A method and device for imaging and processing meibomian glands

Abstract

The present invention discloses a method and device for imaging and processing meibomian glands, the method comprising: first transmitting a first ultrasonic detection signal to the eyelid region, then receiving a first ultrasonic echo detection signal reflected from the eyelid region; preprocessing the first ultrasonic echo detection signal to obtain a first meibomian gland ultrasonic image and first meibomian gland three-dimensional coordinate data; preprocessing the first meibomian gland ultrasonic image to obtain a first image recognition result; based on the first image recognition result and the first meibomian gland three-dimensional coordinate data, transmitting and processing ultrasonic energy to the target meibomian gland. The present invention can use ultrasonic waves to complete the imaging detection and processing of meibomian glands. The imaging detection does not require medical staff to turn over the patient's eyelids, and the practical operability is strong; ultrasonic waves can accurately process the meibomian glands and reduce ultrasonic exposure of surrounding tissues.

Description

Technical Field

[0001] The present invention relates to the technical field of medical devices, and in particular to a method and a device for imaging and processing meibomian glands. Background Art

[0002] Dry eye is a chronic ocular surface disease caused by multiple factors. It is caused by abnormalities in the quality, quantity and dynamics of tears, which leads to tear film instability or imbalance of the ocular surface microenvironment. It may be accompanied by ocular surface inflammatory reactions, tissue damage and nerve abnormalities, causing a variety of ocular discomfort symptoms and / or visual dysfunction.

[0003] Meibomian glands are located in the tarsal plates of the upper and lower eyelids and are responsible for secreting lipids onto the corneal surface, lubricating the corneal surface during blinking and preventing tear evaporation. MGD (meibomian gland dysfunction) is a chronic, nonspecific inflammation of the meibomian glands, characterized by obstruction of the meibomian gland ducts or abnormal secretions of the meibomian glands, and is the main cause of evaporative dry eye.

[0004] In clinical practice, infrared imaging of the meibomian glands allows ophthalmologists to directly observe the structure of the meibomian glands and roughly assess the atrophy and loss of the glands. Given the importance of the meibomian glands in the pathogenesis of dry eye disease, there are already a variety of treatments for MGD, including eyelid hot compress massage and intense pulsed light irradiation.

[0005] In the process of implementing the technical method of the embodiment of the present invention, the inventor of this patent found at least the following technical problems in the prior art: The existing infrared imaging method of meibomian glands requires medical staff to turn the patient's upper and lower eyelids outwards separately during the operation, and take pictures after the eyelids are fixed in position. This process is accompanied by a certain degree of pain and causes discomfort to the patient. Especially for patients with deep eye sockets and younger age, this process is very challenging for the medical staff's skills and the patient's cooperation. Among the treatment methods for MGD, eyelid hot compress massage cannot accurately treat the meibomian glands, and the effect is unsatisfactory, while the charge for intense pulsed light irradiation is too expensive.

[0006] In summary, existing meibomian gland imaging and processing methods cannot meet actual usage needs. Summary of the invention

[0007] The embodiment of the present invention provides a method and device for meibomian gland imaging and processing, which solves the problem that the existing meibomian gland imaging and processing methods cannot meet actual use requirements.

[0008] An embodiment of the present invention provides, on one hand, a method for imaging and processing meibomian glands, comprising: Firstly, a first ultrasonic detection signal is transmitted to the eyelid region, and then a first ultrasonic echo detection signal reflected from the eyelid region is received; Preprocessing the first ultrasonic echo detection signal to obtain the first meibomian gland ultrasonic image and the first meibomian gland three-dimensional coordinate data; Preprocessing the first meibomian gland ultrasound image to obtain a first image recognition result; Based on the first image recognition result and the first meibomian gland three-dimensional coordinate data, ultrasonic energy is emitted to the target meibomian gland.

