, and at the same time to generate various ultrasound images with high accuracy.

4 is a detailed configuration diagram of the image analysis unit 12 of FIG. 1 according to an embodiment of the present invention.

1 and 4, the image analysis unit 12 includes a pre-processor 120, an image feature extraction unit 122, and a determination unit 124. The post- (126).

The image analyzing unit 12 analyzes the ultrasound image according to the additional information received from the user. The additional information is background information for facilitating the functions of the pre-processing unit 120, the image feature extraction unit 122, and the determination unit 124 of the image analysis unit 12. For example, a body part, a kind of transducer, and the like. The application means a medical field such as an obstetrician or an orthopedic surgeon, and a body part means a diagnostic region of a subject such as a joint or an abdomen.

The pre-processing unit 120, the image feature extraction unit 122, and the determination unit 124 may use the additional information, respectively. For example, an object to be processed by the pre-processing unit 120, the image feature extraction unit 122, and the determination unit 124 may vary depending on additional information. The image analysis unit 12 may control the appropriate pre-processing unit 120, the image feature extraction unit 122, and the determination unit 124 integrally according to the additional information.

Hereinafter, the detailed components of the image analysis unit 12 will be described in detail. The pre-processing unit 120 and the image feature extracting unit 122 are used to inform the state of the ultrasound image or to improve the state of the ultrasound image to facilitate the determination of the lesion of the determination unit 124. The pre-processing unit 120 receives the ultrasound image generated through the image generating unit 11 from the control unit 17 and corrects the received ultrasound image. For example, the subcutaneous fat layer is removed from the ultrasound image, the brightness, contrast, etc. are adjusted and the average pixel is removed. The image feature extraction unit 122 extracts the image feature from the corrected ultrasound image through the preprocessing unit 120. The image feature information may be a brightness distribution, an area, a boundary separability, etc. for each band. The image feature extraction unit 122 performs clustering on the corrected ultrasound image through the preprocessing unit 120 and extracts image features for each cluster.

The determination unit 124 determines the presence or absence of a lesion, a change in a position, or a lesion in the ultrasound image using the image feature information extracted through the image feature extraction unit 122. [ The control unit 17 transmits the determination result to the control unit 17, and the control unit 17 displays the result of the determination through the display unit 13. The determination unit 124 according to an exemplary embodiment synthesizes the lesion detection results for each cluster for various images and determines that a lesion has occurred when a predetermined reference value or more is true. For example, if the lesion is judged True in 70% or more of images in the whole image, it is judged that the lesion has occurred. However, the reference value is not limited thereto. The determination unit 124 determines whether a lesion is present using the fuzzy neural network. The fuzzy neural network will be described later with reference to FIG.

The post-processing unit 126 transmits the determination result through the determination unit 124 to the control unit 17, and the control unit 17 receives the result and displays the result through the display unit 13. The post-processing unit 126 according to an exemplary embodiment provides an ultrasound image having a boundary outline in a lesion area to emphasize the lesion area detected when a lesion is detected. The ultrasound image provided at this time may be a live image provided in real time or a Cine image which is a non-real-time image in a state where signal inflow is stopped.

The control unit 17 according to an embodiment provides an alarm signal and corresponding ultrasound image through the display unit 13 when a lesion is detected through the image analysis unit 12. [ The post-processing unit 126 supplies the feedback information received from the user who has confirmed the determination result to the determination unit 124 under the control of the control unit 17 so that the post-processing unit 126 can use it for future determination of the lesion. For example, if the determination result of the determination unit 124 includes information suspicious of the mass, if the user processes the alarm of the controller 17 as True, the corresponding image and the additional information are determined as True Alarm Learning Data Lt; / RTI > On the other hand, if the user processes the alarm of the control unit 17 as False, the corresponding image and the additional information are fed back to the determination unit 124 as false alarm learning data.

The control unit 17 according to the embodiment provides an alarm signal through the display unit 13 when a lesion is detected by the image analyzing unit 12 during the provision of the ultrasound image through the display unit. When the user's alarm confirmation command for the provided alarm signal is inputted, the user acquires the ultrasound image including the ultrasound image and the lesion-detected image which are provided for the user to precisely judge the lesion through comparison between the ultrasound images in the non- And provides it in non-real time through the display unit 13. It is possible to provide multiple ultrasound images obtained through signal processing and re-scan while changing image parameters in the image generating unit 11 when necessary. In addition, the control unit 17 switches the real-time mode while providing the multi-ultrasound image in the non-real-time mode in the non-real-time mode, and displays the ultrasound image including the provided ultrasound image and the lesion- . An embodiment of the above-mentioned multi-ultrasound image providing will be described later in detail with reference to FIG.

