Electronic colposcope image acquisition method and device

Abstract

The present invention provides an electronic colposcope image acquisition method and device, wherein the method includes: performing data acquisition on vaginal structure, constructing a vaginal model corresponding to the vaginal structure; in response to the vaginal structure being in a preset acquisition window, filling the vaginal model into a first display area located in an acquisition guide interface; sending the acquisition guide interface to a display helmet end for display, triggering a vaginal probe to acquire images, and determining whether the acquired acquired image includes the window content corresponding to the preset acquisition window; in response to the acquired image including the window content corresponding to the preset acquisition window, filling the acquired image into a second display area located in the acquisition guide interface, and generating a mapping contour corresponding to the acquired image in the first display area; performing image recognition based on the lesion area on the acquired image, and determining whether the acquired image has a lesion area based on the recognition result. The present invention at least improves the patient's cooperation and comfort.

Description

Technical Field

[0001] The present invention relates to data processing technology, and in particular to an electronic colposcope image acquisition method and device. Background Art

[0002] Among existing medical examination technologies, electronic colposcopy is a common gynecological examination method used to observe and evaluate the health of the female vagina.

[0003] However, in actual operation, medical staff usually need to look directly at the patient's vaginal structure in order to obtain clear images through the electronic colposcope. This direct visual contact and examination posture often make patients feel uncomfortable and embarrassed, especially when patients lack understanding of the examination process or are sensitive to privacy issues. This psychological burden may affect the patient's cooperation and even cause some patients to resist necessary medical examinations, thereby delaying the early detection and treatment of diseases.

[0004] Therefore, how to reduce the discomfort and psychological pressure of patients while ensuring the examination effect has become an urgent problem to be solved in the existing technology. Summary of the invention

[0005] Based on the above problems, the present invention is proposed to provide an electronic colposcopy image acquisition method and device that overcomes the above problems or at least partially solves the above problems.

[0006] According to one aspect of the present invention, there is provided an electronic colposcopy image acquisition method, comprising the following steps:

[0007] Collect multi-dimensional sensing data of the patient's vaginal structure and build a vaginal model based on the obtained vaginal twin data;

[0008] In response to the vagina structure being in the preset acquisition window, calling the acquisition guide interface, and filling the vagina model into a first display area of ​​the acquisition guide interface having the same specifications as the preset acquisition window;

[0009] The acquisition guidance interface is sent to the display helmet of the corresponding medical staff for display, the vaginal probe is triggered to acquire images, and based on the preset probe positioning strategy, it is determined whether the acquired image includes the window content of the corresponding preset acquisition window;

[0010] In response to the captured image including window content corresponding to the preset capture window, the captured image is filled into the second display area located in the capture guide interface, and a mapping contour corresponding to the captured image is generated in the first display area based on the captured image;

[0011] The collected image located in the second display area is subjected to image recognition based on the lesion area, and it is determined whether the collected image has the lesion area based on the recognition result.

[0012] Optionally, in the method according to the present invention, in response to the vagina structure being in a preset acquisition window, calling an acquisition guide interface, and filling the vagina model into a first display area of ​​the acquisition guide interface having the same specifications as the preset acquisition window, comprises:

[0013] In response to receiving the gravity collection value output by the gravity collection unit pre-set on the cabin bearing surface of the corresponding collection operation cabin at any collection time, a communication connection channel is established with the collection medical terminal, and a timed task is generated to maintain a preset determined time period;

[0014] In response to receiving a position determination signal sent by the collection medical terminal based on the communication connection channel within the preset determination period, it is determined that the patient is carried in the collection operation cabin and the vaginal structure of the patient is located in the middle of a preset collection window opened in the collection operation cabin;

[0015] The collection guide interface is called, and the vagina model is filled to the middle of the first display area having the same specifications as the preset collection window.

[0016] Optionally, in the method according to the present invention, triggering the vaginal probe to perform image acquisition, and determining whether the acquired acquired image includes window content corresponding to a preset acquisition window based on a preset probe positioning strategy, comprises:

[0017] Triggering the vaginal probe to perform image acquisition corresponding to an initial acquisition size, and determining each image pixel constituting the acquisition image based on the acquired acquisition image, wherein the initial acquisition size is larger than a preset window size corresponding to the preset acquisition window;

[0018] Perform pixel comparison based on pixel values ​​on each image pixel point and cabin pixel point of the corresponding acquisition operation cabin;

[0019] In response to determining that the collected image simultaneously has at least one image pixel point having the same pixel value as the cabin pixel point and at least one image pixel point having a different pixel value from the cabin pixel point based on the comparison result, all image pixel points having the same pixel value as the cabin pixel point are determined as reference pixel points, and all image pixel points having different pixel values ​​from the cabin pixel point are determined as noise pixel points;

[0020] The image pixels constituting the image frame of the acquired image are determined as a frame pixel group, and in response to the absence of any noise pixel in the frame pixel group, it is determined that the acquired image includes window content corresponding to the preset acquisition window.

[0021] Optionally, in the method according to the present invention, the method further comprises:

[0022] In response to the presence of any noise pixel in the frame pixel group, the image frame is disassembled to obtain left frame lines, right frame lines, upward frame lines and downward frame lines that respectively correspond to the left direction, the right direction, the upward direction and the downward direction of the image frame;

[0023] In response to any noise pixel being located at any frame line, the frame line is determined as a target frame line, and a preset voice table is retrieved, wherein the preset voice table includes left-row same-direction voice, right-row same-direction voice, upward same-direction voice, and downward same-direction voice corresponding to the left row direction, the right row direction, the upward direction, and the downward direction, respectively;

[0024] The preset voice table is traversed, the same-direction voice corresponding to the target frame line is determined as the target voice, and the voice unit pre-installed on the display helmet end is triggered to output the voice corresponding to the target voice.

[0025] Optionally, in the method according to the present invention, in response to the captured image including window content corresponding to a preset capture window, filling the captured image into a second display area located in the capture guide interface, and generating a mapping contour corresponding to the captured image in the first display area based on the captured image, comprising:

[0026] In response to the captured image including the window content corresponding to the preset capture window, the captured image is filled into the second display area of ​​the capture guide interface;

[0027] forming a transparent interactive layer in the second display area, wherein the transparent interactive layer includes an interactively permitted area corresponding to the window content and interactively prohibited content corresponding to other content except the window content;

[0028] In response to the display helmet end interacting with any noise pixel point located in the allowed interaction area, the noise pixel point is determined as an enlargement reference point, and an enlargement multiple input by the display helmet end based on the enlargement reference point is obtained;

[0029] The collected image is enlarged by a corresponding enlargement factor with the enlarged reference point as the center, and the updated image is cropped based on the image frame to obtain an updated collected image, wherein the updated collected image does not contain any cabin pixel point;

[0030] The window acquisition unit is triggered to acquire an image of the preset acquisition window to obtain a window image corresponding to the preset acquisition window;

[0031] Comparing the updated collected image with the window image, and determining a collection position of the updated collected image in the window image based on the comparison result;

[0032] A corresponding mapping position is determined in the first display area based on the acquisition position, and a mapping outline having a dotted line characteristic corresponding to the acquisition image is generated based on the mapping position.

[0033] Optionally, in the method according to the present invention, in response to the display helmet end interacting with any noise pixel point located in the allowed interaction area, determining the noise pixel point as an enlargement reference point, and obtaining the enlargement multiple input by the display helmet end based on the enlargement reference point, comprises:

[0034] In response to the display helmet end performing an interaction with any noise pixel point located in the allowed interaction area for a preset interaction time, the noise pixel point is determined as an enlarged reference point, and an image coordinate system is established based on the collected image;

[0035] Acquire, based on the image coordinate system, an enlarged coordinate point corresponding to the enlarged reference point and each image coordinate point corresponding to each image pixel point located in the frame pixel group;

[0036] Acquire each coordinate distance between the enlarged coordinate point and each image coordinate point, and determine the image coordinate point corresponding to the smallest coordinate distance as the closest coordinate point based on each coordinate distance;

[0037] Establishing a coordinate connection line connecting the enlarged coordinate point and the very close coordinate point, and determining a reference coordinate point corresponding to a reference pixel point located on the coordinate connection line and having an adjacent relationship with the noise pixel point based on the image coordinate system;

[0038] Based on the coordinate connection line, the coordinate distance between the reference coordinate point and the enlarged coordinate point is determined as the enlarged distance, the coordinate distance between the closest coordinate point and the enlarged coordinate point is determined as the reference distance, and the distance multiple between the enlarged distance and the reference distance is determined;

[0039] Creating a numerical interval with the distance ratio as a starting value and a preset ratio as an ending value, and sending the numerical interval to the display helmet end;

[0040] In response to the display helmet end interacting with any numerical value in the numerical range, the numerical value is determined as the enlargement multiple input by the display helmet end based on the enlargement reference point.