[0009] Optionally, after transmitting the treatment ultrasonic energy to the target meibomian gland, the method further comprises: Firstly, a second ultrasonic detection signal is transmitted to the eyelid region, and then a second ultrasonic echo detection signal reflected from the eyelid region is received; Preprocessing the second ultrasonic echo detection signal to obtain the second meibomian gland ultrasonic image and the second meibomian gland three-dimensional coordinate data; Preprocessing the second meibomian gland ultrasound image to obtain a second image recognition result; Based on the first image recognition result and the second image recognition result, ultrasonic treatment effect evaluation data is obtained.

[0010] Optionally, the method further includes: driving the ultrasound probe to move on the XY plane of the meibomian gland to acquire a series of two-dimensional ultrasound images of the meibomian gland; driving the ultrasound probe to move along the Z-axis of the meibomian gland to acquire a series of two-dimensional meibomian gland ultrasound images; According to the identified meibomian gland position, the ultrasound probe position, scanning depth, ultrasound frequency, and focus position are automatically adjusted to obtain the best meibomian gland ultrasound image.

[0011] Optionally, after acquiring a series of two-dimensional meibomian gland ultrasound images, the method further includes: Using a pixel-based reconstruction algorithm, the position and depth information of the pixel in each two-dimensional meibomian gland ultrasound image is obtained; Using a volume rendering algorithm, color and opacity are set for each voxel in the three-dimensional meibomian gland ultrasound data; Based on the two-dimensional meibomian gland image, a three-dimensional reconstruction algorithm is used to generate a three-dimensional ultrasound image of the first meibomian gland; According to the ultrasound imaging depth and the focal length of the ultrasound probe, the three-dimensional coordinate data of the first meibomian gland is constructed to determine the spatial relationship between the meibomian gland and the ultrasound probe.

[0012] Optionally, preprocessing the first meibomian gland ultrasound image to obtain a first image recognition result specifically includes: According to the ultrasound image of the first meibomian gland, the deep learning model based on the convolutional neural network segmented and identified the palpebral conjunctiva area as the ROI through the image segmentation algorithm; All meibomian glands are segmented and identified in the ROI, and the morphological and functional parameters of the segmented meibomian glands are quantitatively analyzed to derive a first image recognition result, which includes: meibomian gland missing area, meibomian gland missing rate, gland distortion, and deformation index.

[0013] Optionally, the emitting of the processed ultrasonic energy to the target meibomian glands is specifically: focusing the ultrasonic waves on the target meibomian glands by means of a curved acoustic lens or an ultrasonic phased array probe.

[0014] Optionally, emitting processing ultrasonic energy to the target meibomian gland based on the first image recognition result and the first meibomian gland three-dimensional coordinate data specifically includes: Based on the first image recognition result, adjusting the energy intensity of the processing ultrasonic energy, wherein the processing ultrasonic energy is low-intensity continuous ultrasonic energy, low-intensity pulsed ultrasonic energy or focused ultrasonic energy; Focusing the processed ultrasound energy at the current meibomian gland according to the three-dimensional coordinate data of the first meibomian gland; Move the treatment ultrasound energy from the current meibomian gland to the next meibomian gland.

[0015] Optionally, after transmitting the treatment ultrasonic energy to the target meibomian gland, the method further comprises: Real-time detection of the skin surface temperature of the eyelid area and output of the eyelid area temperature map; When the skin surface temperature exceeds a preset temperature threshold, the transmission of treatment ultrasound energy to the meibomian glands is stopped.

[0016] In another aspect, an embodiment of the present invention provides a method for treating meibomian glands, comprising: When receiving an instruction to image the meibomian glands without using ultrasound, selecting an output mode of processing ultrasound energy based on the acquired eyelid thickness data, and transmitting processing ultrasound energy to the target meibomian glands; or When an instruction is received not to image the meibomian glands using ultrasound, ultrasound energy is emitted to the target meibomian glands based on the last image recognition result and the last three-dimensional coordinate data of the meibomian glands.

