Ultrasound imaging apparatus, method of generating sectional image of pelvic floor

Abstract

The application provides an ultrasonic imaging device and a pelvic floor cross-section image generation method. The method comprises the following steps: acquiring three-dimensional volume data of a pelvic floor; identifying at least one key anatomical structure associated with the position of a target tissue from the three-dimensional volume data; or generating at least one two-dimensional cross-section according to the prior position of the key anatomical structure associated with the position of the target tissue in the three-dimensional volume data, identifying at least one key anatomical structure associated with the position of the target tissue from the two-dimensional cross-section, and obtaining the spatial position of the key anatomical structure in the three-dimensional volume data; and imaging the cross-section of the target tissue according to the position of the key anatomical structure in the three-dimensional volume data, and obtaining the cross-section image of the target tissue. Therefore, the ultrasonic doctor only needs to complete the acquisition of the three-dimensional volume data of the pelvic floor, and the cross-section image of the target tissue of the pelvic floor can be obtained, which is simple and convenient, and improves the work efficiency of the ultrasonic doctor.

Description

Technical Field

[0001] This invention relates to the field of medical devices, specifically to an ultrasound imaging device and a method for generating cross-sectional images of the pelvic floor. Background Technology

[0002] In modern medical imaging, ultrasound technology has become the most widely used, most frequently used, and fastest-adopting new technology in examinations due to its advantages such as high reliability, speed, convenience, real-time imaging, and repeatability. The development of new ultrasound technologies, especially those based on artificial intelligence, has further promoted the application of ultrasound in clinical diagnosis and treatment. Particularly in my country, in recent years, due to the increasingly serious aging population and large population base, hospitals are facing a growing demand for ultrasound examinations. However, corresponding medical conditions need further improvement, such as the uneven distribution of medical resources between regions and the significant room for improvement in the technical capabilities of primary care physicians. The intelligent development of ultrasound equipment can help doctors improve examination efficiency, bridge the gap between hospitals in different regions, and provide patients with more accurate diagnoses and personalized treatment plans.

[0003] Stress urinary incontinence and pelvic organ prolapse, among other pelvic floor dysfunctions, are common chronic diseases affecting women's physical and mental health, severely impacting their work and social activities. Pelvic floor ultrasound plays a crucial role in the diagnosis of pelvic floor disorders. With current ultrasound technology, clinicians can dynamically observe changes in pelvic organs and the levator ani hiatus during Valsalva maneuvers and anal contractions. Under standard two-dimensional conditions, the anatomical structure and prolapse status of pelvic floor organs can be dynamically displayed in real time, and clinicians can quantitatively assess the degree of prolapse through measurements. Under three-dimensional / four-dimensional conditions, the coronal plane of the minimum hiatus of the levator ani can be displayed, and multiple parallel cross-sectional imaging of the levator ani hiatus allows for layer-by-layer observation of the levator ani muscle for damage during anal contractions. Rotating the probe 90° and tilting it downwards and backwards provides a complete view of the anal canal cross-section, allowing observation of levator ani muscle damage and surrounding lesions. Furthermore, measurements of the levator ani hiatus area and puborectalis muscle thickness under three-dimensional / four-dimensional conditions help physicians quantitatively assess the degree of damage to the levator ani muscle in the pelvic floor.

[0004] However, during pelvic floor ultrasound diagnosis, obtaining standard sections and imaging multiple parallel sections requires manual operation by the physician. This process demands a high level of experience and skill from the doctor, and is time-consuming and labor-intensive. Furthermore, the measurement process for related parameters is extremely complex. Simultaneously, patients often struggle to perform and maintain Valsalva maneuvers and anal sphincter contractions during the examination, necessitating more accurate and rapid imaging of multiple parallel sections by the physician. After manually obtaining the minimum hiatus section of the levator ani muscle and the corresponding parallel sections of the levator ani hiatus and anal canal cross-section, the ultrasound clinician must manually measure various data, significantly consuming examination time and reducing efficiency. Summary of the Invention

[0005] This invention mainly provides an ultrasound imaging device and a method for generating cross-sectional images of the pelvic floor, in order to improve the working efficiency of ultrasound doctors.

[0006] In one embodiment, a method for generating a cross-sectional image of the pelvic floor is provided, characterized by comprising:

[0007] Obtain the three-dimensional volume data of the pelvic floor;

[0008] The pubic symphysis and anorectal angle are identified from the three-dimensional volume data, and their spatial positions in the three-dimensional volume data are obtained.

[0009] Based on the spatial positions of the pubic symphysis and the anorectal angle in the three-dimensional volume data, the position of the minimum hiatus section of the levator ani muscle in the three-dimensional volume data is determined.

[0010] Based on the location of the minimum hiatus section of the levator ani muscle, the image of the minimum hiatus section of the levator ani muscle is obtained according to the three-dimensional volume data.

[0011] In one embodiment, identifying the pubic symphysis and the anorectal angle from the three-dimensional volume data, and obtaining the spatial positions of the pubic symphysis and the anorectal angle in the three-dimensional volume data includes:

[0012] The lower edge of the pubic symphysis, the two pubic branches, and the anorectal angle are identified from the three-dimensional volume data, and the spatial positions of the lower edge of the pubic symphysis, the two pubic branches, and the anorectal angle in the three-dimensional volume data are obtained.

[0013] Determining the position of the minimum hiatus section of the levator ani muscle in the three-dimensional volume data based on the spatial positions of the pubic symphysis and the anorectal angle includes:

[0014] Based on the spatial positions of the lower edge of the pubic symphysis, the two pubic rami, and the anorectal angle in the three-dimensional volume data, a cross-section is determined that passes through at least a portion of the lower edge of the pubic symphysis and the anorectal angle and is parallel to the line connecting two points on the two pubic rami that are symmetrical about the lower edge of the pubic symphysis. The position of this cross-section is taken as the position of the minimum hiatus cross-section of the levator ani muscle.

[0015] In one embodiment, identifying the pubic symphysis and the anorectal angle from the three-dimensional volume data, and obtaining the spatial positions of the pubic symphysis and the anorectal angle in the three-dimensional volume data includes:

[0016] The lower edge of the pubic symphysis and the anorectal angle are identified from the three-dimensional volume data, and their spatial positions in the three-dimensional volume data are obtained.

[0017] Determining the position of the minimum hiatus section of the levator ani muscle in the three-dimensional volume data based on the spatial positions of the pubic symphysis and the anorectal angle includes:

[0018] Based on the spatial positions of the lower edge of the pubic symphysis and the anorectal angle in the three-dimensional volume data, multiple cross-sections passing through a predetermined point on the lower edge of the pubic symphysis and the anorectal angle are determined, and cross-sectional images of the multiple cross-sections are obtained based on the three-dimensional volume data.

[0019] The symmetry of the image regions on both sides of the line connecting the lower edge of the pubic symphysis and the predetermined point in each of the multiple cross-sectional images is determined, and the location of the cross-sectional image that satisfies the predetermined condition is determined as the location of the minimum hiatus cross-section of the levator ani muscle.

[0020] In one embodiment, the method further includes: measuring the area of ​​the levator ani muscle minimum hiatus based on the obtained cross-sectional image of the levator ani muscle minimum hiatus; and / or detecting the levator ani muscle and urethral orifice in the cross-sectional image of the levator ani muscle minimum hiatus, and measuring the distance from the urethral orifice to the leftmost levator ani muscle and / or the distance from the urethral orifice to the rightmost levator ani muscle based on the detected levator ani muscle and urethral orifice; and / or extracting the contour of the levator ani muscle or the levator ani muscle minimum hiatus from the cross-sectional image of the levator ani muscle minimum hiatus, and calculating the distance between the vertical diameters of the levator ani muscle and / or the distance between the horizontal diameters of the levator ani muscle based on the contour.

[0021] In one embodiment, identifying the pubic symphysis from the three-dimensional volume data includes: calculating at least one feature index of the three-dimensional volume data, inputting the feature index into a pre-established model function that corresponds the feature index of the three-dimensional volume data to the pubic symphysis, and obtaining the corresponding pubic symphysis; or, performing target detection or image segmentation on the three-dimensional volume data, performing morphological feature detection on each region obtained after detection or segmentation to obtain multiple candidate regions, determining that each candidate region is the pubic symphysis based on shape features and / or grayscale features, and identifying the candidate region with the highest probability as the pubic symphysis.