[0041] Optionally, in the method according to the present invention, the method further comprises:

[0042] In response to determining that a lesion region exists in the acquired image based on the recognition result, obtaining a region contour corresponding to the lesion region, and comparing the region contour with an image frame corresponding to the acquired image;

[0043] In response to determining that the region outline has an overlapping portion with an image frame corresponding to the acquired image based on the comparison result, obtaining an image center point corresponding to the acquired image, and determining a moving indication line from the image center point to the overlapping portion;

[0044] Acquire a contour center point corresponding to the mapping contour, and generate a mapping indication line in the first display area with the contour center point as a starting point and having the same indication direction as the moving indication line; or,

[0045] In response to the region contour and the image frame corresponding to the acquired image having no overlapping portion, a lesion model corresponding to the lesion region is generated in the vaginal model.

[0046] Optionally, in the method according to the present invention, determining a moving indication line pointing from the center point of the image to the overlapping portion comprises:

[0047] Taking the center point of the image as a starting point, determining image division lines of a corresponding preset number of divisions extending toward the image frame, and dividing the image frame based on the image division lines to obtain divided frames with the same division length;

[0048] Acquire each lesion pixel point constituting the lesion area based on the acquired image, and in response to any division frame including any lesion pixel point, aggregate the division frame into a movement indication group;

[0049] Acquire the number of pixels corresponding to each lesion pixel point respectively included in each division frame located in the movement indication group, and sort the division frames located in the movement indication group from most to least based on the number of pixels to obtain a division sequence;

[0050] Taking the center point of the image as a starting point, determining each moving indication line pointing to each dividing frame line located in the moving indication group, and performing a gradual adjustment on the thickness specification and the length specification of each moving indication line corresponding to the region sequence.

[0051] Optionally, in the method according to the present invention, the method further comprises:

[0052] In response to receiving a tissue sampling signal sent by the display helmet, performing image recognition based on the sampling instrument on the collected image located in the second display area;

[0053] In response to determining that the end of the corresponding sampling instrument contacts the vaginal structure based on the recognition result, determining the contact point corresponding to the end of the instrument based on the collected image;

[0054] Determine a sampling identification circle corresponding to a preset sampling radius based on the contact point, and generate a mapping identification circle corresponding to the sampling identification circle in a mapping contour located in the first display area;

[0055] In response to the mapping identification circle being located at the lesion model, the mapping identification circle is modified from an initial preset first color to a preset second color.

[0056] According to another aspect of the present invention, there is provided an electronic colposcopy image acquisition device, comprising:

[0057] A model building module is configured to collect multi-dimensional sensing data of the patient's vaginal structure and build a vaginal model based on the obtained vaginal twin data;

[0058] The model filling module is configured to call up a collection guide interface in response to the vagina structure being in a preset collection window, and fill the vagina model into a first display area of ​​the collection guide interface having the same specifications as the preset collection window;

[0059] The image acquisition module is configured to send the acquisition guidance interface to the display helmet of the corresponding medical staff for display, trigger the vaginal probe to acquire images, and determine whether the acquired image includes the window content of the corresponding preset acquisition window based on the preset probe positioning strategy;

[0060] an image mapping module, configured to fill the acquired image into the second display area located in the acquisition guide interface in response to the acquired image including window content corresponding to the preset acquisition window, and generate a mapping outline corresponding to the acquired image in the first display area based on the acquired image;

[0061] The lesion recognition module is configured to perform image recognition based on the lesion area on the collected image located in the second display area, and determine whether the collected image has a lesion area based on the recognition result.

[0062] According to the solution of the present invention, firstly, the present invention realizes the non-direct visual data collection of the patient's vaginal structure through multi-dimensional sensing means, avoiding the embarrassing situation that medical staff need to look directly at the patient's vaginal structure during traditional examinations. This not only greatly protects the patient's privacy, but also significantly reduces the patient's psychological burden during the examination process, and improves the patient's cooperation and comfort;

[0063] Secondly, the vaginal model constructed by combining vaginal twin data in this embodiment provides an intuitive and accurate vaginal structure reference for medical staff. During the examination process, when the patient's vaginal structure is in the preset acquisition window, this embodiment will automatically call the acquisition guidance interface and fill the vaginal model into the first display area to help medical staff obtain the complete morphology of the corresponding vaginal structure based on the vaginal model displayed in the first display area;

[0064] Furthermore, by sending the acquisition guidance interface to the display helmet end of the medical staff for display, remote and contactless image acquisition guidance is realized, which not only improves the professionalism and safety of the acquisition process, but also further reduces the psychological pressure of the patient. At the same time, the present invention can automatically determine whether the acquired image includes the content of the preset acquisition window based on the preset probe positioning strategy, thereby ensuring the integrity and accuracy of the acquisition, and after determining that the acquired image includes the window content corresponding to the preset acquisition window, the corresponding acquired image will be sent to the second display area, thereby helping the medical staff to realize the real-time positioning of the vaginal probe based on the first display area and the second display area, so as to perform a medical examination of the vaginal structure through the vaginal probe;

[0065] Finally, the present invention also has the function of intelligently identifying the lesion area. By performing image recognition on the collected image located in the second display area, the present invention can automatically determine whether there is a lesion area, providing corresponding assistance for the diagnosis of medical staff. This function not only improves the accuracy and efficiency of diagnosis, but also provides valuable time for early treatment of patients. BRIEF DESCRIPTION OF THE DRAWINGS

[0066] Figure 1 A flow chart of an electronic colposcopy image acquisition method according to an embodiment of the present invention is shown;

[0067] Figure 2 The structure diagram of the acquisition operation cabin in this embodiment is shown;

[0068] Figure 3 A structural block diagram of an electronic colposcopy image acquisition device according to another embodiment of the present invention is shown. DETAILED DESCRIPTION

[0069] The exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0070] To solve the problems in the above background technology, the inventor proposes the solution of the present invention. An embodiment of the present invention provides an electronic colposcopy image acquisition method, which can be executed in a computing device, wherein the computing device can be understood as a terminal with data processing function.

[0071] Figure 1 FIG. 4 is a flow chart showing a method for collecting electronic colposcopy images according to an embodiment of the present invention. Figure 1 As shown, the electronic colposcopy image acquisition method proposed in this embodiment begins with step S101. In step S101, the following contents are included:

[0072] Multi-dimensional sensing data is collected on the patient's vaginal structure, and a vaginal model is constructed based on the obtained vaginal twin data.

[0073] For example, in this embodiment, in order to perform a corresponding medical diagnosis on the patient's vaginal structure, a corresponding vaginal model may be first constructed based on the patient's vaginal structure for real-time display, wherein the corresponding creation process may be described as follows:

[0074] First, data on the patient's vaginal structure can be collected through a multi-dimensional sensing device, wherein the multi-dimensional sensing device includes but is not limited to an optical sensor, a pressure sensor, a temperature sensor, and a deformation sensor, which are used to obtain the geometric shape, tissue elasticity, temperature distribution, and dynamic change data of the vaginal structure;

[0075] Next, the data collected by the multidimensional sensing device can be preprocessed, including data filtering, denoising and normalization, to eliminate noise and errors in the collection process and obtain high-quality vaginal twin data;

[0076] Next, based on the vaginal twin data obtained above, a vaginal model corresponding to the vaginal structure is constructed using a three-dimensional modeling algorithm, wherein the three-dimensional modeling algorithm includes but is not limited to finite element analysis, mesh generation, and surface fitting technology to achieve accurate geometric reconstruction and physical property simulation of the vaginal structure;

[0077] Finally, the obtained vaginal model can be optimized accordingly, which can include model accuracy verification and parameter adjustment. By comparing the actual measurement data with the model output data, the model parameters can be adjusted to improve the accuracy and reliability of the model.