[0017] In a third aspect, an embodiment of the present invention further provides a device for imaging and processing meibomian glands, comprising: An ultrasonic probe, used to first transmit a first ultrasonic detection signal to the eyelid region, and then receive a first ultrasonic echo detection signal reflected from the eyelid region; An ultrasonic imaging module, used for preprocessing the first ultrasonic echo detection signal to obtain the first meibomian gland ultrasonic image and the first meibomian gland three-dimensional coordinate data; A preprocessing calculation module, used for preprocessing the first meibomian gland ultrasonic image to obtain a first image recognition result; An ultrasonic processing module, configured to transmit ultrasonic energy to a target meibomian gland based on the first image recognition result and the three-dimensional coordinate data of the first meibomian gland; The ultrasonic processing module is a flexible or non-flexible ultrasonic processing module, and the ultrasonic probe and the ultrasonic processing module are the same or different ultrasonic instruments.

[0018] One or more technical solutions provided in the embodiments of the present invention have at least the following technical effects or advantages: The present invention innovatively uses ultrasound to complete the imaging detection and processing of the meibomian glands. After the meibomian glands are located and imaged by the ultrasound detection signal, the ultrasonic energy is transmitted to the meibomian glands in a targeted manner to stimulate the secretion of tears and meibomian gland lipids. During the entire process, medical staff no longer need to turn over the patient's eyelids. The conduction of ultrasound can reach the delicate tissue of the meibomian glands, with high imaging efficiency, no requirements for the medical staff's skills, and strong practical operability. In addition, ultrasound can accurately treat the meibomian glands by accurately locating the meibomian glands for targeted treatment, reducing the ultrasonic exposure of surrounding tissues, and is low in cost compared to intense pulsed light irradiation. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 A flowchart of a method for imaging and processing meibomian glands according to an embodiment of the present invention; Figure 2 This is a schematic diagram of a precision stepping motor driving an ultrasonic probe in one embodiment of the present invention; Figure 3 This is a schematic diagram of the connection between an ultrasonic probe and an infrared thermal imaging component in one embodiment of the present invention; In the figure: 61 - Z-axis precision stepper motor; 62 - XY plane precision stepper motor; 51 - eyelid; 52 - meibomian gland; 41 - display screen; 31 - processor; 21 - thermal imaging component; 11 - ultrasound probe. DETAILED DESCRIPTION

[0020] The embodiment of the present invention provides a method and device for meibomian gland imaging and processing, which solves the problem that the existing meibomian gland imaging and processing methods cannot meet actual use requirements.

[0021] In order to better understand the above-mentioned method and device for meibomian gland imaging and processing, the following will be described in detail in conjunction with the drawings and specific implementation methods of the specification. Obviously, the embodiments described in the present invention are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0022] Ultrasonic imaging is a non-invasive imaging technology widely used in medical monitoring. It does not require the patient's eyelids to be manually turned outward. Ultrasonic waves are transmitted and received by an ultrasonic probe. The reflection and scattering characteristics of different media to ultrasonic waves are utilized. The transmission of ultrasonic waves can reach the delicate tissue of the meibomian glands and generate images of the internal parts of the meibomian glands. Therefore, ultrasonic waves can be used for meibomian gland imaging.

[0023] In addition, in previous studies, it was found that ultrasound has heating and stimulating effects, and can be used to stimulate the secretion of tears and meibomian gland lipids in rabbits, so ultrasound can be used for meibomian gland treatment.

[0024] Please refer to the following Figure 1 , a method for meibomian gland imaging and processing in an embodiment of the present invention includes steps 101 to 104.

[0025] Step 101: first transmit a first ultrasonic detection signal to the eyelid area, and then receive a first ultrasonic echo detection signal reflected from the eyelid area.