[0022] In one embodiment, identifying the anorectal angle from the three-dimensional volume data includes: calculating at least one feature index of the three-dimensional volume data, inputting the feature index into a pre-established model function that corresponds the feature index of the three-dimensional volume data to the anorectal angle, and obtaining the corresponding anorectal angle; or, performing target detection or image segmentation on the three-dimensional volume data, performing morphological feature detection on each region obtained after detection or segmentation to obtain multiple candidate regions, determining the probability that each candidate region is an anorectal angle based on shape features and / or grayscale features, and determining the candidate region with the highest probability as the anorectal angle.

[0023] In one embodiment, obtaining an image of the levator ani muscle minimum hiatus section based on the location of the levator ani muscle minimum hiatus section from the three-dimensional volume data includes: obtaining image data containing the levator ani muscle minimum hiatus section from the three-dimensional volume data based on the location of the levator ani muscle minimum hiatus section, and performing volume rendering on the image data containing the minimum hiatus section to obtain a volume-rendered image of the minimum hiatus section; and / or, obtaining a two-dimensional image of the levator ani muscle minimum hiatus section from the three-dimensional volume data based on the location of the levator ani muscle minimum hiatus section; and / or, performing thick-layer imaging on the levator ani muscle minimum hiatus section according to a preset imaging thickness to obtain a thick-layer image of the levator ani muscle minimum hiatus section.

[0024] In one embodiment, a method for generating a cross-sectional image of the pelvic floor is provided, comprising:

[0025] Obtain the three-dimensional volume data of the pelvic floor;

[0026] Identify at least one key anatomical structure associated with the location of the target tissue from the three-dimensional volumetric data, and obtain the spatial location of the key anatomical structure in the three-dimensional volumetric data; or, generate at least one two-dimensional section based on the prior location of the key anatomical structure associated with the location of the target tissue in the three-dimensional volumetric data, identify at least one key anatomical structure associated with the location of the target tissue from the two-dimensional section, and obtain the spatial location of the key anatomical structure in the three-dimensional volumetric data.

[0027] Based on the position of the key anatomical structure in the three-dimensional volume data, the cross-section of the target tissue is imaged to obtain a cross-sectional image of the target tissue.

[0028] In one embodiment, the target tissue is the levator ani muscle, and the method further includes:

[0029] The levator ani muscle is detected based on the cross-sectional image of the levator ani muscle, and the minimum hole area of ​​the detected levator ani muscle is measured.

[0030] Alternatively, the levator ani muscle and urethral opening can be detected based on the cross-sectional image of the levator ani muscle, and the distance from the urethral opening to the leftmost levator ani muscle and the distance from the urethral opening to the rightmost levator ani muscle can be measured.

[0031] In one embodiment, the target tissue is the anal canal, and the cross-sectional image of the target tissue is a cross-section of the anal canal; the method further includes:

[0032] The puborectalis muscle was examined based on the cross-section of the anal canal, and the thickness of the puborectalis muscle was measured.

[0033] In one embodiment, identifying at least one key anatomical structure associated with the location of the target tissue from the three-dimensional volumetric data includes:

[0034] Calculate at least one feature index of the three-dimensional volume data, input the feature index into a pre-established model function that maps the feature index of the three-dimensional volume data to the key anatomical structure, and obtain the corresponding key anatomical structure.

[0035] In one embodiment, identifying at least one key anatomical structure associated with the location of the target tissue from the three-dimensional volumetric data includes:

[0036] The three-dimensional volume data is segmented into images. Morphological features are detected in each segmented region to obtain multiple candidate regions. The probability that a candidate region is a key anatomical structure is determined based on its shape features and / or grayscale features. The candidate region with the highest probability is identified as the key anatomical structure.

[0037] In one embodiment, generating at least one two-dimensional section based on the prior location of a key anatomical structure associated with the location of the target tissue in the three-dimensional volumetric data, and identifying at least one key anatomical structure associated with the location of the target tissue from the two-dimensional section includes:

[0038] Based on the prior location of key anatomical structures associated with the location of the target tissue in the three-dimensional volumetric data, the most similar pair of two-dimensional cross-sections are detected on both sides of the prior location, and the key anatomical structures are identified from the central plane of the pair of two-dimensional cross-sections.

[0039] In one embodiment, imaging a cross-section of the target tissue to obtain a cross-sectional image of the target tissue includes:

[0040] Perform volumetric rendering imaging on a cross-section of the target tissue to obtain a volumetric rendering image of the cross-section of the target tissue; and / or,

[0041] Perform grayscale imaging on a cross-section of the target tissue to obtain a grayscale image of the cross-section of the target tissue; and / or,

[0042] Based on user input commands, a reference line is determined; the key anatomical structure is then sectioned and imaged along the reference line to obtain a cross-sectional image of the target tissue; and / or,

[0043] Image the cross-sections of the target tissue to obtain images of multiple parallel cross-sections of the target tissue; and / or,

[0044] Based on a pre-set imaging thickness, a thick-layer image of the cross-section of the target tissue is obtained by performing thick-layer imaging on the cross-section of the target tissue.

[0045] In one embodiment, the key anatomical structure is the target tissue or tissue capable of locating the target tissue.

[0046] In one embodiment, an ultrasound imaging device is provided, comprising:

[0047] An ultrasonic probe is used to emit ultrasonic waves toward an object to be imaged in order to scan the object and to receive ultrasonic echoes returned from the object to be imaged.

[0048] Transmit / receive circuitry is used to control the ultrasonic probe to transmit ultrasonic waves toward the object to be imaged and to receive the echoes of the ultrasonic waves.

[0049] A processor is configured to perform 3D reconstruction based on the echo of the ultrasound to obtain three-dimensional volume data of the object to be imaged; the object to be imaged includes the pelvic floor.

[0050] The processor is further configured to identify at least one key anatomical structure associated with the location of the target tissue from the three-dimensional volumetric data, and obtain the spatial location of the key anatomical structure in the three-dimensional volumetric data; or, based on the prior location of the key anatomical structure associated with the location of the target tissue in the three-dimensional volumetric data, generate at least one two-dimensional cross-section, identify at least one key anatomical structure associated with the location of the target tissue from the two-dimensional cross-section, and obtain the spatial location of the key anatomical structure in the three-dimensional volumetric data; and image the cross-section of the target tissue based on the location of the key anatomical structure in the three-dimensional volumetric data to obtain a cross-sectional image of the target tissue.

[0051] A display for showing cross-sectional images of the target tissue.

[0052] In one embodiment of the ultrasound imaging device, the target tissue is the levator ani muscle, and the processor is further configured to:

[0053] The levator ani muscle is detected based on the cross-sectional image of the levator ani muscle, and the minimum hole area of ​​the detected levator ani muscle is measured.

[0054] Alternatively, the levator ani muscle and urethral opening can be detected based on the cross-sectional image of the levator ani muscle, and the distance from the urethral opening to the leftmost levator ani muscle and the distance from the urethral opening to the rightmost levator ani muscle can be measured.

[0055] In one embodiment of the ultrasound imaging device, the target tissue is the anal canal, and the cross-sectional image of the target tissue is a cross-section of the anal canal; the processor is further configured to:

[0056] The puborectalis muscle was examined based on the cross-section of the anal canal, and the thickness of the puborectalis muscle was measured.

[0057] In one embodiment of the ultrasound imaging apparatus, the processor identifies at least one key anatomical structure associated with the location of the target tissue from the three-dimensional volumetric data, including:

[0058] Calculate at least one feature index of the three-dimensional volume data, input the feature index into a pre-established model function that maps the feature index of the three-dimensional volume data to the key anatomical structure, and obtain the corresponding key anatomical structure.

[0059] In one embodiment of the ultrasound imaging apparatus, the processor identifies at least one key anatomical structure associated with the location of the target tissue from the three-dimensional volumetric data, including:

[0060] The three-dimensional volume data is segmented into images. Morphological features are detected in each segmented region to obtain multiple candidate regions. The probability that a candidate region is a key anatomical structure is determined based on its shape features and / or grayscale features. The candidate region with the highest probability is identified as the key anatomical structure.

[0061] In one embodiment of the ultrasound imaging apparatus, the processor generates at least one two-dimensional section based on the prior location of a key anatomical structure associated with the location of the target tissue in the three-dimensional volumetric data, and identifies at least one key anatomical structure associated with the location of the target tissue from the two-dimensional section, including:

[0062] Based on the prior location of key anatomical structures associated with the location of the target tissue in the three-dimensional volumetric data, the most similar pair of two-dimensional cross-sections are detected on both sides of the prior location, and the key anatomical structures are identified from the central plane of the pair of two-dimensional cross-sections.