[0078] In step S102, the following contents are included:

[0079] In response to the vagina structure being in the preset acquisition window, the acquisition guide interface is called, and the vagina model is filled into a first display area of ​​the acquisition guide interface having the same specifications as the preset acquisition window.

[0080] For example, in this embodiment, after completing the construction of the corresponding vaginal model, it is necessary to make the patient's vaginal structure located in the preset acquisition window. Here, it should be noted that the preset acquisition window can be understood as a corresponding window that can expose the vaginal structure, so that in the subsequent operation process, the vaginal structure can be correspondingly imaged based on the preset acquisition window to determine whether there is a corresponding lesion area in the vaginal structure; further, after determining that the vaginal structure is in the preset acquisition window, the pre-stored acquisition guide interface can be retrieved from the server, and the corresponding vaginal model can be filled into the first display area of ​​the acquisition guide interface that has the same specifications as the preset acquisition window; it should be noted that the first display area having the same specifications as the preset acquisition window can be understood as having the same aspect ratio, so that the vaginal model located in the first display area can form a corresponding mapping relationship with the vaginal structure located in the preset acquisition window.

[0081] Furthermore, in this embodiment, the above-mentioned “in response to the vagina structure being in the preset acquisition window, calling the acquisition guide interface, and filling the vagina model into the first display area of ​​the acquisition guide interface having the same specifications as the preset acquisition window” may also include the following steps:

[0082] In response to receiving the gravity collection value output by the gravity collection unit pre-set on the cabin bearing surface of the corresponding collection operation cabin at any collection time, a communication connection channel is established with the collection medical terminal, and a timed task is generated to maintain a preset determined time period;

[0083] In response to receiving a position determination signal sent by the collection medical terminal based on the communication connection channel within the preset determination period, it is determined that the patient is carried in the collection operation cabin and the vaginal structure of the patient is located in the middle of a preset collection window opened in the collection operation cabin;

[0084] The collection guide interface is called, and the vagina model is filled to the middle of the first display area having the same specifications as the preset collection window.

[0085] For example, in this embodiment, in order to determine whether the vaginal structure is in the corresponding preset acquisition window, the following method steps may be used:

[0086] First, at any collection time, when receiving the gravity collection value output by the gravity collection unit set on the cabin bearing surface of the collection operation cabin, the server automatically triggers the establishment of a communication connection channel with the collection medical terminal, and generates a timed task that maintains a preset duration; wherein the gravity collection unit is used to monitor the gravity changes on the cabin bearing surface in real time to determine whether the patient has been correctly carried in the collection operation cabin. Here, if Figure 2As shown, the collection operation cabin can be understood as being used to carry the patient, and the corresponding collection operation cabin is also provided with a corresponding preset collection window, so that when the patient is carried in the collection operation cabin, the vaginal structure to be examined can be exposed through the preset collection window, so as to improve the privacy of the corresponding patient;

[0087] Then, within the preset time period, if a location determination signal sent by the acquisition medical terminal is received through the communication connection channel, it means that the medical staff using the acquisition medical terminal has confirmed that the patient is carried in the acquisition operation cabin and the patient's vaginal structure is located in the middle of the preset acquisition window of the acquisition operation cabin. Here, the acquisition medical terminal can be understood as a terminal with data processing function used by the medical staff, such as a mobile phone or a computer;

[0088] Finally, after completing the determination of the patient's position, the server can call up the pre-stored acquisition guidance interface and map the pre-constructed vaginal model to the middle of the first display area of ​​the acquisition guidance interface that has the same specifications as the preset acquisition window, thereby enabling the vaginal model located in the first display area to form a corresponding mapping relationship with the vaginal structure located in the preset acquisition window, so that when medical staff conduct corresponding inspections on the vaginal structure later, they can perform corresponding mapping displays on the corresponding vaginal model based on the inspection results, thereby achieving the corresponding positioning effect.

[0089] In step S103, the following contents are included:

[0090] The acquisition guidance interface is sent to the display helmet of the corresponding medical staff for display, the vaginal probe is triggered to acquire images, and based on the preset probe positioning strategy, it is determined whether the acquired image includes the window content of the corresponding preset acquisition window.

[0091] For example, in this embodiment, after the vaginal model is filled into the first display area located at the collection guidance interface, the corresponding collection guidance interface can be sent to the medical staff for display. Here, in order to improve the privacy of the patient, the medical staff can obtain the corresponding collection guidance interface based on wearing the display helmet end. That is, after the medical staff completes wearing the display helmet end, the medical staff can only view the corresponding screen displayed by the corresponding display helmet end, and cannot view other screens accordingly. Here, the display helmet end can be a VR helmet or an AR helmet.

[0092] After the display of the acquisition guidance interface is completed, the vaginal probe (that is, the electronic colposcope mentioned in the background technology) used by the medical staff can be triggered to acquire images. Here, the vaginal probe can be understood as a device unit with an image acquisition function, which can be installed on a control arm extending upward from the acquisition operation cabin (in order to ensure the neatness of the diagram, Figure 2The corresponding control arm and the vaginal probe mounted on the control arm are not shown in the figure). The control arm can move horizontally under the control of the medical staff, and the control arm has a preset distance from the preset collection window.

[0093] It should be noted that, in this embodiment, the medical staff can output corresponding control signals through the handheld controller to control the control arm to move horizontally, so that the vaginal probe can move along with the movement of the control arm;

[0094] In this embodiment, since medical staff can only view the acquisition guidance interface when wearing the corresponding display helmet end and cannot view the specific position of the vaginal probe, it is necessary to determine whether the acquired image obtained through image acquisition by the vaginal probe includes the window content of the corresponding preset acquisition window based on the corresponding preset probe positioning strategy; that is, when the acquired image includes the window content of the corresponding preset acquisition window, it means that the corresponding vaginal probe is in the position for performing corresponding image acquisition on the preset acquisition window, so as to determine the accuracy and real-time performance of the operation process and ensure that the acquired image can meet the corresponding inspection standards.

[0095] Furthermore, in this embodiment, the above-mentioned “triggering the vaginal probe to perform image acquisition, and determining whether the acquired image includes the window content corresponding to the preset acquisition window based on the preset probe positioning strategy” may also include the following steps:

[0096] Triggering the vaginal probe to perform image acquisition corresponding to an initial acquisition size, and determining each image pixel constituting the acquisition image based on the acquired acquisition image, wherein the initial acquisition size is larger than a preset window size corresponding to the preset acquisition window;

[0097] Perform pixel comparison based on pixel values ​​on each image pixel point and cabin pixel point of the corresponding acquisition operation cabin;

[0098] In response to determining that the collected image simultaneously has at least one image pixel point having the same pixel value as the cabin pixel point and at least one image pixel point having a different pixel value from the cabin pixel point based on the comparison result, all image pixel points having the same pixel value as the cabin pixel point are determined as reference pixel points, and all image pixel points having different pixel values ​​from the cabin pixel point are determined as noise pixel points;

[0099] The image pixels constituting the image frame of the acquired image are determined as a frame pixel group, and in response to the absence of any noise pixel in the frame pixel group, it is determined that the acquired image includes window content corresponding to the preset acquisition window.

[0100] For example, in this embodiment, whether the acquired image includes the window content of the corresponding preset acquisition window can be determined based on the corresponding preset probe positioning strategy by the following content:

[0101] First, after the acquisition guidance interface is sent to the display helmet for corresponding display, the vaginal probe used by the medical staff can be triggered to perform corresponding image acquisition, wherein the acquisition process follows the preset initial acquisition size, which is set larger than the preset window size of the preset acquisition window to be matched later, so as to ensure that the acquired image information is rich and comprehensive enough;

[0102] Subsequently, the acquired image is carefully analyzed to determine each image pixel that constitutes the acquired image. This step is the basis of image processing and aims to provide an accurate data basis for subsequent pixel comparison.