[0026] In the specific implementation process, for example: when the patient is within the detection range, the ultrasonic probe transmits a first ultrasonic detection signal to the patient's eyelid area. After the first ultrasonic detection signal reaches the patient's eyelid area, it reaches the delicate tissue of the meibomian gland through ultrasonic conduction. The reflected first ultrasonic echo detection signal is received by the phased array element in the ultrasonic probe.

[0027] In order to capture the best meibomian gland ultrasound image and facilitate understanding of the actual situation of the meibomian gland, the method of the present invention also includes: driving the ultrasound probe to move on the XY plane of the meibomian gland to collect a series of two-dimensional meibomian gland ultrasound images; driving the ultrasound probe to move on the Z axis of the meibomian gland to collect a series of two-dimensional meibomian gland ultrasound images; and automatically adjusting the ultrasound probe position, scanning depth, ultrasound frequency, and focus position according to the identified meibomian gland position to obtain the best meibomian gland ultrasound image.

[0028] In the specific implementation process, for example: Figure 2As shown, a precision stepper motor is used to drive a mechanical device to move the ultrasonic probe for linear scanning. The XY plane precision stepper motor 62 drives the ultrasonic probe 11 to move on the XY plane of the meibomian gland 52 in the eyelid 51, and collects a series of two-dimensional meibomian gland ultrasound images to obtain a two-dimensional image sequence; the Z axis precision stepper motor 61 drives the ultrasonic probe 11 to move along the Z axis of the meibomian gland 52 in the eyelid 51, and collects a series of two-dimensional meibomian gland ultrasound images to obtain a two-dimensional image sequence. Through image processing techniques, such as image segmentation and feature extraction, the position of the meibomian gland 52 in the image is identified. According to the identified meibomian gland position, the position of the ultrasonic probe 11 is automatically adjusted to align it with the meibomian gland 52. By adjusting the scanning depth, it is ensured that the ultrasonic detection signal can penetrate to the depth where the meibomian gland 52 is located. By adjusting the ultrasonic frequency, it is adapted to the reflection and absorption characteristics of different tissues to the ultrasonic detection signal. By adjusting the focus position, it is ensured that the ultrasonic detection signal forms a clear image in the meibomian gland 52. Finally, the ultrasound probe 11 acquires a series of two-dimensional meibomian gland ultrasound images that are equally spaced and parallel to each other. The information of the two-dimensional meibomian gland ultrasound images includes the direction and relative position of the sequence.

[0029] Step 102: pre-processing the first ultrasonic echo detection signal to obtain the first meibomian gland ultrasonic image and the first meibomian gland three-dimensional coordinate data.

[0030] In order to generate three-dimensional meibomian gland ultrasound images and three-dimensional coordinate data of meibomian glands, so as to facilitate observation of the actual situation of the meibomian glands and determination of the actual position of the meibomian glands, after acquiring a series of two-dimensional meibomian gland ultrasound images, the method also includes: using a pixel-based reconstruction algorithm to obtain the position and depth information of the pixel points in each two-dimensional meibomian gland ultrasound image; using a volume rendering algorithm to set the color and opacity for each voxel in the three-dimensional meibomian gland ultrasound data; based on the two-dimensional meibomian gland image, using a three-dimensional reconstruction algorithm to generate a three-dimensional first meibomian gland ultrasound image; constructing the three-dimensional coordinate data of the first meibomian gland according to the ultrasound imaging depth and the focal length of the ultrasound probe, and determining the spatial relationship between the meibomian gland and the ultrasound probe.

[0031] In the specific implementation process, for example: the ultrasonic imaging module uses a pixel-based reconstruction algorithm, each two-dimensional meibomian gland ultrasound image is inserted into a predefined three-dimensional volume lattice, and each pixel point of the two-dimensional meibomian gland ultrasound image is traversed. According to the position and depth information of the pixel point in the image, the pixel value is assigned to one or more voxels in the three-dimensional volume lattice to form three-dimensional meibomian gland ultrasound data.