[0063] In one embodiment of the ultrasound imaging device, the processor images a cross-section of the target tissue to obtain a cross-sectional image of the target tissue, including:

[0064] Perform volumetric rendering imaging on a cross-section of the target tissue to obtain a volumetric rendering image of the cross-section of the target tissue; and / or,

[0065] Perform grayscale imaging on a cross-section of the target tissue to obtain a grayscale image of the cross-section of the target tissue; and / or,

[0066] Based on user input commands, a reference line is determined; the key anatomical structure is then sectioned and imaged along the reference line to obtain a cross-sectional image of the target tissue; and / or,

[0067] Image the cross-sections of the target tissue to obtain images of multiple parallel cross-sections of the target tissue; and / or,

[0068] Based on a pre-set imaging thickness, a thick-layer image of the cross-section of the target tissue is obtained by performing thick-layer imaging on the cross-section of the target tissue.

[0069] In one embodiment of the ultrasound imaging device, the key anatomical structure is the target tissue or tissue capable of locating the target tissue.

[0070] In one embodiment, an ultrasound imaging device is provided, comprising:

[0071] Memory, used to store programs;

[0072] A processor for executing a program stored in the memory to implement the method described above.

[0073] In one embodiment, a computer-readable storage medium is provided, including a program that can be executed by a processor to implement the method described above.

[0074] According to the ultrasound imaging device and pelvic floor cross-sectional image generation method of the above embodiments, firstly, three-dimensional volume data of the pelvic floor is acquired; at least one key anatomical structure associated with the location of the target tissue is identified from the three-dimensional volume data, and the spatial position of the key anatomical structure in the three-dimensional volume data is obtained; or, based on the prior position of the key anatomical structure associated with the location of the target tissue in the three-dimensional volume data, at least one two-dimensional cross-section is generated, at least one key anatomical structure associated with the location of the target tissue is identified from the two-dimensional cross-section, and the spatial position of the key anatomical structure in the three-dimensional volume data is obtained; based on the position of the key anatomical structure in the three-dimensional volume data, the cross-section of the target tissue is imaged to obtain a cross-sectional image of the target tissue. It is evident that ultrasound physicians only need to complete the acquisition of three-dimensional volume data of the pelvic floor to obtain a cross-sectional image of the target tissue of the pelvic floor, which is simple and convenient to operate and improves the work efficiency of ultrasound physicians. Attached Figure Description

[0075] Figure 1 This is a structural block diagram of an ultrasound imaging device in one embodiment;

[0076] Figure 2 This is a flowchart of a method for generating a cross-sectional image of the pelvic floor in one embodiment;

[0077] Figure 3 An ultrasound image of a sagittal section of the basin;

[0078] Figure 4 A volumetric rendering image of the smallest hiatus section of the levator ani muscle.

[0079] Figure 5 An image showing the effect of an arbitrary cross-section of the smallest hiatus of the levator ani muscle;

[0080] Figure 6 An image showing the effect of multiple parallel cross-sections of the minimum hiatus of the levator ani muscle;

[0081] Figure 7 This is a schematic diagram of a cross-sectional image of the minimum opening of the levator ani muscle in one embodiment. Detailed Implementation

[0082] The present invention will now be described in further detail with reference to specific embodiments and accompanying drawings. Similar elements in different embodiments are referred to by associated similar element reference numerals. In the following embodiments, many details are described to facilitate a better understanding of this application. However, those skilled in the art will readily recognize that some features may be omitted in different situations, or may be replaced by other elements, materials, or methods. In some cases, certain operations related to this application are not shown or described in the specification. This is to avoid obscuring the core parts of this application with excessive description. For those skilled in the art, detailed description of these related operations is not necessary; they can fully understand the related operations based on the description in the specification and general technical knowledge in the art.

[0083] Furthermore, the features, operations, or characteristics described in the specification can be combined in any suitable manner to form various embodiments. At the same time, the steps or actions in the method description can be rearranged or adjusted in a manner obvious to those skilled in the art. Therefore, the various orders in the specification and drawings are only for the clear description of a particular embodiment and do not imply a necessary order, unless otherwise stated that a particular order must be followed.

[0084] The serial numbers assigned to components in this document, such as "first" and "second," are used only to distinguish the described objects and have no sequential or technical meaning. The terms "connection" and "linkage" used in this application, unless otherwise specified, include both direct and indirect connections (linkages).

[0085] This invention acquires three-dimensional volume data of the pelvic floor; automatically detects and identifies key anatomical structures within the three-dimensional volume data; automatically images the pelvic floor based on these key anatomical structures; performs measurements based on the automatic imaging results; and finally displays the imaging and measurement results on a monitor. This achieves automatic imaging and measurement of pelvic floor cross-sectional images, improving the work efficiency of ultrasound physicians. The three-dimensional volume data can be generated by ultrasound imaging equipment, MRI equipment, or computed tomography (CT). The following section uses ultrasound imaging equipment as an example to describe in detail the ultrasound imaging equipment and its method for generating pelvic floor cross-sectional images.

[0086] like Figure 1 As shown, the ultrasound imaging device provided by the present invention includes an ultrasound probe 30, a transmitting / receiving circuit 40 (i.e., a transmitting circuit 410 and a receiving circuit 420), a beamforming module 50, an IQ demodulation module 60, a processor 20, a human-computer interaction device 70, and a memory 80.

[0087] The ultrasound probe 30 includes a transducer (not shown) composed of multiple array elements arranged in an array. These elements can be arranged in a row to form a linear array, or in a two-dimensional matrix to form a planar array. They can also form a convex array. Each element is used to emit an ultrasonic beam according to an excitation electrical signal, or to convert a received ultrasonic beam into an electrical signal. Therefore, each element can be used to convert between electrical pulse signals and ultrasonic beams, thereby emitting ultrasonic waves towards the object to be imaged (e.g., the pelvic floor of the human body in this embodiment), or to receive echoes of ultrasonic waves reflected back from tissues. During ultrasound detection, the transmitting circuit 410 and the receiving circuit 420 can control which elements are used to emit ultrasonic beams and which are used to receive ultrasonic beams, or control the elements to be used in time-slotted manner to emit ultrasonic beams or receive echoes of ultrasonic beams. Elements participating in ultrasonic wave emission can be simultaneously excited by electrical signals to emit ultrasonic waves simultaneously; or elements participating in ultrasonic wave emission can be excited by several electrical signals with a certain time interval to continuously emit ultrasonic waves with a certain time interval.

[0088] The array elements, for example, employ piezoelectric crystals, which convert electrical signals into ultrasonic signals according to the transmission sequence transmitted by the transmitting circuit 410. Depending on the application, the ultrasonic signals may include one or more scanning pulses, one or more reference pulses, one or more driving pulses, and / or one or more Doppler pulses. Depending on the wave morphology, the ultrasonic signals include focused waves and plane waves.

[0089] The user selects a suitable position and angle by moving the ultrasonic probe 30 to emit ultrasonic waves toward the object to be imaged 10 and receives the echo of the ultrasonic waves returned by the object to be imaged 10, and outputs an ultrasonic echo signal. The ultrasonic echo signal is a channel analog electrical signal formed by the receiving array element, which carries amplitude information, frequency information and time information.

[0090] The transmitting circuit 410 generates a transmission sequence under the control of the processor 20. This transmission sequence controls some or all of the multiple array elements to transmit ultrasound waves to the biological tissue. The transmission sequence parameters include the position and number of array elements, and the ultrasound beam transmission parameters (e.g., amplitude, frequency, number of transmissions, transmission interval, transmission angle, waveform, focusing position, etc.). In some cases, the transmitting circuit 410 also performs phase delay on the transmitted beam, allowing different transmitting array elements to transmit ultrasound waves at different times, so that each transmitted ultrasound beam can be focused in a predetermined region of interest. Different operating modes, such as B-image mode, C-image mode, and D-image mode (Doppler mode), may have different transmission sequence parameters. After the echo signal is received by the receiving circuit 420 and processed by subsequent modules and corresponding algorithms, a B-image reflecting the tissue anatomy, a C-image reflecting the tissue anatomy and blood flow information, and a D-image reflecting the Doppler spectrum can be generated.

[0091] The receiving circuit 420 receives and processes ultrasonic echo signals from the ultrasonic probe 30. The receiving circuit 420 may include one or more amplifiers, analog-to-digital converters (ADCs), etc. The amplifier amplifies the received echo signal after appropriate gain compensation. The amplifier samples the analog echo signal at predetermined time intervals, converting it into a digitized signal. The digitized echo signal still retains amplitude, frequency, and phase information. The data output from the receiving circuit 420 can be sent to the beamforming module 50 for processing or to the memory 80 for storage.