[0103] Next, each image pixel of the collected image is compared with the cabin pixel of the corresponding collection operation cabin one by one, and the comparison is based on the pixel values ​​of the two. This process is intended to identify the similar parts and the different parts of the collected image and the collection operation cabin. Here, it should be noted that the corresponding collection operation cabin can be composed of cabin pixel points with the same pixel value.

[0104] Based on the above comparison results, the present embodiment further makes a judgment: if the collected image simultaneously has at least one image pixel point having the same pixel value as the cabin pixel point and at least one image pixel point having a different pixel value, it means that at least part of the content of the corresponding preset collection window appears in the collected image, therefore, the image pixel points having the same corresponding pixel values ​​can be further classified as reference pixel points, and the image pixel points having different corresponding pixel values ​​can be classified as noise pixel points;

[0105] Finally, in order to determine whether the acquired image completely includes all the contents of the corresponding preset acquisition window, each image pixel point of the image frame constituting the edge of the acquired image can be defined as a frame pixel group, and check whether the group contains any noise pixel points. If there are no noise pixel points in the frame pixel group, it is determined that the acquired image has included all the contents of the corresponding preset acquisition window (that is, the aforementioned window content), thereby indicating that the corresponding vaginal probe is already in the position for performing corresponding image acquisition on the preset acquisition window, ensuring that the acquired image can meet the corresponding inspection standards.

[0106] It should be noted that, in the above process, when the captured image simultaneously contains at least one image pixel point having the same pixel value as the cabin pixel point, and at least one image pixel point having a different pixel value from the cabin pixel point, it indicates that the captured image includes at least part of the content of the corresponding preset capture window, so that further position determination can be performed based on the subsequently obtained frame pixel group; and when the captured image only contains image pixel points having the same pixel value as the cabin pixel point, it indicates that the captured image does not include at least part of the content of the corresponding preset capture window, that is, the vaginal probe only performs corresponding image capture on the acquisition operation cabin. At this time, the corresponding acquisition status can be fed back to the medical staff through the display helmet end, thereby prompting the medical staff to control the control arm to move horizontally to adjust the position of the corresponding vaginal probe.

[0107] Furthermore, in this embodiment, the following steps may also be included:

[0108] In response to the presence of any noise pixel in the frame pixel group, the image frame is disassembled to obtain left frame lines, right frame lines, upward frame lines and downward frame lines that respectively correspond to the left direction, the right direction, the upward direction and the downward direction of the image frame;

[0109] In response to any noise pixel being located at any frame line, the frame line is determined as a target frame line, and a preset voice table is retrieved, wherein the preset voice table includes left-row same-direction voice, right-row same-direction voice, upward same-direction voice, and downward same-direction voice corresponding to the left row direction, the right row direction, the upward direction, and the downward direction, respectively;

[0110] The preset voice table is traversed, the same-direction voice corresponding to the target frame line is determined as the target voice, and the voice unit pre-installed on the display helmet end is triggered to output the voice corresponding to the target voice.

[0111] For example, in this embodiment, based on the above content, if there is no noise pixel point in the frame pixel group, it is determined that the acquired image has included all the contents of the corresponding preset acquisition window (that is, the aforementioned window content), thereby indicating that the corresponding vaginal probe is already in a position for performing corresponding image acquisition on the preset acquisition window; and if there is any noise pixel point in the frame pixel group, it can be determined that the acquired image only includes part of the content of the corresponding preset acquisition window, and does not include all the content of the corresponding preset acquisition window. At this time, the medical staff can be instructed to adjust the position of the vaginal probe based on the following method steps:

[0112] First, when it is detected that there are noise pixels in the frame pixel group corresponding to the image frame, the present embodiment needs to disassemble the image frame to obtain its constituent elements, namely, the left frame line, the right frame line, the upward frame line and the downward frame line corresponding to the left direction, the right direction, the upward direction and the downward direction respectively;

[0113] Next, by performing corresponding checks on the obtained frame lines, if a noise pixel point is found on any of the frame lines, the frame line is determined to be a target frame line with a problem. Subsequently, the present embodiment calls a preset voice table, which has pre-set same-direction voice instructions corresponding to four basic directions (left, right, upward, and downward), specifically left-direction same-direction voice, right-direction same-direction voice, upward same-direction voice, and downward same-direction voice;

[0114] Finally, by traversing the preset voice table, the corresponding same-direction voice is found according to the specific direction of the target frame line, and it is determined as the target voice, thereby triggering the voice unit installed on the display helmet end to play sound information matching the target voice. In this way, medical staff can receive specific direction feedback about the problem of the image frame through hearing, and take corresponding adjustment measures for the vaginal probe.

[0115] For example, when there are corresponding noise pixels in the left-direction frame line, it indicates that there may be content of the corresponding preset acquisition window on the left side of the left-direction frame line. Therefore, the voice unit may be triggered to output the left-line same-direction voice, so that after receiving the corresponding left-line same-direction voice, the medical staff can make the corresponding left-direction position adjustment to the vaginal probe until it can be ensured that the left-direction frame line of the acquisition image acquired again by the vaginal probe after the position adjustment does not have the corresponding noise pixels; and after completing the position adjustment of the vaginal probe based on all the frame lines with noise pixels, it can be ensured that the frame pixel value of the acquisition image obtained by the subsequent image acquisition of the corresponding vaginal probe does not have any noise pixels, that is, the acquisition image already contains all the content of the corresponding preset acquisition window, thereby realizing effective monitoring of the integrity of image acquisition based on the preset acquisition window, and providing intuitive and accurate position adjustment prompts for medical staff through voice feedback.

[0116] In step S104, the following contents are included:

[0117] In response to the captured image including window content corresponding to the preset capture window, the captured image is filled into the second display area located in the capture guide interface, and a mapping outline corresponding to the captured image is generated in the first display area based on the captured image.

[0118] For example, in this embodiment, when the acquired image includes the window content corresponding to the preset acquisition window, the acquired image can be filled into the second display area located in the acquisition guide interface to ensure that the medical staff can intuitively see the acquired image on the acquisition guide interface, so as to facilitate the medical staff to perform subsequent confirmation or adjustment operations; in addition, a corresponding mapping contour can be generated in the first display area according to the acquired image, where "mapping contour" can refer to a corresponding contour line or boundary box drawn in the first display area according to the shape, size and position information of the acquired image; since the first display area has the same specifications as the preset acquisition window, and there is a vaginal model corresponding to the vaginal structure in the first display area, the generation of the mapping contour is intended to provide a visual guide or reference for the medical staff, so that the medical staff can determine the image position of the currently acquired acquired image in the first display area to achieve the positioning of the acquired image, so that the medical staff can adjust the position of the vaginal probe according to the mapping contour to complete the medical examination of the entire vaginal structure.

[0119] Further, in this embodiment, the above-mentioned “in response to the captured image including the window content corresponding to the preset capture window, filling the captured image into the second display area located in the capture guide interface, and generating a mapping contour corresponding to the captured image in the first display area based on the captured image” may also include the following steps:

[0120] In response to the captured image including the window content corresponding to the preset capture window, the captured image is filled into the second display area of ​​the capture guide interface;

[0121] forming a transparent interactive layer in the second display area, wherein the transparent interactive layer includes an interactively permitted area corresponding to the window content and interactively prohibited content corresponding to other content except the window content;

[0122] In response to the display helmet end interacting with any noise pixel point located in the allowed interaction area, the noise pixel point is determined as an enlargement reference point, and an enlargement multiple input by the display helmet end based on the enlargement reference point is obtained;

[0123] The collected image is enlarged by a corresponding enlargement factor with the enlarged reference point as the center, and the updated image is cropped based on the image frame to obtain an updated collected image, wherein the updated collected image does not contain any cabin pixel point;

[0124] The window acquisition unit is triggered to acquire an image of the preset acquisition window to obtain a window image corresponding to the preset acquisition window;

[0125] Comparing the updated collected image with the window image, and determining a collection position of the updated collected image in the window image based on the comparison result;

[0126] A corresponding mapping position is determined in the first display area based on the acquisition position, and a mapping outline having a dotted line characteristic corresponding to the acquisition image is generated based on the mapping position.