[0032] The ultrasound imaging module uses a volume rendering algorithm to set color and opacity values ​​for each voxel in the 3D meibomian gland ultrasound data for 3D visualization. The specific implementation process is as follows: Based on the value of the voxel, a color and opacity value are assigned to each voxel, and the reflection and transmission of each voxel for light are calculated to simulate the behavior of light when passing through the volume data.

[0033] The ultrasound imaging module performs two-dimensional image conversion and three-dimensional imaging image reconstruction to convert ultrasound data into a form that is easier to understand and visualize.

[0034] The ultrasonic imaging module uses the ultrasonic imaging depth and the focal length of the ultrasonic probe to calculate the actual spatial position corresponding to each pixel in the ultrasonic image. By identifying and analyzing the features of the meibomian glands (such as shape, boundaries, etc.) in the ultrasonic image of the first meibomian gland, combined with the ultrasonic imaging depth and the focal length of the ultrasonic probe, the three-dimensional coordinates of the meibomian glands are constructed. Based on the three-dimensional coordinate data of the first meibomian gland and the position information of the ultrasonic probe, the spatial relationship between the meibomian gland and the ultrasonic probe, such as distance, direction, etc., is determined.

[0035] Step 103: pre-process the first meibomian gland ultrasound image to obtain a first image recognition result.

[0036] Step 103 specifically includes the following steps: based on the first meibomian gland ultrasound image, a deep learning model based on a convolutional neural network uses an image segmentation algorithm to segment and identify the palpebral conjunctiva area as a ROI (Region of Interest); segment and identify all meibomian glands in the ROI, perform quantitative analysis on the morphological and functional parameters of the segmented meibomian glands, and derive a first image recognition result.

[0037] Among them, the first image recognition results include: meibomian gland loss area, meibomian gland loss rate, gland distortion, and deformation index. The meibomian gland loss area and meibomian gland loss rate are used to reflect the proportion of the gland in the ROI area, indicating the different severity of MGD. Gland distortion is used to evaluate the degree of gland distortion and is a clear sign of moderate to severe MGD. The deformation index is used to measure the degree of change in gland diameter, which is related to the obstruction of the meibomian gland opening, local dilation caused by meibum accumulation, and a small amount of local acinar atrophy.

[0038] Step 104: Based on the first image recognition result and the meibomian gland three-dimensional coordinate data, transmit processing ultrasonic energy to the target meibomian gland.

[0039] Step 104 specifically includes the following steps: adjusting the energy intensity of the processed ultrasonic energy based on the first image recognition result; focusing the processed ultrasonic energy on the current meibomian gland according to the three-dimensional coordinate data of the meibomian gland; and moving the processed ultrasonic energy from the current meibomian gland to the next meibomian gland.

[0040] It should be noted that the processing ultrasonic energy can be low-intensity continuous ultrasound, or low-intensity pulsed ultrasound, focused ultrasound, etc. The processing ultrasonic energy is used to process the meibomian glands in two ways: contact and non-contact. When contact treatment is selected, a flexible ultrasonic processing module is used to fit closely with the eyelid skin, which can reduce the loss of ultrasonic energy. The flexible ultrasonic processing module generates an ultrasonic processing signal and radiates it to the meibomian gland tissue deep in the eyelid, thereby generating mechanical and thermal effects. When non-contact treatment is selected, the non-flexible ultrasonic processing module moves between the upper and lower eyelid areas without contacting the patient's eyelid skin, and focuses on treating one meibomian gland each time it moves. One meibomian gland refers to a meibomian gland or a group of meibomian glands on the same plane, because ultrasonic focusing is not limited to a specific meibomian gland, and ultrasonic focusing can achieve point, line, and surface focusing, which can target a certain meibomian gland of MGD, and can also treat the meibomian gland group on the same plane at the same time. In practical applications, it can be selected as needed. The target meibomian gland refers to a meibomian gland or a group of meibomian glands on the same plane that needs to be treated. The current meibomian gland site refers to a meibomian gland being processed or a group of meibomian glands on the same plane. The next meibomian gland site refers to a meibomian gland waiting to be processed or a group of meibomian glands on the same plane.