[0092] The beamforming module 50 is signal-connected to the receiving circuit 420 and is used to perform beamforming processing on the echo signal, including delay and weighted summation. Because the distance from the ultrasonic receiving point in the tested tissue to the receiving array elements varies, the channel data from the same receiving point output by different receiving array elements has delay differences, requiring delay processing to align the phases. Weighted summation of the different channel data from the same receiving point is then performed to obtain the beamformed ultrasonic image data. The ultrasonic image data output by the beamforming module 50 is also called radio frequency (RF) data. The beamforming module 50 outputs the RF data to the IQ demodulation module 60. In some embodiments, the beamforming module 50 can also output the RF data to the memory 80 for caching or storage, or directly output the RF data to the processor 20 for image processing.

[0093] The beamforming module 50 can perform the above functions in hardware, firmware or software. For example, the beamforming module 50 may include a central controller circuit (CPU) capable of processing input data according to specific logic instructions, one or more microprocessor chips or any other electronic components. When the beamforming module 50 is implemented in software, it can execute instructions stored on a tangible and non-transitory computer-readable medium (e.g., memory) to perform beamforming calculations using any appropriate beamforming method.

[0094] The IQ demodulation module 60 removes the signal carrier through IQ demodulation, extracts the tissue structure information contained in the signal, and filters to remove noise. The signal obtained at this time is called the baseband signal (IQ data pair). The IQ demodulation module 60 outputs the IQ data pair to the processor 20 for image processing.

[0095] In some embodiments, the IQ demodulation module 60 also caches or saves the IQ data output to the memory 80 so that the processor 20 can read the data from the memory 80 for subsequent image processing.

[0096] The IQ demodulation module 60 can also perform the above functions in hardware, firmware or software. In some embodiments, the IQ demodulation module 60 can also be integrated with the beamforming module 50 in a single chip.

[0097] The processor 20 is configured to process input data according to specific logical instructions. It is a central controller circuit (CPU), one or more microprocessors, a graphics controller circuit (GPU), or any other electronic component. It can control peripheral electronic components according to input instructions or predetermined instructions, or perform data reading and / or saving on the memory 80. It can also process input data by executing programs in the memory 80. For example, it can perform one or more processing operations on the acquired ultrasound data according to one or more operating modes. The processing operations include, but are not limited to, adjusting or limiting the form of ultrasound waves emitted by the ultrasound probe 30, generating various image frames for display on the display of the human-machine interface device 70, or adjusting or limiting the content and form displayed on the display, or adjusting one or more image display settings displayed on the display (e.g., ultrasound images, interface components, locating regions of interest).

[0098] When an echo signal is received, the acquired ultrasound data can be processed in real time by the processor 20 during scanning or treatment, or it can be temporarily stored in the memory 80 and processed in a near real-time manner during online or offline operation.

[0099] In this embodiment, the processor 20 controls the operation of the transmitting circuit 410 and the receiving circuit 420, for example, controlling the transmitting circuit 410 and the receiving circuit 420 to operate alternately or simultaneously. The processor 20 can also determine a suitable operating mode according to the user's selection or the program settings, form a transmission sequence corresponding to the current operating mode, and send the transmission sequence to the transmitting circuit 410 so that the transmitting circuit 410 can use the appropriate transmission sequence to control the ultrasonic probe 30 to emit ultrasonic waves.

[0100] The processor 20 is also used to process the ultrasound data to generate a grayscale image showing the changes in signal intensity within the scanning range. This grayscale image reflects the anatomical structure inside the tissue and is called a B-image. The processor 20 can output the B-image to the display of the human-computer interaction device 70 for display.

[0101] The human-computer interaction device 70 is used for human-computer interaction, that is, to receive user input and output visual information; it can receive user input through a keyboard, operation buttons, mouse, trackball, etc., or it can use a touch screen integrated with the display; it can output visual information through a display.

[0102] based on Figure 1 The ultrasound imaging device shown in the diagram generates a cross-sectional image of the pelvic floor as follows: Figure 2As shown, it includes the following steps:

[0103] Step 1: The processor 20 performs 3D reconstruction based on the ultrasound data stored in the memory 80, or based on the ultrasound data output by the beamforming module 50, or based on the ultrasound data output by the IQ demodulation module 60, to obtain the three-dimensional volume data (three-dimensional volume data) of the pelvic floor, and outputs the three-dimensional volume data to the display of the human-computer interaction device 70 for display.

[0104] Step 2: The processor 20 identifies key anatomical structures from the three-dimensional volumetric data. In this embodiment, the target tissue is the levator ani muscle. The key anatomical structures associated with it can be the target tissue itself or tissues that can locate the target tissue. In this embodiment, there can be two key anatomical structures associated with it: the lower border of the pubic symphysis (or its surrounding tissues) and the anorectal angle. For the identification of key anatomical structures in the pelvic floor ultrasound, the lower border of the pubic symphysis can be an independent coordinate point, or it can include / or include the anatomical structures around the lower border (such as the pubic ramus structure at the lower border of the pubic symphysis). Furthermore, the detection of key anatomical structures is not limited to the lower border of the pubic symphysis and the anorectal angle. In this embodiment, for ease of explanation, the key anatomical structures of the levator ani muscle mainly refer to the lower border of the pubic symphysis and the anorectal angle. Step 2 specifically includes two methods: direct identification and indirect identification.

[0105] The direct identification method is described in step 2.1, and includes:

[0106] The processor 20 identifies at least one key anatomical structure associated with the location of the target tissue from the three-dimensional volumetric data and obtains the spatial location of the key anatomical structure in the three-dimensional volumetric data. Specifically, the detection of key anatomical structures can be achieved based on machine learning or deep learning methods. For example, at least one feature index of the three-dimensional volumetric data is calculated, and the feature index is input into a pre-established model function that maps the feature indexes of the three-dimensional volumetric data to the key anatomical structures to obtain the corresponding key anatomical structure. The main steps are as follows:

[0107] Database construction steps: The database typically contains multiple sets of three-dimensional volume data of the pelvic floor from ultrasound and the calibration results of key anatomical structures. Among them, the calibration results can be set according to the actual task requirements. It can be a ROI (region of interest) containing the target, or a mask that precisely segments the posterior inferior margin of the pubic symphysis (or its surrounding tissues) and the anorectal angle region.

[0108] Localization and recognition steps: After building the database, a machine learning algorithm is designed to learn the features or patterns in the database that can distinguish target regions (key anatomical structure regions) from non-target regions (background regions) to achieve image localization and recognition. This implementation step includes, but is not limited to, the following situations.

[0109] The first scenario can employ a traditional sliding window-based approach, which typically involves: firstly, extracting features from the region within the sliding window. Feature extraction methods can include traditional PCA, LDA, Haar features, texture features, etc., or deep neural networks can be used for feature extraction. Then, the extracted features are matched with a database, and discriminators such as KNN, SVM, random forest, and neural networks are used for classification to determine whether the current sliding window is a region of interest and to obtain its corresponding category.

[0110] The second approach involves using deep learning-based bounding-box detection and recognition. A common approach is to learn features and regress parameters on a constructed database by stacking convolutional and fully connected layers. For an input image, the network can directly regress the bounding-box of the corresponding region of interest and simultaneously obtain the category of the tissue structure within that region of interest. Common networks include R-CNN, Fast R-CNN, Faster R-CNN, SSD, and YOLO.

[0111] The third type is the end-to-end semantic segmentation network method based on deep learning. This type of method is similar to the structure of the second type of deep learning-based bounding-box method. The difference is that the fully connected layer is removed and an upsampling or deconvolution layer is added to make the input and output sizes the same, so as to directly obtain the region of interest and its corresponding category of the input image. Common networks include FCN, U-Net, Mask R-CNN, etc.

[0112] The fourth scenario involves using only the methods from the first, second, or third scenarios to locate the target, and then designing an additional classifier to classify the target based on the localization results. Common classification methods include: first, extracting features from the target ROI or mask. Feature extraction methods can include traditional PCA, LDA, Haar features, texture features, etc., or deep neural networks can be used for feature extraction. Then, the extracted features are matched against a database, and discriminators such as KNN, SVM, random forest, and other neural networks are used for classification.

[0113] The machine learning or deep learning algorithms described above can automatically locate the positions of key anatomical structures such as the posteroinferior margin of the pubic symphysis and the anorectal angle in three-dimensional volumetric data, thus facilitating the subsequent localization of the standard minimum hiatus area of ​​the levator ani muscle, which can then serve as the basis for subsequent imaging.