[0127] For example, in this embodiment, the generation process of the corresponding mapping contour may be described as follows:

[0128] First, when it is detected that the collected image contains the window content corresponding to the preset collection window, the collected image can be filled into the second display area located on the collection guide interface, thereby ensuring that the medical staff can intuitively see the collected image on the collection vagina interface through the display helmet end, where the first display area and the second display area can be arranged horizontally;

[0129] Subsequently, a corresponding transparent interactive layer can be created on the second display area, wherein the transparent interactive layer contains two key areas: one is the allowed interactive area, which covers the window content of the corresponding preset acquisition window to allow medical staff to perform interactive operations; the other is the prohibited interactive area, which covers all other content except the window content, prohibiting medical staff from interacting in these areas; this ensures the accuracy of user operations and avoids unnecessary misoperations;

[0130] Then, the medical staff can interact with any noise pixel point in the allowed interaction area through the display helmet end to mark the noise pixel point as the expansion reference point, and at the same time receive the expansion multiple input by the medical staff through the display helmet end, so as to further expand the collected image by a corresponding multiple with the expansion reference point as the center, and perform image cropping based on the image frame on the obtained updated image to obtain the updated collected image. It is worth noting that the updated collected image no longer contains any cabin pixel points;

[0131] Then, in order to determine the acquisition position of the acquisition image corresponding to the preset acquisition window, the window acquisition unit can be triggered to perform a new round of image acquisition on the preset acquisition window, and a window image that accurately corresponds to the preset acquisition window is obtained, and the updated acquisition image is compared and analyzed with the window image, and the acquisition position of the updated acquisition image in the window image is determined based on the comparison result; here, the window acquisition unit can be installed in advance, for example, it can be installed on the ceiling of the corresponding medical room, or a fixed arm extending upward can be set on the acquisition operation cabin, and the window acquisition unit can be fixed on the fixed arm;

[0132] Finally, based on the foregoing, it can be known that the first display area has the same specifications as the preset acquisition window, and the corresponding first display area is filled with a vaginal model corresponding to the vaginal structure. Therefore, the corresponding mapping position is found in the first display area based on the acquired acquisition position, and a mapping contour with a dotted line characteristic is generated based on the mapping position to achieve the positioning function for the acquired image.

[0133] Furthermore, in this embodiment, the above-mentioned “in response to the display helmet end interacting with any noise pixel point located in the allowed interaction area, determining the noise pixel point as an enlargement reference point, and obtaining the enlargement multiple input by the display helmet end based on the enlargement reference point” may also include the following steps:

[0134] In response to the display helmet end performing an interaction with any noise pixel point located in the allowed interaction area for a preset interaction time, the noise pixel point is determined as an enlarged reference point, and an image coordinate system is established based on the collected image;

[0135] Acquire, based on the image coordinate system, an enlarged coordinate point corresponding to the enlarged reference point and each image coordinate point corresponding to each image pixel point located in the frame pixel group;

[0136] Acquire each coordinate distance between the enlarged coordinate point and each image coordinate point, and determine the image coordinate point corresponding to the smallest coordinate distance as the closest coordinate point based on each coordinate distance;

[0137] Establishing a coordinate connection line connecting the enlarged coordinate point and the very close coordinate point, and determining a reference coordinate point corresponding to a reference pixel point located on the coordinate connection line and having an adjacent relationship with the noise pixel point based on the image coordinate system;

[0138] Based on the coordinate connection line, the coordinate distance between the reference coordinate point and the enlarged coordinate point is determined as the enlarged distance, the coordinate distance between the closest coordinate point and the enlarged coordinate point is determined as the reference distance, and the distance multiple between the enlarged distance and the reference distance is determined;

[0139] Creating a numerical interval with the distance ratio as a starting value and a preset ratio as an ending value, and sending the numerical interval to the display helmet end;

[0140] In response to the display helmet end interacting with any numerical value in the numerical range, the numerical value is determined as the enlargement multiple input by the display helmet end based on the enlargement reference point.

[0141] For example, in this embodiment, in order to ensure that when the acquired image is enlarged based on the enlargement reference point, the acquired updated image does not contain any cabin pixel point, the corresponding enlargement reference point and the enlargement multiple can be determined based on the following steps:

[0142] First, when medical staff interact with any noise pixel point in the allowed interaction area through the display helmet, and the interaction lasts for a preset interaction time, the noise pixel point can be automatically identified as an expanded reference point, and the corresponding image coordinate system can be constructed based on the currently collected image content. Here, setting the corresponding preset interaction time can prevent accidental touches and improve the accuracy of the interaction;

[0143] Subsequently, the enlarged coordinate point corresponding to the enlarged reference point is determined through the image coordinate system, and at the same time, the corresponding image coordinate point is determined for each image pixel point constituting the frame pixel group; this step ensures the accurate positioning of all relevant pixel points in the image coordinate system;

[0144] Next, the coordinate distance between the enlarged coordinate point and each image coordinate point can be calculated, and the image coordinate point with the smallest coordinate distance can be selected as the closest coordinate point;

[0145] Afterwards, a coordinate connection line connecting the magnified coordinate point and the extremely close coordinate point is drawn on the acquired image based on the image coordinate system, and along the coordinate connection line, a reference coordinate point corresponding to a reference pixel point having an adjacent relationship with the noise pixel point is determined. Here, the reference coordinate point can be understood as a critical point between a preset acquisition window and an acquisition operation cabin;

[0146] Next, based on the coordinate connection line, the enlarged distance between the reference coordinate point and the enlarged coordinate point, and the reference distance between the very close coordinate point and the enlarged coordinate point are respectively obtained, and the distance multiple between the enlarged distance and the reference distance is further determined;

[0147] Subsequently, a numerical interval can be created based on the obtained distance multiple, the starting value of the numerical interval can be the distance ratio calculated above, and the ending value is the preset ratio; this numerical interval provides an optional magnification range for the display helmet end, which can be explained that even if the medical staff selects the distance ratio as the magnification through the display helmet end, it can be ensured that the updated collected image does not contain any cabin pixel points;

[0148] Finally, when the display helmet interacts with any value within the numerical range, the value can be determined as the magnification factor input by the display helmet based on the magnification reference point, so that the magnification reference point and the magnification factor can be determined separately, so that the updated image can be enlarged and cropped accordingly based on the magnification reference point and the magnification factor in the subsequent process to obtain an updated captured image without any cabin pixel point.

[0149] It should be noted that after the corresponding updated collected image is obtained, the updated collected image will also be displayed in the second display area accordingly.

[0150] In step S105, the following contents are included:

[0151] The collected image located in the second display area is subjected to image recognition based on the lesion area, and it is determined whether the collected image has the lesion area based on the recognition result.

[0152] For example, in this embodiment, after completing the filling of the vaginal model of the first display area and the filling of the corresponding captured image of the second display area, the medical staff can determine the positional relationship between the corresponding vaginal probe and the vaginal structure based on the first display area, so as to control the vaginal probe to perform multi-directional image capture of the vaginal structure through continuous movement, and update the captured image located in the second display area in real time based on the image capture, so as to perform a medical examination of the vaginal structure based on the captured image. Here, in order to improve the efficiency and accuracy of the corresponding medical examination, corresponding image recognition can be introduced to determine whether there is a corresponding lesion area in the captured image.

[0153] It should be noted that, in this embodiment, the image recognition model can be constructed based on a neural network learning model or a machine learning model. Since the establishment and training process of the corresponding image recognition model are existing technologies, this embodiment does not elaborate on this process.