FIG. 5 is a block diagram of a medical image processing apparatus 1b according to another embodiment of the present invention, which further includes detailed components in the medical image processing apparatus 1a of FIG. 1 in detail.

5, the medical image processing apparatus 1b includes a front end processor 10, an image generating unit 11, an image analyzing unit 12, a display unit 13, an input unit 14, a first memory 15 ), A second memory 16, and a control unit 17.

The front-end processor 10 acquires channel data from the ultrasound signals reflected from the object, and the image generating unit 11 manipulates the image parameters of the channel data acquired through the front-end processor 10 to generate an ultrasound image . The front end processor 10 may store the obtained channel data in the second memory 16. [ The second memory 16 provides channel data for the ultrasound image of the image generating unit 11 to the beam former 110 of the image generating unit 11. The second memory 16 may be located on the main board or graphics card of the host computer, but the position of the second memory 16 is not limited thereto.

The image generation unit 11 includes a beamformer 110, a signal processing unit 112, and a scan conversion unit 114. The beam former 110 receives and focuses channel data output from the front end processor 10 and the signal processing unit 112 receives and processes a reception focusing signal to generate image data. The image data input from the signal processing unit 112 is scanned and converted for display through the display unit 13. [ At this time, when the signal processing unit 112 receives the channel data from the second memory 16 via the beam former 110 and receives the image parameter from the first memory 15, the signal processing unit 112 and the scan conversion unit 114 may generate a new image by performing signal processing and scan conversion on the channel data by reflecting preset image parameters. The display unit 13 receives and displays the scan-converted image data through the scan conversion unit 114. Then, the analysis result of the image analysis unit 12 is displayed.

The image analyzer 12 analyzes the ultrasound image generated through the image generator 11 to detect a lesion and provides a detection result to the controller 17. [ At this time, the control unit 17 generates an alarm signal indicating the occurrence of the lesion together with the ultrasound image in which the lesion is detected, and provides the generated alarm signal to the display unit 13. The display unit 13 can provide a multi-ultrasound image including an ultrasound image and a lesion-detected image that are provided to the user in real time. This function is called a multi live function, and the multi live function will be described later in detail with reference to FIG.

An input unit 14 according to an embodiment receives an alarm confirmation command for an alarm signal from a user. The input unit 14 may receive feedback information from a user who has confirmed the detection result, and may provide the feedback information to the image analysis unit 12 via the control unit 17. In this case, the image analysis unit 12 may reflect the inputted feedback information to the lesion judgment.

The input unit 14 according to an exemplary embodiment receives preset image parameters from a user and stores the input image parameters in the first memory 15. Image parameters are stored in the first memory 15 and the stored image parameters are provided to the beam former 110 and the signal processing unit 112 of the image generating unit 11. In the first memory 15, usage information for each image parameter may be stored. Usage information for each image parameter relates to various usage details such as the frequency of use of the user, the use time, and the like. In this case, the image analyzing unit 12 may reflect the usage information for each image parameter in the ultrasound image analysis. In the second memory 16, ultrasound images detected by lesion detection through the analysis of the image analyzing unit 12 are stored in the second memory 16 on the channel data obtained through the front-end processor 10 and the ultrasound images generated from the obtained channel data do. The first memory 15 and the second memory 16 may be located in the host computer only in different regions thereof.

6 is a reference diagram for explaining the principle of a fuzzy neural network used for lesion judgment according to an embodiment of the present invention.

Referring to FIG. 6, a fuzzy neural network includes a first layer 610, a second layer 620, and a third layer 630. The lesion determination process using the fuzzy neural net will be described. The first layer 610 includes n input nodes into which input patterns having n characteristics are inputted according to image feature parameters. The second layer 620 is composed of a plurality of layers that are hidden layers. The third layer 630 is configured as an output node for outputting a lesion judgment result. The lesion judgment result of the third layer 630 may be provided as True or False. Whether or not the trunk connection between the nodes can be determined according to a predetermined reasoning rule.

Fuzzy neural networks are robust to noise because they can be processed quickly and with high accuracy through non-deterministic approaches and learning. In addition, medical knowledge used in actual image diagnosis can be conceptually applied, and verbal analysis of the final generated neural network is possible.

FIG. 7 is a reference diagram illustrating a multi-image providing screen and user input means for explaining a multi-live technique according to an exemplary embodiment of the present invention.

Referring to FIG. 7, the medical image processing apparatus provides an ultrasound image 700 in a single image form in a live or non-real-time state, Is activated (ON), the lesion is automatically detected and an alarm signal is provided to the user (720). The user who has confirmed the alarm signal requests the alarm confirmation to check whether the lesion judgment result of the medical image processing apparatus is correct, and the medical image processing apparatus receives the alarm confirmation command (720).