[0114] In ultrasound pelvic floor data, if the pubic rami are parallel and symmetrical, then the plane of symmetry corresponding to the pubic rami is a standard midsagittal section, such as... Figure 3 As shown, after determining the midsagittal section, the echoes of the pubic symphysis and the anorectal angle show significant differences from the surrounding tissues: the pubic symphysis appears as a bright oval, the fascia at the outer edge of the pubic ramus is a bright tissue surrounding the outer edge of the fascia, and the posteroinferior border of the pubic symphysis is located at the intersection of the midline of the pubic ramus and the fascia; when the examinee is in different action states (resting, anal contraction, and valsalva), the anorectal angle presents a certain angle, and this angle is generally no greater than 180°. Therefore, traditional grayscale and / or morphological feature detection methods can be used to detect the posteroinferior border of the pubic symphysis and the anorectal angle. For example, firstly, image segmentation is performed on the three-dimensional volume data, specifically binary segmentation. Morphological feature detection is performed on each segmented region to obtain multiple candidate regions. Then, for each candidate region, the probability that the candidate region is a key anatomical structure (the posteroinferior border of the pubic symphysis and the anorectal angle) is determined based on shape features and / or grayscale features, and the candidate region with the highest probability is identified as the key anatomical structure. Of course, other traditional grayscale detection and segmentation methods can also be used, such as OTSU thresholding, level set, graph cut, and snake.

[0115] The indirect identification method is described in step 2.2: The processor 20 generates at least one two-dimensional section based on the prior location of key anatomical structures associated with the target tissue in the three-dimensional volumetric data. From these two-dimensional sections, at least one key anatomical structure associated with the target tissue is identified, and the spatial location of the key anatomical structure in the three-dimensional volumetric data is obtained. Specifically, based on the prior location of the key anatomical structure associated with the target tissue in the three-dimensional volumetric data, the processor 20 detects the most similar pair of two-dimensional sections on both sides of the prior location, and identifies the key anatomical structure from the central plane of this pair of two-dimensional sections. In this embodiment, the processor 20 detects the most similar pair of two-dimensional sections on both sides of the sagittal section of the pelvic floor, and identifies the posterior inferior margin of the pubic symphysis and the anorectal angle from the central plane of this pair of two-dimensional sections. Directly identifying the midsagittal section can lead to problems with accuracy, making it difficult to find the ideal midsagittal section. However, this invention, through symmetrical searching, accurately identifies the key anatomical structure without needing to find the midsagittal section.

[0116] The aforementioned method for identifying key anatomical structures is automated, but it can also be done manually. Manually acquiring anatomical structures involves users using tools such as a keyboard and mouse to click or draw lines on specific anatomical structures within the 3D volumetric data, based on a defined workflow, to inform the device of the type and location of key structural elements.

[0117] In one embodiment, when identifying key anatomical structures, the lower border of the pubic symphysis, the two pubic rami, and the anorectal angle can be directly identified from the three-dimensional volume data, thereby obtaining their spatial positions in the three-dimensional volume data. Then, based on the spatial positions of the lower border of the pubic symphysis, the two pubic rami, and the anorectal angle in the three-dimensional volume data, a section passing through at least a portion of the lower border of the pubic symphysis and the anorectal angle and parallel to the line connecting two points symmetrical to the lower border of the pubic symphysis relative to the lower border of the pubic symphysis can be determined. This section can be considered as the section of the minimum hiatus of the levator ani muscle, and therefore, the position of this section can be taken as the position of the section of the minimum hiatus of the levator ani muscle.

[0118] In one embodiment, when identifying key anatomical structures, the lower border of the pubic symphysis and the anorectal angle can be directly identified from the three-dimensional volume data, thereby obtaining their spatial positions in the three-dimensional volume data. Then, based on the spatial positions of the lower border of the pubic symphysis and the anorectal angle in the three-dimensional volume data, multiple cross-sections passing through a predetermined point on the line connecting the lower border of the pubic symphysis and the anorectal angle can be determined, and cross-sectional images of these multiple cross-sections can be obtained based on the three-dimensional volume data. Subsequently, based on the image data of these cross-sectional images, the symmetry of the image regions (all or part of the image regions) located on both sides of the line connecting the lower border of the pubic symphysis and the predetermined point in each cross-sectional image can be determined, and the cross-sections whose symmetry satisfies predetermined conditions are considered as the levator ani muscle minimum hiatus cross-sections. Therefore, the location of the cross-section image whose symmetry satisfies predetermined conditions can be determined as the location of the levator ani muscle minimum hiatus cross-section. Here, the "predetermined conditions" can be set according to actual circumstances. For example, in one embodiment, the predetermined condition may be set to "maximum symmetry". In another embodiment, the predetermined condition may be set to symmetry greater than a specific value, symmetry less than a specific value, or symmetry within a specific range, and so on. Conventional methods can be used to determine the symmetry of an image region, and will not be elaborated upon here.

[0119] Step 3: Based on the location of key anatomical structures in the three-dimensional volumetric data, processor 20 images the cross-section of the target tissue, obtaining a cross-sectional image of the target tissue, and displays the cross-sectional image of the target tissue on a monitor. For example, from... Figure 3 After identifying the lower border of the pubic symphysis and the anorectal angle, the three-dimensional volume data is sectioned with a plane. The resulting section is the required minimum hiatus section of the levator ani muscle. This plane is aligned with... Figure 3 The intersecting straight lines in the planes pass through the lower border of the pubic symphysis and the anorectal angle, such as... Figure 3 As shown by the straight line, the cross-sectional image of the target tissue can be presented in the following ways: volumetric rendering imaging, cross-sectional imaging, arbitrary cross-section (planar or curved) imaging, multi-parallel plane imaging and cross-sectional imaging, arbitrary cross-section (planar or curved) imaging, and a combination of multiple parallel plane imaging and thick-layer imaging.

[0120] Volumetric rendering imaging adjusts the position, size, and curvature of the volume of interest (VOI) and imaging curve based on key anatomical structures detected in 3D volumetric data to display a rendered image of the levator ani hiatus. Sectional imaging adjusts the spatial position of the section based on key anatomical structures detected in 3D volumetric data to generate a sectional image of the levator ani hiatus. Arbitrary section (planar or curved) imaging generates reference lines (straight lines or curves) based on key anatomical structures detected in 3D volumetric data, and obtains the minimum hiatus section of the levator ani muscle through these reference lines. Multi-parallel plane imaging, based on detected key anatomical locations, uses the minimum hiatus area section of the levator ani muscle as a reference to achieve multiple parallel section images at equal intervals.

[0121] The effect of volumetric rendering imaging of the minimum hiatus area of ​​the levator ani muscle by processor 20 is as follows: Figure 4 As shown. For 3D stereoscopic imaging, volumetric rendering imaging displays the 3D volumetric data within the VOI frame a using algorithms such as ray tracing through different imaging modes. Obtaining a good volumetric rendering image requires adjusting the orientation of the volume data, as well as setting the size and position of the VOI frame a. For ultrasound pelvic floor 3D volumetric data, after detecting anatomical structures such as the pubic symphysis and anorectal angle, the orientation of the volume data can be automatically adjusted based on the specific locations of the posterior inferior border of the pubic symphysis and the anorectal angle, ensuring that the posterior inferior border of the pubic symphysis and the anorectal angle are at the same horizontal level, and that the VOI imaging curve coincides as closely as possible with the straight line containing the inferior border of the pubic symphysis and the anorectal angle. Simultaneously, adjusting the VOI frame a to a suitable size (generally 2.5mm) allows for the rendering of the minimum hiatus section of the levator ani muscle, such as... Figure 5 As shown in the bottom right corner of the image.

[0122] Processor 20 performs grayscale imaging of the levator ani muscle minimum hiatus area section: a cross-sectional image refers to a cross-sectional image at a specific location and orientation in the three-dimensional volume data. By adjusting rotation and translation, cross-sectional images at different locations and orientations can be obtained. Based on the anatomical structures such as the lower border of the pubic symphysis and the anorectal angle detected in the three-dimensional volume data, the standard section of the levator ani muscle minimum hiatus area can be directly obtained through planar imaging. Typically, the levator ani muscle minimum hiatus section is a two-dimensional section through the lower border of the pubic symphysis and the anorectal angle. Therefore, by using the key anatomical structure locations detected in the three-dimensional volume data in the previous step, a plane can be generated that passes through (or approximately passes through) the section containing the lower border of the pubic symphysis and the anorectal angle. The plane equation can be obtained by solving the equation or fitting it. After obtaining the plane equation, the grayscale image corresponding to this plane can be extracted from the three-dimensional volume data, thus obtaining the levator ani muscle minimum hiatus area section. Based on automatic imaging, users can also manually fine-tune the levator ani muscle hiatus section position based on the location of the lower border of the pubic symphysis and the anorectal angle in the three-dimensional volume data.