[0154] Furthermore, in this embodiment, the following steps may also be included:

[0155] In response to determining that a lesion region exists in the acquired image based on the recognition result, obtaining a region contour corresponding to the lesion region, and comparing the region contour with an image frame corresponding to the acquired image;

[0156] In response to determining that the region outline has an overlapping portion with an image frame corresponding to the acquired image based on the comparison result, obtaining an image center point corresponding to the acquired image, and determining a moving indication line from the image center point to the overlapping portion;

[0157] Acquire a contour center point corresponding to the mapping contour, and generate a mapping indication line in the first display area with the contour center point as a starting point and having the same indication direction as the moving indication line; or,

[0158] In response to the region contour and the image frame corresponding to the acquired image having no overlapping portion, a lesion model corresponding to the lesion region is generated in the vaginal model.

[0159] For example, in this embodiment, when it is determined based on the recognition result that a corresponding lesion area exists in the acquired image, in order to improve the visual identification of the lesion area, the following method steps may be taken:

[0160] First, it should be noted that, since the acquisition image displayed in the second display area is obtained based on the magnification input by the medical staff, the acquisition image cannot include all the contents of the corresponding vaginal structure. Based on this, when the corresponding lesion area appears in the acquisition image, it is necessary to determine whether the corresponding acquisition image includes all the contents of the lesion area. If not, the medical staff can be prompted accordingly to help them find other parts of the corresponding lesion area that are not included in the acquisition image by moving the vaginal probe. Therefore, when it is determined that there is a lesion area in the acquisition image, it is necessary to obtain the regional contour corresponding to the lesion area and compare the regional contour with the image frame of the corresponding acquisition image.

[0161] Next, this embodiment will respond according to the following two situations:

[0162] First, when it is determined based on the comparison result that the regional contour and the image frame have an overlapping part, it means that the acquired image may not include all the contents of the lesion area, and the remaining contents of the corresponding lesion area not included in the acquired image may be located in the area outside the overlapping part. Therefore, the image center point corresponding to the acquired image can be obtained, and a moving indication line pointing from the image center point to the overlapping part can be further determined; after the acquisition of the moving indication direction is completed, the contour center point of the corresponding mapping contour can be obtained in the first display area, and a mapping indication line with the contour center point as the starting point and the same indication direction as the moving indication line can be generated in the first display area, thereby giving corresponding movement instructions to medical staff, so that the medical staff can move the vaginal probe accordingly based on the moving indication line to help complete the entire acquisition of the lesion area;

[0163] Secondly, when it is determined based on the comparison results that the regional outline and the image frame have no overlapping parts, it means that the acquired image may have included all the contents of the lesion area. At this time, in order to improve the visual identification of the lesion area, a corresponding lesion model can be generated in the vaginal model based on the lesion area.

[0164] Furthermore, in this embodiment, the above-mentioned “determining a moving indication line pointing from the center point of the image to the overlapping portion” includes:

[0165] Taking the center point of the image as a starting point, determining image division lines of a corresponding preset number of divisions extending toward the image frame, and dividing the image frame based on the image division lines to obtain divided frames with the same division length;

[0166] Acquire each lesion pixel point constituting the lesion area based on the acquired image, and in response to any division frame including any lesion pixel point, aggregate the division frame into a movement indication group;

[0167] Acquire the number of pixels corresponding to each lesion pixel point respectively included in each division frame located in the movement indication group, and sort the division frames located in the movement indication group from most to least based on the number of pixels to obtain a division sequence;

[0168] Taking the center point of the image as a starting point, determining each moving indication line pointing to each dividing frame line located in the moving indication group, and performing a gradual adjustment on the thickness specification and the length specification of each moving indication line corresponding to the region sequence.

[0169] For example, in this embodiment, the determination of the moving indication line may be implemented based on the following process:

[0170] First, by taking the center point of the image as the starting point, the image division lines corresponding to the preset number of divisions can be determined. Each image division line starts from the center point of the image and extends in different directions toward the image frame of the corresponding acquired image. The image frame of the corresponding acquired image can be divided into a number of division frames using the corresponding image division lines, wherein:

[0171] Next, based on the acquired image obtained above, the lesion pixels constituting the lesion area are obtained, and the positions of the lesion pixels are further compared with each division frame. When it is determined that any division frame includes at least one lesion pixel, the division frame can be aggregated into a pre-created movement indication group;

[0172] Next, it can be explained that when the number of lesion pixels included in any division frame is greater, it means that the area outside the division frame may contain a larger area of ​​content corresponding to the lesion area. Therefore, the division frames can be sorted from the most to the least by obtaining the number of pixels corresponding to the lesion pixels included in each division frame in the moving indication group, and the corresponding division sequence can be obtained;

[0173] Finally, the center point of the image can be used as the starting point to determine the moving indication lines pointing to the dividing frame lines in the moving indication group, and the corresponding thickness specifications and length specifications can be gradually adjusted based on the division sequence obtained above, so as to visually reflect the importance of the indications corresponding to different moving indication lines, thereby improving the indication effect for medical staff.

[0174] For example, the more pixels a lesion pixel point includes in any division frame, the closer its position in the division sequence is, and the thicker and longer the generated moving indicator line should be, so as to have a better indication effect, to remind medical staff that they should preferentially control the movement of the vaginal probe toward the moving indicator line, so as to improve the inspection efficiency of the corresponding medical examination.

[0175] Here, the above-mentioned lesion area may include different types of lesions corresponding to the vaginal structure, and this embodiment does not limit the specific types of the lesion area.

[0176] Furthermore, in this embodiment, after completing the medical examination of the patient's vaginal structure, in order to perform further pathological analysis on the corresponding vaginal structure, the medical staff generally uses a corresponding sampling instrument to perform tissue sampling, and the medical staff may perform multiple samplings when performing tissue sampling to improve the accuracy of the corresponding analysis. In order to prevent the medical staff from repeatedly sampling the same location, in this embodiment, visual sampling marking of the corresponding tissue sampling can be performed based on the following method steps:

[0177] In response to receiving a tissue sampling signal sent by the display helmet, performing image recognition based on the sampling instrument on the collected image located in the second display area;

[0178] In response to determining that the end of the corresponding sampling instrument contacts the vaginal structure based on the recognition result, determining the contact point corresponding to the end of the instrument based on the collected image;

[0179] Determine a sampling identification circle corresponding to a preset sampling radius based on the contact point, and generate a mapping identification circle corresponding to the sampling identification circle in a mapping contour located in the first display area;

[0180] In response to the mapping identification circle being located at the lesion model, the mapping identification circle is modified from an initial preset first color to a preset second color.

[0181] For example, in this embodiment, when the medical staff needs to perform tissue sampling of the vaginal structure based on the acquired image, the medical staff can send a corresponding tissue sampling signal to the server based on the display helmet they are wearing, so that the server will perform image recognition based on the sampling instrument on the acquired image located in the second display area to determine whether the sampling instrument appears in the acquired image; when it is determined based on the recognition result that the sampling instrument appears in the acquired image, it can be determined that the medical staff has used the corresponding sampling instrument to reach the corresponding preset acquisition window to prepare for tissue sampling of the vaginal structure; further, when it is determined based on the acquired image that the instrument end of the sampling instrument has contacted the vaginal structure, it indicates that the sampling instrument has begun to sample the vaginal structure. At this time, it can be determined based on the sampling image whether the sampling instrument is in contact with the instrument. The contact point corresponding to the end, here, the contact point can also be understood as the sampling position of the sampling instrument for tissue sampling; because medical staff generally use a smaller sampling area when performing tissue sampling to minimize the degree of tissue damage to the patient, therefore, a sampling identification circle corresponding to a preset sampling radius can be determined based on the contact point, so as to complete the visual identification of this tissue sampling based on the sampling identification circle; further, in order to enable medical staff to know the sampling position corresponding to the sampling identification circle, a mapping identification circle corresponding to the sampling identification circle can be generated in the mapping contour located in the first display area, so that medical staff can determine the sampling position corresponding to this tissue sampling based on the viewing of the vaginal model located in the first display area, thereby improving the corresponding sampling identification.

[0182] Furthermore, since medical staff may sample based on the lesion area or other areas except the lesion area during the corresponding tissue sampling process, in order to distinguish the corresponding sampling situations, when the mapping identification circle is located in the lesion model, the corresponding mapping identification circle can be changed from the initial preset first color to the preset second color.