[0041] Transmitting and processing ultrasonic energy to the target meibomian gland, specifically: focusing the ultrasonic wave on the target meibomian gland through a curved acoustic lens or an ultrasonic phased array probe.

[0042] In the specific implementation process, for example: when non-contact processing is adopted, a curved acoustic lens can be added to the non-flexible ultrasonic probe to focus the ultrasound on the target meibomian gland through geometric physical focusing. An ultrasonic phased array probe can also be used to control the different delay times of the pulses emitted by each element in the transducer array, thereby changing the phase relationship when the sound wave reaches a certain point in the object, and then the ultrasound focus can be focused on the target meibomian gland through the phased array focusing method.

[0043] In order to evaluate the treatment effect of ultrasonic energy on the meibomian glands by comparing the meibomian glands before and after treatment, after step 104, it also includes: first transmitting a second ultrasonic detection signal to the eyelid area, and then receiving a second ultrasonic echo detection signal reflected from the eyelid area; preprocessing the second ultrasonic echo detection signal to obtain a second meibomian gland ultrasonic image and second meibomian gland three-dimensional coordinate data; preprocessing the second meibomian gland ultrasonic image to obtain a second image recognition result; based on the first image recognition result and the second image recognition result, obtaining ultrasonic treatment effect evaluation data. Since the processing of the second ultrasonic detection signal, the second ultrasonic echo detection signal, the second meibomian gland ultrasonic image, and the second meibomian gland three-dimensional coordinate data are similar to steps 101, 102, and 103, they are not repeated for the sake of brevity. By comparing the meibomian glands in the first image recognition result and the second image recognition result, ultrasonic treatment effect evaluation data can be obtained.

[0044] In order to intuitively demonstrate the ultrasonic treatment effect and avoid the adverse effects of high temperature, after step 104, it also includes: real-time detection of the skin surface temperature of the eyelid area, and output of the eyelid area temperature map; when the skin surface temperature exceeds the preset temperature threshold, stop emitting ultrasonic energy for treatment to the meibomian glands.

[0045] In the specific implementation process, for example: Figure 3 As shown, the infrared thermal imaging component 21 detects the temperature of the upper and lower eyelid skin area, and the processor 31 superimposes the collected temperature data on the upper and lower eyelid skin area to form an eyelid area temperature map, and images it on the display screen 41. According to the position information and temperature information on the eyelid area temperature map, the ultrasonic treatment effect can be intuitively displayed. The preset temperature threshold is set to 42.5°C. During the ultrasonic treatment process, if the skin surface temperature of the eyelid exceeds 42.5°C, the processor 31 transmits a control instruction to the ultrasonic probe 11 to stop transmitting and processing ultrasonic energy to avoid excessively high skin surface temperature. It should be noted that the temperature of the upper and lower eyelid skin area can be detected by a non-contact temperature sensor to determine whether the skin surface temperature exceeds the preset temperature threshold.

[0046] In order to directly perform targeted treatment on the meibomian glands according to different eyelid thicknesses without using ultrasound to image the meibomian glands, so as to facilitate patients to select a suitable output mode for processing ultrasonic energy according to their own eyelid conditions and realize a personalized and targeted meibomian gland ultrasonic treatment plan, the present invention also provides a meibomian gland treatment method, comprising: upon receiving an instruction to image the meibomian glands without using ultrasound, based on the acquired eyelid thickness data, selecting an output mode for processing ultrasonic energy, and emitting the processed ultrasonic energy to the target meibomian glands.