[0123] Processor 20 performs automatic cross-sectional imaging of the minimum hiatus area of ​​the levator ani muscle: Besides displaying a volumetric rendering image of the minimum hiatus area reconstructed in 3D, the minimum hiatus area of ​​the levator ani muscle can also be displayed using arbitrary cross-sectional imaging, as shown in the following specific forms: Figure 5 As shown. Arbitrary profile imaging involves taking one or more reference lines (straight lines or curves) from a certain cross-section of the three-dimensional volumetric data. Figure 5 The reference line (represented by a straight line) will form a cross-section (planar or curved) with the 3D data. This cross-section is then extracted and straightened into a plane for display. Typically, the formation of the minimum hiatus cross-section of the levator ani muscle involves placing the reference line at the straight line formed by the lower edge of the pubic symphysis and the anorectal angle. After obtaining key anatomical structures such as the lower edge of the pubic symphysis and the anorectal angle through the automatic recognition method described above, the imaging line can be placed on the straight line formed by these two anatomical positions, achieving automated imaging of any cross-section of the minimum hiatus of the levator ani muscle. Alternatively, the area cross-section of the minimum hiatus of the levator ani muscle can be obtained manually. For example, a reference line can be determined based on user input, and the key anatomical structures can be sectioned and imaged using the reference line to obtain the area cross-section of the minimum hiatus of the levator ani muscle.

[0124] Processor 20 performs multi-parallel plane automatic imaging of the levator ani muscle minimum hiatus area: During clinical ultrasound examinations, ultrasound clinicians typically need to observe the levator ani muscle hiatus area based on multiple parallel planes to obtain more comprehensive diagnostic information. Multi-parallel plane imaging refers to an imaging method that simultaneously displays multiple parallel planes, with equal distances between adjacent planes, and the user can adjust the distance between parallel planes. After detecting key anatomical locations such as the pubic symphysis and anorectal angle, the levator ani muscle minimum hiatus area section can be obtained using the aforementioned automatic imaging method. Then, using this section as a reference, equidistant multi-plane parallel imaging of the levator ani muscle is performed, such as... Figure 6 As shown. Users can change the position of the reference plane or reselect the reference plane according to the actual situation. At the same time, the spacing between frames can be fixed and equal, or it can be adjusted automatically or manually according to the actual situation.

[0125] The processor 20 combines the aforementioned cross-sectional imaging, arbitrary cross-sectional imaging, and multiple parallel section imaging with thick-slice imaging: In addition to the four methods mentioned above, the imaging mode for the levator ani hiatus section can also combine cross-sectional imaging, arbitrary cross-sectional imaging, and multiple parallel section imaging with thick-slice imaging. Thick-slice imaging refers to adding a certain thickness to a tomographic grayscale image and displaying it using surface mode, X-ray mode, or a fusion of both. The thickness can be a pre-set imaging thickness. This mode can effectively improve the contrast resolution of the image, enhancing the display of key anatomical structures and features, but attenuating some image details. Simultaneously, the image thickness of thick-slice imaging can be set to a fixed value (e.g., 2.5 mm) according to actual clinical needs, or the thickness parameter can be adaptively adjusted according to the actual anatomical structures and features. Of course, users can also manually set this parameter based on personal needs and operating habits.

[0126] During ultrasound scanning, after scanning the minimum hiatus area of ​​the levator ani muscle, rotating the ultrasound probe 90° and tilting it downwards and backwards allows for a complete display of the anal canal's cross-section. Therefore, after obtaining the minimum hiatus area of ​​the levator ani muscle, the processor 20 can generate a cross-sectional image of the anal canal based on the positional relationship between the anal canal's cross-section and the minimum hiatus area of ​​the levator ani muscle. Alternatively, the target tissue can be set as the anal canal, and the cross-sectional image of the anal canal can be obtained using the above method. The process is similar to obtaining the minimum hiatus area of ​​the levator ani muscle, and therefore will not be elaborated upon. The imaging method for the puborectalis muscle section is the same as that for the levator ani muscle minimum hiatus area section. That is, after detecting the key anatomical sites related to the anal sphincter, the puborectalis muscle can be imaged in a variety of ways, namely, automatic imaging of the puborectalis muscle section volume rendering, section imaging, arbitrary section (planar or curved) imaging, multi-parallel plane imaging section imaging, arbitrary section (planar or curved) imaging, and the combination of multiple parallel plane imaging with thick-slice imaging.

[0127] For cases with poor image quality or deviations in the location of key anatomical structures detected by the algorithm, users can use tools such as keyboard and mouse to move, scale, delete, and recalibrate the VOI region or arbitrary cross-sectional imaging curve in the detected section, achieving semi-automatic VOI imaging or arbitrary cross-sectional imaging. For standard cross-sectional plane imaging of the pelvic floor and imaging of multiple parallel cross-sections, users can also adjust the cross-section using a knob.

[0128] Step 4: Measure the cross-sectional image of the target tissue. After obtaining the minimum hiatus section of the levator ani muscle, doctors usually need to perform relevant measurements on the corresponding section to quantitatively assess the prolapse and tearing of pelvic floor organs. However, the relevant measurements are often complex, time-consuming, and labor-intensive, requiring a high level of experience and skill from the doctor, making it difficult to guarantee the robustness and accuracy of the measurement results. Therefore, intelligent measurement of corresponding items based on the pelvic floor section obtained above is crucial.

[0129] Based on the minimum hiatus section of the levator ani muscle, ultrasound clinicians currently routinely measure two items: the area of ​​the minimum hiatus and the distance from the urethral orifice to the leftmost and rightmost levator ani muscles. For volumetric rendering imaging, sectional imaging, and arbitrary sectional imaging, intelligent measurement of these two items only needs to be implemented on a single frame image. For multiple parallel sectional imaging, intelligent measurement needs to be performed on one or more frames. The selection of the measurement frame and the number of frames can be set by the user based on their needs; the default setting is... Figure 6 The middle three consecutive frames of images.

[0130] Therefore, in this embodiment, step 4 includes: the processor 20 detects the levator ani muscle based on the cross-sectional image of the levator ani muscle (the minimum hiatus cross-section of the levator ani muscle), performs area statistics on the detected levator ani muscles, and obtains the minimum hiatus area of ​​the levator ani muscle; the processor 20 detects the levator ani muscle and the urethral opening based on the cross-sectional image of the levator ani muscle, and measures the distance from the urethral opening to the leftmost levator ani muscle and the distance from the urethral opening to the rightmost levator ani muscle.

[0131] Specifically, processor 20 identifies the target to be measured, such as the levator ani muscle, based on traditional grayscale and / or morphological target segmentation methods, and then performs measurement: In the levator ani hiatus section, the levator ani muscle presents as a bright muscle fiber acoustic beam, and its ultrasound imaging features are significantly different from the surrounding anatomical structures. Therefore, similar to the key anatomical structure detection approach mentioned earlier, processor 20 uses traditional grayscale and / or morphological feature detection and segmentation methods to detect and segment the levator ani muscle. For example, first, the levator ani hiatus section image is binarized and segmented. After some necessary morphological operations, multiple candidate regions are obtained. Then, for each candidate region, the probability that the region is the levator ani muscle is determined based on features such as shape and grayscale brightness. The region with the highest probability is selected as the target segmentation region. Of course, other traditional grayscale detection and segmentation methods can also be used, such as OTSU thresholding, level set, graph cut, and Snake.

[0132] The processor 20 can also identify the target to be measured, such as the levator ani muscle, based on target segmentation methods such as machine learning or deep learning, and then perform measurement. Besides the traditional image segmentation methods mentioned above, similar to the target detection methods described, the levator ani muscle can also be segmented based on target segmentation methods such as machine learning and deep learning. This segmentation method can refer to the third and fourth cases in the corresponding target detection methods mentioned earlier. That is, the levator ani muscle can be segmented directly based on an end-to-end deep learning semantic segmentation network; alternatively, the target can be located first based on end-to-end network segmentation, and then a classifier can be designed to perform pixel-level classification and interpretation based on the location results, achieving levator ani muscle segmentation in two steps. These two approaches are similar to the target detection methods mentioned earlier and will not be elaborated upon here.