Then, the medical image processing apparatus enters the Cine mode, which is a non-real-time mode, and the ultrasound image 730, which is being provided, and the image 740 in which the lesion is detected, so that the user can precisely judge the lesion through comparison between the ultrasound images, Provides ultrasound images in non-real time. Ultrasound images obtained through signal processing and re-scanning can be provided while changing the image parameters 750 when necessary. At this time, the screen providing the multi-ultrasound image may be modified in various forms such as a dual mode and a quad mode.

Then, while the multi-ultrasound image is being provided in the non-real-time mode in the non-real-time mode, the medical image processing apparatus switches to the live mode 760 and includes the ultrasound image 770 and the detected image 780 It is possible to provide multiple ultrasound images in real time.

8 is a flowchart illustrating a medical image processing method for medical diagnosis according to an embodiment of the present invention.

Referring to FIG. 8, the medical image processing apparatus obtains channel data from ultrasound signals reflected from a target object (800). At this time, the obtained channel data can be stored in the memory.

Next, the medical image processing apparatus generates an ultrasound image by manipulating the image parameters of the obtained channel data (810). In operation 810 of generating an ultrasound image according to an exemplary embodiment, the medical image processing apparatus reads channel data stored in a memory and generates new ultrasound images by manipulating image parameters of the read channel data.

In operation 810 of generating an ultrasound image according to an exemplary embodiment, the medical image processing apparatus performs software processing of channel data signal processing and ultrasound image generation through manipulation of image parameters through a computer.

Next, the medical image processing apparatus analyzes the generated ultrasound image to detect a lesion and provides a detection result (820). In a step 820 of providing a detection result according to an embodiment, the medical image processing apparatus provides an alarm signal when a lesion is detected while providing an ultrasound image. When a user's alarm confirmation command for the provided alarm signal is input, the user acquires a multi-ultrasound image including an ultrasound image and a lesion-detected image, which the user is providing to accurately determine the lesion through comparison between the ultrasound images And provides multiple ultrasound images while changing image parameters as needed. In addition, in the non-real-time mode, the multi-ultrasound image can be provided in real time while providing the multi-ultrasound image in non-real-time mode, and the ultrasound image and the ultrasound image including the detected lesion can be provided in real time.

The embodiments of the present invention have been described above. It will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the invention as defined by the appended claims. Therefore, the disclosed embodiments should be considered in an illustrative rather than a restrictive sense. The scope of the present invention is defined by the appended claims rather than by the foregoing description, and all differences within the scope of equivalents thereof should be construed as being included in the present invention.

1a, 1b: medical image processing apparatus 10: front-end processor
11: Image generation unit 12: Image analysis unit
13: display section 14: input section
15: first memory 16: second memory
17: control unit 102: ADC
104: front end processing unit 110: beam former
112: signal processing unit 114: scan conversion unit
120: Preprocessing unit 122: Image feature extraction unit
124: Determination unit 126: Post processing unit

Claims (29)