[0183] It should be noted that, in this embodiment, the above-mentioned sampling instrument may include medical surgical forceps, and the instrument end of the corresponding sampling instrument may be a forceps head or a forceps tail; and the corresponding preset sampling radius may be set based on the sampling experience of medical staff, and this embodiment does not make any specific limitations on this.

[0184] Furthermore, in this embodiment, the specific process of generating a mapping identification circle corresponding to the sampling identification circle can be based on obtaining the corresponding position of the corresponding sampling identification circle in the acquired image, and based on the position, generating a corresponding mapping identification circle in the mapping contour corresponding to the acquired image.

[0185] In summary, firstly, this embodiment realizes the non-direct visual data collection of the patient's vaginal structure through multi-dimensional sensing means, avoiding the embarrassing situation that medical staff need to look directly at the patient's vaginal structure during traditional examinations. This not only greatly protects the patient's privacy, but also significantly reduces the patient's psychological burden during the examination process, and improves the patient's cooperation and comfort;

[0186] Secondly, the vaginal model constructed by combining vaginal twin data in this embodiment provides an intuitive and accurate vaginal structure reference for medical staff. During the examination process, when the patient's vaginal structure is in the preset acquisition window, this embodiment will automatically call the acquisition guidance interface and fill the vaginal model into the first display area to help medical staff obtain the complete morphology of the corresponding vaginal structure based on the vaginal model displayed in the first display area;

[0187] Furthermore, by sending the acquisition guidance interface to the display helmet end of the medical staff for display, remote and contactless image acquisition guidance is realized, which not only improves the professionalism and safety of the acquisition process, but also further reduces the psychological pressure of the patient. At the same time, this embodiment can also automatically determine whether the acquired image includes the content of the preset acquisition window based on the preset probe positioning strategy, thereby ensuring the integrity and accuracy of the acquisition, and after determining that the acquired image includes the window content corresponding to the preset acquisition window, the corresponding acquired image will be sent to the second display area, thereby helping the medical staff to realize the real-time positioning of the vaginal probe based on the first display area and the second display area, so as to perform a medical examination of the vaginal structure through the vaginal probe;

[0188] Finally, this embodiment also has the function of intelligently identifying the lesion area. By performing image recognition on the collected image located in the second display area, this embodiment can automatically determine whether there is a lesion area, providing corresponding assistance for the diagnosis of medical staff. This function not only improves the accuracy and efficiency of diagnosis, but also provides valuable time for early treatment of patients.

[0189] Another embodiment of the present invention provides an electronic colposcopy image acquisition device, Figure 3 The corresponding device block diagram includes:

[0190] A model building module is configured to collect multi-dimensional sensing data of the patient's vaginal structure and build a vaginal model based on the obtained vaginal twin data;

[0191] The model filling module is configured to call up a collection guide interface in response to the vagina structure being in a preset collection window, and fill the vagina model into a first display area of ​​the collection guide interface having the same specifications as the preset collection window;

[0192] The image acquisition module is configured to send the acquisition guidance interface to the display helmet of the corresponding medical staff for display, trigger the vaginal probe to acquire images, and determine whether the acquired image includes the window content of the corresponding preset acquisition window based on the preset probe positioning strategy;

[0193] an image mapping module, configured to fill the acquired image into the second display area located in the acquisition guide interface in response to the acquired image including window content corresponding to the preset acquisition window, and generate a mapping outline corresponding to the acquired image in the first display area based on the acquired image;

[0194] The lesion recognition module is configured to perform image recognition based on the lesion area on the collected image located in the second display area, and determine whether the collected image has a lesion area based on the recognition result.

[0195] In the description provided herein, algorithms and displays are not inherently related to any particular computer, virtual system or other device. Various general purpose systems can also be used together with the examples of the present invention. According to the above description, it is obvious that the structure required for constructing such systems. In addition, the present invention is not directed to any specific programming language either. It should be understood that various programming languages ​​can be utilized to implement the content of the present invention described herein, and the above description of specific languages ​​is for the purpose of disclosing the preferred embodiment of the present invention.

[0196] In the description provided herein, a large number of specific details are described. However, it is understood that embodiments of the present invention can be practiced without these specific details. In some instances, well-known methods, structures and techniques are not shown in detail so as not to obscure the understanding of this description.

[0197] Similarly, it should be understood that in order to streamline the present disclosure and aid in understanding one or more of the various inventive aspects, in the above description of exemplary embodiments of the present invention, various features of the present invention are sometimes grouped together into a single embodiment, figure, or description thereof.

[0198] Those skilled in the art will appreciate that the modules or units or components of the devices of the examples disclosed herein may be arranged in the devices described in the embodiments, or alternatively may be located in one or more devices different from the devices of the examples. The modules of the foregoing examples may be combined into one module or may be divided into multiple submodules.

[0199] Those skilled in the art will appreciate that the modules of the device of the embodiment can be adaptively changed and arranged in one or more devices different from the embodiment. The modules or units or components of the embodiment can be combined into one module or unit or component, and further they can be divided into multiple sub-modules or sub-units or sub-components.

[0200] Furthermore, those skilled in the art will appreciate that although some embodiments described herein include certain features included in other embodiments but not other features, the combination of features from different embodiments is meant to be within the scope of the present invention and to form different embodiments.

[0201] In addition, some of the embodiments are described herein as methods or combinations of method elements that can be implemented by a processor of a computer system or by other devices that perform the functions. Therefore, a processor with necessary instructions for implementing the method or method elements forms a device for implementing the method or method elements. In addition, the elements described herein of the device embodiments are examples of devices for implementing the functions performed by the elements for the purpose of implementing the invention.

[0202] As used herein, unless otherwise specified, the use of ordinal numbers "first," "second," "third," etc. to describe common objects merely indicates that different instances of similar objects are involved, and is not intended to imply that the objects so described must have a given order in time, space, order, or in any other manner.

[0203] Although the present invention has been described according to a limited number of embodiments, it will be apparent to those skilled in the art, with the benefit of the above description, that other embodiments may be envisioned within the scope of the invention thus described. In addition, it should be noted that the language used in this specification is primarily selected for readability and instructional purposes, rather than for explaining or limiting the subject matter of the present invention.

Claims

1. An electronic colposcopy image acquisition method, characterized in that: include: Collect multi-dimensional sensing data of the patient's vaginal structure and build a vaginal model based on the obtained vaginal twin data; In response to receiving the gravity collection value output by the gravity collection unit pre-set on the cabin bearing surface of the corresponding collection operation cabin at any collection time, a communication connection channel is established with the collection medical terminal, and a timed task is generated to maintain a preset determined time period; In response to receiving a position determination signal sent by the collection medical terminal based on the communication connection channel within a preset period of time, it is determined that the patient is carried in the collection operation cabin and the vaginal structure of the patient is located in the middle of a preset collection window opened in the collection operation cabin; Calling up the acquisition guide interface, and filling the vaginal model to the middle of the first display area having the same specifications as the preset acquisition window; Sending the acquisition guidance interface to the display helmet end of the corresponding medical staff, triggering the vaginal probe to acquire an image of the corresponding initial acquisition size, and determining each image pixel based on the acquired image, wherein the initial acquisition size is larger than the preset window size of the corresponding preset acquisition window; Perform pixel comparison between each image pixel and the cabin pixel of the corresponding acquisition operation cabin; In response to determining that the captured image simultaneously has at least one image pixel point having the same pixel value as the cabin pixel point and at least one image pixel point having a different pixel value from the cabin pixel point based on the comparison result, all image pixel points having the same pixel value are determined as reference pixel points, and all image pixel points having different pixel values ​​are determined as noise pixel points; Determine each image pixel point of the image frame constituting the acquired image as a frame pixel group, and in response to the absence of any noise pixel point in the frame pixel group, determine that the acquired image includes window content corresponding to the preset acquisition window; In response to the captured image including window content corresponding to the preset capture window, the captured image is filled into the second display area located in the capture guide interface, and a mapping outline corresponding to the captured image is generated in the first display area; Image recognition is performed on the collected image located in the second display area, and it is determined whether there is a lesion area in the collected image based on the recognition result.