[0133] After segmenting the levator ani muscle, other measurements can be calculated based on the levator ani hiatus section. These include the distance from the urethral opening to the leftmost side of the levator ani muscle, and the distance from the urethral opening to the rightmost side of the levator ani muscle. It is important to note that the automatic measurement of these measurements requires the algorithm system to automatically detect the urethral opening. The automatic detection method is the same as the automatic detection method for key anatomical structures described above, and will not be repeated here.

[0134] In one embodiment, the contour of the levator ani muscle or the levator ani muscle's minimum hiatus can also be extracted from the cross-sectional image of the levator ani muscle's minimum hiatus, and the distance between the superior and inferior diameters of the levator ani muscle and / or the distance between the left and right diameters of the levator ani muscle can be calculated based on this contour. Figure 7 As shown, A represents the extracted outline of the levator ani muscle or the smallest hiatus of the levator ani muscle, B represents the left-right diameter of the levator ani muscle, and C represents the upper-lower diameter of the levator ani muscle. Here, the distance between the upper and lower diameters can be the length of the upper-lower diameter C, and the distance between the left and right diameters can be the length of the left-right diameter B.

[0135] In the embodiment where the target tissue is the anal canal, the cross-sectional image of the target tissue is a cross-section of the anal canal. The processor 20 detects the puborectalis muscle based on the cross-section of the anal canal and measures the thickness of the detected puborectalis muscle. Specifically, the processor 20 measures the thickness of the puborectalis muscle based on multiple parallel cross-sections. This thickness measurement is performed three times consecutively in the corresponding peripheral area of ​​the same frame image, and the results of each measurement are displayed and averaged. The selection of the measurement frame, frame number, and number of measurements can be customized by the user based on their needs; the default is to take three measurement results from the middle frame. In the embodiment where the target tissue is the anal canal, the automatic measurement process is similar to that of the levator ani muscle embodiment, only the object being detected differs, so it will not be described in detail here.

[0136] The measurements described above can be performed fully automatically or semi-automatically. Semi-automatic measurement allows users to set one or more input points on the corresponding interface and obtain the corresponding measurement results based on those input points. Furthermore, for cases where image quality is poor and the measurement results obtained through intelligent algorithms are inaccurate, users can also use tools such as the keyboard and mouse to delete, modify, or re-enter the results.

[0137] Step 5: The processor 20 displays the cross-sectional image of the target tissue and its measurement results on the display of the human-computer interaction device 70. Since doctors may need to measure different target tissues, a pre-defined workflow can be established, integrating the doctor's selection function for the target tissue into the workflow for the doctor to choose freely. The image and measurement results corresponding to the selected function are then displayed on the monitor.

[0138] In summary, the ultrasound imaging device and the method for generating cross-sectional images of the pelvic floor provided by this invention can not only automatically generate cross-sections of target tissues in the pelvis, but also perform corresponding measurements, greatly improving the working efficiency of ultrasound physicians.

[0139] Those skilled in the art will understand that all or part of the functions of the various methods in the above embodiments can be implemented by hardware or by computer programs. When all or part of the functions in the above embodiments are implemented by computer programs, the program can be stored in a computer-readable storage medium, which may include: read-only memory, random access memory, disk, optical disk, hard disk, etc., and the program is executed by a computer to achieve the above functions. For example, the program can be stored in the memory of a device, and when the program in the memory is executed by the processor, all or part of the above functions can be achieved. In addition, when all or part of the functions in the above embodiments are implemented by computer programs, the program can also be stored in a server, another computer, disk, optical disk, flash drive, or external hard drive, etc., and can be downloaded or copied to the memory of a local device, or the system of the local device can be updated. When the program in the memory is executed by the processor, all or part of the functions in the above embodiments can be achieved.

[0140] This document describes various exemplary embodiments with reference to them. However, those skilled in the art will recognize that changes and modifications can be made to the exemplary embodiments without departing from the scope of this document. For example, various operational steps and components for performing operational steps can be implemented in different ways depending on the specific application or considering any number of cost functions associated with the operation of the system (e.g., one or more steps can be deleted, modified, or combined with other steps).

[0141] Furthermore, as those skilled in the art will understand, the principles herein can be reflected in a computer program product on a computer-readable storage medium pre-loaded with computer-readable program code. Any tangible, non-transitory computer-readable storage medium may be used, including magnetic storage devices (hard disks, floppy disks, etc.), optical storage devices (CD-ROMs, DVDs, Blu-ray discs, etc.), flash memory, and / or the like. These computer program instructions may be loaded onto a general-purpose computer, a special-purpose computer, or other programmable data processing apparatus to form a machine, such that instructions, which execute on the computer or other programmable data processing apparatus, can generate means for performing a specified function. These computer program instructions may also be stored in a computer-readable storage medium that can instruct the computer or other programmable data processing apparatus to operate in a particular manner, such that instructions stored in the computer-readable storage medium can form an article of manufacture, including means for implementing the specified function. The computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to perform a series of operational steps on the computer or other programmable apparatus to produce a computer-implemented process, such that instructions, which execute on the computer or other programmable apparatus, can provide steps for implementing the specified function.

[0142] While the principles herein have been illustrated in various embodiments, numerous modifications to the structures, arrangements, proportions, elements, materials, and components, particularly suited to specific environments and operational requirements, may be used without departing from the principles and scope of this disclosure. These modifications and other alterations or alterations will be included within the scope of this document.

[0143] The foregoing specific descriptions have been described with reference to various embodiments. However, those skilled in the art will recognize that various modifications and changes can be made without departing from the scope of this disclosure. Therefore, considerations for this disclosure are to be illustrative rather than restrictive, and all such modifications are to be included within its scope. Similarly, advantages, other advantages, and solutions to problems with respect to various embodiments have been described above. However, benefits, advantages, solutions to problems, and any elements that produce these, or make them more explicit, should not be construed as critical, essential, or necessary. The term “comprising” and any other variations thereof as used herein are non-exclusive inclusion, meaning that a process, method, article, or apparatus that includes a list of elements includes not only those elements but also other elements not expressly listed or not part of the process, method, system, article, or apparatus. Furthermore, the term “coupled” and any other variations thereof as used herein refer to physical connections, electrical connections, magnetic connections, optical connections, communication connections, functional connections, and / or any other connections.

[0144] Those skilled in the art will recognize that many changes can be made to the details of the above embodiments without departing from the basic principles of the invention. Therefore, the scope of the invention should be determined according to the following claims.

Claims

1. A method for generating a cross-sectional image of the bottom of a basin, characterized in that, include: Obtain the three-dimensional volume data of the pelvic floor; The spatial position of the lower edge of the pubic symphysis, the two pubic branches, and the anorectal angle is obtained from the three-dimensional volume data by identifying at least a portion of them. Based on the spatial position of the lower margin of the pubic symphysis, the two pubic branches, and at least a portion of the anorectal angle in the three-dimensional volume data, one or more reference lines are obtained from a cross section of the three-dimensional volume data, the reference lines passing through the lower margin of the pubic symphysis and at least a portion of the anorectal angle; the cross section is displayed and the one or more reference lines are displayed in the cross section; The three-dimensional volume data is sectioned by a plane or curved surface passing through the reference line to obtain the minimum hiatus section of the levator ani muscle. The minimum hiatus section of the levator ani muscle is parallel to the line connecting two points on the two pubic rami that are symmetrical about the lower edge of the pubic symphysis. The minimum hiatus section of the levator ani muscle is imaged to obtain a two-dimensional image of the minimum hiatus section of the levator ani muscle. Based on the obtained two-dimensional cross-sectional image of the minimum opening of the levator ani muscle, the area of ​​the minimum opening of the levator ani muscle is measured. and / or The levator ani muscle and urethral orifice are detected in the two-dimensional levator ani minimum hiatus section image, and the distance from the urethral orifice to the leftmost levator ani muscle and / or the distance from the urethral orifice to the rightmost levator ani muscle are measured based on the detected levator ani muscle and urethral orifice. and / or The contour of the levator ani muscle or the levator ani muscle minimum opening is extracted from the two-dimensional cross-sectional image of the levator ani muscle, and the distance between the upper and lower diameters of the levator ani muscle and / or the distance between the left and right diameters of the levator ani muscle is calculated based on the contour.