A front end processor for acquiring channel data from the ultrasound signal reflected from the object;
An image generator for generating an ultrasound image by manipulating image parameters of the channel data obtained through the front-end processor;
An image analyzer for analyzing an ultrasound image generated through the image generator to detect a lesion; And
A display unit for providing an analysis result of the image analysis unit;
The medical image processing apparatus comprising: The method according to claim 1,
Wherein the channel data is raw data that has not been subjected to signal processing including beamforming in the image generating unit. 3. The method of claim 2,
Wherein the channel data is a digital RF signal obtained by digitizing an analog RF electric signal converted from an ultrasonic wave reflection signal through a transducer. 3. The method of claim 2,
Wherein the channel data is IQ data obtained by demodulating the analog RF signal converted from the ultrasound reflection signal through the transducer or the digital RF signal obtained by digitizing the analog RF signal to the baseband. 2. The apparatus of claim 1, wherein the front end processor
An analog / digital converter for converting an analog RF electric signal converted from an ultrasonic wave reflected signal through a transducer into a digital RF electric signal;
The medical image processing apparatus comprising: 6. The apparatus of claim 5, wherein the front end processor
A front-end processing unit located at a front end or a rear end of the analog-to-digital conversion unit and demodulating an analog RF electric signal or a digital RF signal to a baseband to generate IQ data;
The medical image processing apparatus further comprising: The apparatus of claim 1, wherein the image generating unit
Wherein processing of channel data signal processing and generation of ultrasound image through manipulation of image parameters is performed through a computer. 2. The apparatus of claim 1, wherein the front end processor
And stores the acquired channel data in a memory of the host computer. The apparatus of claim 8, wherein the image generating unit
And a plurality of new ultrasound images are generated by manipulating image parameters of the channel data read in and read from the memory of the host computer. The apparatus of claim 1, wherein the image generating unit
And generates an ultrasound image from the channel data according to the image parameter set by the user. The apparatus of claim 1, wherein the image generating unit
And generates an ultrasound image from the channel data according to an image parameter corresponding to the lesion-detected ultrasound image through the image analyzer. The apparatus of claim 1, wherein the image analyzing unit
And a computer aided diagnosis (CAD) apparatus for receiving the ultrasound image generated through the image generating unit through a method call through a software interface or a network and analyzing the received ultrasound image. The apparatus of claim 1, wherein the image analyzing unit
Wherein the ultrasound image is analyzed by reflecting additional information received from a user with respect to the ultrasound image. The apparatus of claim 1, wherein the image analyzing unit
A preprocessor for correcting the ultrasound image generated through the image generator;
An image feature extraction unit for extracting an image feature of the ultrasound image corrected through the preprocessing unit; And
A determination unit for determining whether a lesion is present in the ultrasound image using the extracted image feature information;
The medical image processing apparatus comprising: 15. The apparatus of claim 14, wherein the determination unit
Wherein the fuzzy neural network is used to determine whether or not a lesion is present. 15. The apparatus of claim 14, wherein the image analyzing unit
A post-processing unit for providing the determination result through the display unit and providing the feedback information received from the user who has confirmed the determination result to the determination unit for later determination;
The medical image processing apparatus further comprising: The apparatus as claimed in claim 16, wherein the post-processing unit
And provides an ultrasound image having a boundary outline in the lesion area through the display unit to emphasize the lesion area when a lesion is detected. 2. The medical image processing apparatus according to claim 1,
A control unit for requesting the image analysis unit to analyze the image, receiving the image analysis result, generating an alarm signal according to the received analysis result, and providing the generated alarm signal through the display unit;
The medical image processing apparatus further comprising: 19. The apparatus of claim 18, wherein the image analysis unit
Wherein when the lesion is detected on the ultrasound image displayed on the screen through the display unit, the control unit provides the detected image and the identifier of the detected image. 20. The apparatus of claim 19, wherein the control unit
Wherein the image analyzing unit identifies the image and the manipulated image parameter through the identifier of the image in which the lesion is detected by the image analyzing unit, and provides the alarm signal together with the detected image through the display unit. 19. The apparatus of claim 18, wherein the controller
When a lesion is detected by the image analysis unit during the provision of the ultrasound image through the display unit, an alarm signal is provided through the display unit, and when a user's alarm confirmation command for the provided alarm signal is input, A multi-ultrasound image including an ultrasound image and an image of a lesion being provided is acquired through the display unit in non-real-time so as to precisely judge the lesion through comparison between the ultrasound images, and if necessary, And provides a multi-ultrasound image obtained through signal processing and re-scan. 22. The apparatus of claim 21, wherein the control unit
Real-time mode in which a multi-ultrasound image is provided in a non-real-time mode in a non-real-time mode, and a multi-ultrasound image including an ultrasound image and a lesion- Device. 2. The medical image processing apparatus according to claim 1,
A first memory for storing preset image parameters and usage information for each image parameter and providing the stored image parameters to the image generator;
The medical image processing apparatus further comprising: 2. The medical image processing apparatus according to claim 1,
The channel data obtained through the front-end processor and the ultrasound image generated from the obtained channel data are stored in the ultrasound image detected through the analysis of the image analysis unit, and the stored ultrasound images are provided to the image generation unit A second memory;
The medical image processing apparatus further comprising: The display device according to claim 1, wherein the display unit
Wherein the ultrasound image analyzing unit provides the ultrasound image analyzed through the image analyzing unit as a real-time image, a non-real-time image, a review image, or a combination thereof in the form of a multi-ultrasound image. Obtaining channel data from ultrasound signals reflected from a target object;
Generating an ultrasound image by manipulating image parameters of the obtained channel data; And
Analyzing the ultrasound image to detect a lesion and providing an analysis result;
Wherein the medical image processing method comprises the steps of: 27. The method of claim 26,
Wherein the channel data is raw data that does not include signal processing including beamforming. 27. The method of claim 26, wherein providing the analysis results comprises:
Providing an alarm signal when a lesion is detected during the provision of an ultrasound image; and
When a user's alarm confirmation command for the provided alarm signal is inputted, the user acquires a multi-ultrasound image including the ultrasound image and the lesion-detected image so that the user can precisely judge the lesion by comparing the ultrasound images in non- Providing a multi-ultrasound image obtained through signal processing and re-scanning while providing image data in real time and changing image parameters when necessary;
Wherein the medical image processing method comprises the steps of: 29. The method of claim 28, wherein providing the analysis results comprises:
Providing a multi-ultrasound image including non-real-time ultrasound images and lesion-detected ultrasound images in real time while providing multi-ultrasound images in non-real-time mode in a non-real-time mode;
Wherein the medical image processing method further comprises:

Images