[0047] In the specific implementation process, for example, the thickness of the eyelids of Chinese people is 2.5mm-5.5mm. Clinical research results show that the thickness of the eyelids will become thinner with age, and the position of the meibomian glands inside the eyelids will also change accordingly. In order to achieve the ultrasonic processing depth that can adapt to the meibomian glands in different positions, an age-eyelid thickness database is established in advance, and the eyelid thickness of the user group is divided into three levels: 2.5mm-3.5mm, 3.5mm-4.5mm, and 4.5mm-5.5mm. The corresponding three-level ultrasonic parameters are set to suit users with different eyelid thicknesses. Patients can test their eyelid thickness in the hospital, or they can determine their own eyelid thickness based on the age-eyelid thickness database, and then choose the output mode for processing ultrasonic energy. When the patient or medical staff triggers the corresponding button or option to choose not to use ultrasound to image the meibomian glands, an instruction is generated not to use ultrasound to image the meibomian glands. Based on the instruction not to use ultrasound to image the meibomian glands, according to the age data output by the patient or the medical staff, the corresponding eyelid thickness data is selected from the age-eyelid thickness database, and the ultrasonic energy output mode corresponding to the eyelid thickness data is selected to transmit the processing ultrasonic energy to the target meibomian gland. Or when the patient or the medical staff triggers the corresponding button or option to select not to use ultrasound to image the meibomian glands, an instruction not to use ultrasound to image the meibomian glands is generated. Based on the instruction not to use ultrasound to image the meibomian glands, according to the eyelid thickness data output by the patient or the medical staff, and the ultrasonic energy output mode corresponding to the eyelid thickness data is selected, the processing ultrasonic energy is transmitted to the target meibomian gland.

[0048] In order to facilitate the subsequent meibomian gland processing process, after the first meibomian gland imaging, it is not necessary to perform meibomian gland imaging before the subsequent meibomian gland processing to improve the efficiency of meibomian gland processing. The present invention also provides a meibomian gland processing method, including: when receiving an instruction to image the meibomian gland without using ultrasound, based on the last image recognition result and the last meibomian gland three-dimensional coordinate data, emitting processing ultrasonic energy to the target meibomian gland.

[0049] In the specific implementation process, for example: when the patient's meibomian glands need to be processed four times, after the meibomian glands are imaged using ultrasound for the first time, the image recognition results and the meibomian gland three-dimensional coordinate data are automatically saved. When the patient's meibomian glands are subsequently processed, the first image recognition results and the meibomian gland three-dimensional coordinate data are used to obtain the location of the meibomian glands, so that the target meibomian glands can be accurately processed.

[0050] Another embodiment of the present invention provides a device for imaging and processing meibomian glands, comprising: An ultrasonic probe, used to first transmit a first ultrasonic detection signal to the eyelid region, and then receive a first ultrasonic echo detection signal reflected from the eyelid region; An ultrasonic imaging module, used for preprocessing the first ultrasonic echo detection signal to obtain the first meibomian gland ultrasonic image and the first meibomian gland three-dimensional coordinate data; A preprocessing calculation module, used for preprocessing the first meibomian gland ultrasonic image to obtain a first image recognition result; An ultrasonic processing module, configured to transmit ultrasonic energy to a target meibomian gland based on the first image recognition result and the three-dimensional coordinate data of the first meibomian gland; The ultrasonic processing module is a flexible or non-flexible ultrasonic processing module, and the ultrasonic probe and the ultrasonic processing module are the same or different ultrasonic instruments.

[0051] When a flexible ultrasonic treatment module is used, the flexible ultrasonic treatment module fits closely with the eyelid skin, which can reduce the loss of ultrasonic energy. The flexible ultrasonic treatment module generates ultrasonic treatment signals and radiates them to the meibomian gland tissue deep in the eyelid, thereby generating mechanical benefits and thermal effects.

[0052] When a non-flexible ultrasonic treatment module is used, the non-flexible ultrasonic treatment module focuses the ultrasound waves at the target meibomian glands without contacting the patient's eyelid skin.

[0053] The ultrasonic probe and the ultrasonic processing module are the same or different ultrasonic instruments, that is, the ultrasonic probe and the ultrasonic processing module can share the same ultrasonic instrument or use different ultrasonic instruments separately to be applied to different usage scenarios.