2. The method as described in claim 1, characterized in that, Identifying the posterior inferior border of the pubic symphysis from the three-dimensional volume data includes: Calculate at least one feature index of the three-dimensional volume data, input the feature index into a pre-established model function that corresponds the feature index of the three-dimensional volume data to the posterior inferior border of the pubic symphysis, and obtain the corresponding posterior inferior border of the pubic symphysis; or The three-dimensional volume data is subjected to target detection or image segmentation. Morphological feature detection is performed on each region obtained after detection or segmentation to obtain multiple candidate regions. For each candidate region, it is determined that the candidate region is the posterior inferior margin of the pubic symphysis based on shape features and / or grayscale features. The candidate region with the highest probability is determined as the posterior inferior margin of the pubic symphysis.

3. The method as described in claim 1, characterized in that, Identifying at least a portion of the anorectal angle from the three-dimensional volume data includes: Calculate at least one feature index of the three-dimensional volume data, input the feature index into a pre-established model function that maps the feature indexes of the three-dimensional volume data to at least a portion of the anorectal angle, and obtain at least a portion of the corresponding anorectal angle; or The three-dimensional volume data is subjected to target detection or image segmentation. Morphological feature detection is performed on each region obtained after detection or segmentation to obtain multiple candidate regions. For each candidate region, the probability that the candidate region is at least a part of the anorectal angle is determined based on shape features and / or grayscale features. The candidate region with the highest probability is determined as at least a part of the anorectal angle.

4. The method as described in claim 1, characterized in that, Identifying at least a portion of the posteroinferior margin of the pubic symphysis, the two pubic rami, and the anorectal angle from the three-dimensional volume data, thereby obtaining the spatial location of the posteroinferior margin of the pubic symphysis, the two pubic rami, and at least a portion of the anorectal angle in the three-dimensional volume data, includes: The spatial positions of at least a portion of the posterior inferior border of the pubic symphysis, the two pubic rami, and the anorectal angle within the three-dimensional volume data can be directly identified from the three-dimensional volume data; or, Based on the prior positions of the posterior inferior margin of the pubic symphysis, the two pubic branches, and at least a portion of the anorectal angle in the three-dimensional volume data, at least one two-dimensional section is generated. From the two-dimensional section, at least a portion of the posterior inferior margin of the pubic symphysis, the two pubic branches, and the anorectal angle is identified, and the spatial positions of the posterior inferior margin of the pubic symphysis, the two pubic branches, and at least a portion of the anorectal angle in the three-dimensional volume data are obtained.

5. The method as described in claim 1, characterized in that, Based on the spatial positions of at least a portion of the lower edge of the pubic symphysis, the two pubic rami, and the anorectal angle in the three-dimensional volume data, one or more reference lines are obtained from a cross-section of the three-dimensional volume data, including: Obtain a sagittal section image of the levator ani muscle from the three-dimensional volume data, and identify at least a portion of the posteroinferior margin of the pubic symphysis, the two pubic branches, and the anorectal angle in the sagittal section image of the levator ani muscle; Based on the identified posterior inferior margin of the pubic symphysis, the two pubic branches, and at least a portion of the anorectal angle, one or more reference lines are obtained from the sagittal section image of the levator ani muscle.

6. The method as described in claim 5, characterized in that, The pubic bone based on the identification By combining the posterior inferior margin, the two pubic rami, and at least a portion of the anorectal angle, one or more reference lines are obtained from the sagittal section image of the levator ani muscle, including: In response to a user's operational command regarding at least a portion of the identified posteroinferior border of the pubic symphysis, the two pubic rami, and the anorectal angle, one or more reference lines are acquired from the sagittal section image of the levator ani muscle according to the user's operational command; and / or Based on the identified lower posterior margin of the pubic symphysis, the two pubic branches, and at least a portion of the anorectal angle, one or more reference lines are automatically acquired.

7. An ultrasound imaging device, characterized in that... include: An ultrasonic probe is used to emit ultrasonic waves toward an object to be imaged in order to scan the object and to receive ultrasonic echoes returned from the object to be imaged. Transmit / receive circuitry is used to control the ultrasonic probe to transmit ultrasonic waves toward the object to be imaged and to receive the echoes of the ultrasonic waves. monitor; A processor is configured to perform 3D reconstruction based on the echo of the ultrasound to obtain three-dimensional volume data of the object to be imaged; the object to be imaged includes the pelvic floor. The processor is also used for: The spatial position of the lower edge of the pubic symphysis, the two pubic branches, and the anorectal angle is obtained from the three-dimensional volume data by identifying at least a portion of them. Based on the spatial position of the lower margin of the pubic symphysis, the two pubic branches, and at least a portion of the anorectal angle in the three-dimensional volume data, one or more reference lines are obtained from a cross section of the three-dimensional volume data, the reference lines passing through the lower margin of the pubic symphysis and at least a portion of the anorectal angle; the cross section is displayed on the display and the one or more reference lines are displayed in the cross section; The three-dimensional volume data is sectioned using a plane or curved surface passing through the reference line to obtain a section of the levator ani muscle minimum hiatus. The section of the levator ani muscle minimum hiatus is parallel to the line connecting two points symmetrical about the lower edge of the pubic symphysis on the two pubic rami. The section of the levator ani muscle minimum hiatus is imaged to obtain a two-dimensional image of the levator ani muscle minimum hiatus. The image of the levator ani muscle minimum hiatus is displayed on the display. Based on the obtained two-dimensional cross-sectional image of the minimum opening of the levator ani muscle, the area of ​​the minimum opening of the levator ani muscle is measured. and / or The levator ani muscle and urethral orifice are detected in the two-dimensional levator ani minimum hiatus section image, and the distance from the urethral orifice to the leftmost levator ani muscle and / or the distance from the urethral orifice to the rightmost levator ani muscle are measured based on the detected levator ani muscle and urethral orifice. and / or The contour of the levator ani muscle or the levator ani muscle minimum opening is extracted from the two-dimensional cross-sectional image of the levator ani muscle, and the distance between the upper and lower diameters of the levator ani muscle and / or the distance between the left and right diameters of the levator ani muscle is calculated based on the contour.

8. The device as described in claim 7, characterized in that, The processor is used to extract from the three Identifying at least a portion of the posteroinferior margin of the pubic symphysis, the two pubic rami, and the anorectal angle from the three-dimensional volumetric data, thereby obtaining the spatial positions of the posteroinferior margin of the pubic symphysis and at least a portion of the anorectal angle in the three-dimensional volumetric data, including: The processor is used to directly identify the spatial positions of at least a portion of the posteroinferior margin of the pubic symphysis, the two pubic rami, and the anorectal angle in the three-dimensional volume data; or, The processing is used to generate at least one two-dimensional cross-section based on the prior position of the posteroinferior margin of the pubic symphysis, the two pubic branches and at least a portion of the anorectal angle in the three-dimensional volume data, to identify the pubic symphysis, the two pubic branches and the anorectal angle from the two-dimensional cross-section, and to obtain the spatial position of the posteroinferior margin of the pubic symphysis, the two pubic branches and at least a portion of the anorectal angle in the three-dimensional volume data.

9. The device as described in claim 7, characterized in that, The processor is configured to acquire one or more reference lines from a cross section of the three-dimensional volume data based on the spatial positions of at least a portion of the lower edge of the pubic symphysis, the two pubic rami, and the anorectal angle in the three-dimensional volume data, including: The processor acquires a sagittal section image of the levator ani muscle from the three-dimensional volume data, and identifies at least a portion of the posteroinferior margin of the pubic symphysis, the two pubic branches, and the anorectal angle in the sagittal section image of the levator ani muscle; Based on the identified posterior inferior margin of the pubic symphysis, the two pubic branches, and at least a portion of the anorectal angle, one or more reference lines are obtained from the sagittal section image of the levator ani muscle.

10. The device as described in claim 9, characterized in that, The processor is used based on the Identify at least a portion of the posteroinferior border of the pubic symphysis, the two pubic rami, and the anorectal angle, and obtain one or more reference lines from the sagittal section image of the levator ani muscle, including: The processor is configured to respond to a user's operation command regarding at least a portion of the identified inferior border of the pubic symphysis, the two pubic branches, and the anorectal angle, and to acquire one or more reference lines from the sagittal section image of the levator ani muscle according to the user's operation command; and / or The processor automatically acquires one or more reference lines based on the identified posteroinferior margin of the pubic symphysis, the two pubic branches, and at least a portion of the anorectal angle.

11. An ultrasonic imaging device, characterized in that... include: Memory, used to store programs; A processor for executing a program stored in the memory to implement the method as described in any one of claims 1-6.

12. A computer-readable storage medium, characterized in that, Includes a program that can be executed by a processor to implement the method as described in any one of claims 1-6.

Images