2. The electronic colposcopy image acquisition method according to claim 1, characterized in that: The method further comprises: In response to the presence of any noise pixel in the frame pixel group, the image frame is disassembled to obtain left frame lines, right frame lines, upward frame lines and downward frame lines that respectively correspond to the left direction, the right direction, the upward direction and the downward direction of the image frame; In response to any noise pixel being located at any frame line, the frame line is determined as a target frame line, and a preset voice table is retrieved, wherein the preset voice table includes left-row same-direction voice, right-row same-direction voice, upward same-direction voice, and downward same-direction voice corresponding to the left row direction, the right row direction, the upward direction, and the downward direction, respectively; The preset voice table is traversed, the same-direction voice corresponding to the target frame line is determined as the target voice, and the voice unit pre-installed on the display helmet end is triggered to output the voice corresponding to the target voice.

3. The electronic colposcopy image acquisition method according to claim 1, characterized in that: In response to the captured image including the window content corresponding to the preset capture window, the captured image is filled into the second display area located in the capture guide interface, and a mapping outline corresponding to the captured image is generated in the first display area, including: In response to the captured image including the window content corresponding to the preset capture window, the captured image is filled into the second display area of ​​the capture guide interface; forming a transparent interactive layer in the second display area, wherein the transparent interactive layer includes an interactively permitted area corresponding to the window content and interactively prohibited content corresponding to other content except the window content; In response to the display helmet end interacting with any noise pixel point located in the allowed interaction area, the noise pixel point is determined as an enlargement reference point, and an enlargement multiple input by the display helmet end based on the enlargement reference point is obtained; The collected image is enlarged by a corresponding enlargement factor with the enlarged reference point as the center, and the updated image is cropped based on the image frame to obtain an updated collected image, wherein the updated collected image does not contain any cabin pixel point; The window acquisition unit is triggered to acquire an image of the preset acquisition window to obtain a window image corresponding to the preset acquisition window; Comparing the updated collected image with the window image, and determining a collection position of the updated collected image in the window image based on the comparison result; A corresponding mapping position is determined in the first display area based on the acquisition position, and a mapping outline having a dotted line characteristic corresponding to the acquisition image is generated based on the mapping position.

4. The electronic colposcopy image acquisition method according to claim 3, characterized in that: In response to the display helmet end interacting with any noise pixel point located in the allowed interaction area, the noise pixel point is determined as an enlargement reference point, and an enlargement multiple input by the display helmet end based on the enlargement reference point is obtained, including: In response to the display helmet end performing an interaction with any noise pixel point located in the allowed interaction area for a preset interaction time, the noise pixel point is determined as an enlarged reference point, and an image coordinate system is established based on the collected image; Acquire, based on the image coordinate system, an enlarged coordinate point corresponding to the enlarged reference point and each image coordinate point corresponding to each image pixel point located in the frame pixel group; Acquire each coordinate distance between the enlarged coordinate point and each image coordinate point, and determine the image coordinate point corresponding to the smallest coordinate distance as the closest coordinate point based on each coordinate distance; Establishing a coordinate connection line connecting the enlarged coordinate point and the very close coordinate point, and determining a reference coordinate point corresponding to a reference pixel point located on the coordinate connection line and having an adjacent relationship with the noise pixel point based on the image coordinate system; Based on the coordinate connection line, the coordinate distance between the reference coordinate point and the enlarged coordinate point is determined as the enlarged distance, the coordinate distance between the closest coordinate point and the enlarged coordinate point is determined as the reference distance, and the distance multiple between the enlarged distance and the reference distance is determined; Creating a numerical interval with the distance ratio as a starting value and a preset ratio as an ending value, and sending the numerical interval to the display helmet end; In response to the display helmet end interacting with any numerical value in the numerical range, the numerical value is determined as the enlargement multiple input by the display helmet end based on the enlargement reference point.

5. The electronic colposcopy image acquisition method according to claim 1, characterized in that: The method further comprises: In response to determining that a lesion region exists in the acquired image based on the recognition result, obtaining a region contour corresponding to the lesion region, and comparing the region contour with an image frame corresponding to the acquired image; In response to determining that the region outline has an overlapping portion with an image frame corresponding to the acquired image based on the comparison result, obtaining an image center point corresponding to the acquired image, and determining a moving indication line from the image center point to the overlapping portion; Acquire a contour center point corresponding to the mapping contour, and generate a mapping indication line in the first display area with the contour center point as a starting point and having the same indication direction as the moving indication line; or, In response to the region contour and the image frame corresponding to the acquired image having no overlapping portion, a lesion model corresponding to the lesion region is generated in the vaginal model.

6. The electronic colposcopy image acquisition method according to claim 5, characterized in that: Determining a moving indication line from the center point of the image to the overlapping portion includes: Taking the center point of the image as a starting point, determining image division lines of a corresponding preset number of divisions extending toward the image frame, and dividing the image frame based on the image division lines to obtain divided frames with the same division length; Acquire the lesion pixels constituting the lesion area based on the acquired image, and in response to any division frame including any lesion pixel, aggregate the division frame into a movement indication group; Acquire the number of pixels corresponding to each lesion pixel point respectively included in each division frame located in the movement indication group, and sort the division frames located in the movement indication group from most to least based on the number of pixels to obtain a division sequence; Taking the center point of the image as a starting point, determining each moving indication line pointing to each dividing frame line located in the moving indication group, and performing a gradual adjustment on the thickness specification and the length specification of each moving indication line corresponding to the region sequence.

7. The electronic colposcopy image acquisition method according to claim 6, characterized in that: The method further comprises: In response to receiving a tissue sampling signal sent by the display helmet, performing image recognition based on the sampling instrument on the collected image located in the second display area; In response to determining that the end of the corresponding sampling instrument contacts the vaginal structure based on the recognition result, determining the contact point corresponding to the end of the instrument based on the collected image; Determine a sampling identification circle corresponding to a preset sampling radius based on the contact point, and generate a mapping identification circle corresponding to the sampling identification circle in a mapping contour located in the first display area; In response to the mapping identification circle being located at the lesion model, the mapping identification circle is modified from an initial preset first color to a preset second color.

8. An electronic colposcopy image acquisition device, characterized in that: include: A model building module is configured to collect multi-dimensional sensing data of the patient's vaginal structure and build a vaginal model based on the obtained vaginal twin data; The model filling module is configured to respond to receiving the gravity collection value output by the gravity collection unit pre-set on the cabin bearing surface of the corresponding collection operation cabin at any collection time, establish a communication connection channel with the collection medical terminal, and generate a timed task to maintain a preset determined time period; In response to receiving a position determination signal sent by the collection medical terminal based on the communication connection channel within a preset period of time, it is determined that the patient is carried in the collection operation cabin and the vaginal structure of the patient is located in the middle of a preset collection window opened in the collection operation cabin; Calling up the acquisition guide interface, and filling the vaginal model to the middle of the first display area having the same specifications as the preset acquisition window; The image acquisition module is configured to send the acquisition guidance interface to the display helmet of the corresponding medical staff for display, trigger the vaginal probe to acquire an image of the corresponding initial acquisition size, and determine each image pixel based on the acquired image, wherein the initial acquisition size is larger than the preset window size of the corresponding preset acquisition window; Perform pixel comparison between each image pixel and the cabin pixel of the corresponding acquisition operation cabin; In response to determining that the captured image simultaneously has at least one image pixel point having the same pixel value as the cabin pixel point and at least one image pixel point having a different pixel value from the cabin pixel point based on the comparison result, all image pixel points having the same pixel value are determined as reference pixel points, and all image pixel points having different pixel values ​​are determined as noise pixel points; Determine each image pixel point of the image frame constituting the acquired image as a frame pixel group, and in response to the absence of any noise pixel point in the frame pixel group, determine that the acquired image includes window content corresponding to the preset acquisition window; An image mapping module is configured to fill the acquired image into the second display area located in the acquisition guide interface in response to the acquired image including the window content corresponding to the preset acquisition window, and generate a mapping outline corresponding to the acquired image in the first display area; The lesion recognition module is configured to perform image recognition on the collected image located in the second display area, and determine whether there is a lesion area in the collected image based on the recognition result.

Images