[0054] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A method for imaging and processing meibomian glands, characterized in that: include: Firstly, a first ultrasonic detection signal is transmitted to the eyelid region, and then a first ultrasonic echo detection signal reflected from the eyelid region is received; Preprocessing the first ultrasonic echo detection signal to obtain the first meibomian gland ultrasonic image and the first meibomian gland three-dimensional coordinate data; Preprocessing the first meibomian gland ultrasound image to obtain a first image recognition result; Based on the first image recognition result and the first meibomian gland three-dimensional coordinate data, ultrasonic energy is emitted to the target meibomian gland.

2. The method according to claim 1, characterized in that After transmitting the treatment ultrasound energy to the target meibomian gland, the method further comprises: Firstly, a second ultrasonic detection signal is transmitted to the eyelid region, and then a second ultrasonic echo detection signal reflected from the eyelid region is received; Preprocessing the second ultrasonic echo detection signal to obtain the second meibomian gland ultrasonic image and the second meibomian gland three-dimensional coordinate data; Preprocessing the second meibomian gland ultrasound image to obtain a second image recognition result; Based on the first image recognition result and the second image recognition result, ultrasonic treatment effect evaluation data is obtained.

3. The method according to claim 1, characterized in that The method further comprises: driving the ultrasound probe to move on the XY plane of the meibomian gland to acquire a series of two-dimensional ultrasound images of the meibomian gland; driving the ultrasound probe to move along the Z-axis of the meibomian gland to acquire a series of two-dimensional meibomian gland ultrasound images; According to the identified meibomian gland position, the ultrasound probe position, scanning depth, ultrasound frequency, and focus position are automatically adjusted to obtain the best meibomian gland ultrasound image.

4. The method according to claim 3, characterized in that After acquiring a series of two-dimensional meibomian gland ultrasound images, the method further includes: Using a pixel-based reconstruction algorithm, the position and depth information of the pixel in each two-dimensional meibomian gland ultrasound image is obtained; Using a volume rendering algorithm, color and opacity are set for each voxel in the three-dimensional meibomian gland ultrasound data; Based on the two-dimensional meibomian gland image, a three-dimensional reconstruction algorithm is used to generate a three-dimensional ultrasound image of the first meibomian gland; According to the ultrasound imaging depth and the focal length of the ultrasound probe, the three-dimensional coordinate data of the first meibomian gland is constructed to determine the spatial relationship between the meibomian gland and the ultrasound probe.

5. The method according to claim 1, characterized in that The preprocessing of the first meibomian gland ultrasound image to obtain the first image recognition result specifically includes: According to the ultrasound image of the first meibomian gland, the deep learning model based on the convolutional neural network segmented and identified the palpebral conjunctiva area as the ROI through the image segmentation algorithm; All meibomian glands are segmented and identified in the ROI, and the morphological and functional parameters of the segmented meibomian glands are quantitatively analyzed to derive a first image recognition result, which includes: meibomian gland missing area, meibomian gland missing rate, gland distortion, and deformation index.

6. The method according to claim 1, characterized in that The emitting of ultrasonic energy to the target meibomian glands is specifically: focusing the ultrasonic waves on the target meibomian glands through a curved acoustic lens or an ultrasonic phased array probe.

7. The method according to claim 1, characterized in that The method of transmitting the processing ultrasonic energy to the target meibomian gland based on the first image recognition result and the first meibomian gland three-dimensional coordinate data specifically includes: Based on the first image recognition result, adjusting the energy intensity of the processing ultrasonic energy, wherein the processing ultrasonic energy is low-intensity continuous ultrasonic energy, low-intensity pulsed ultrasonic energy or focused ultrasonic energy; Focusing the processed ultrasound energy at the current meibomian gland according to the three-dimensional coordinate data of the first meibomian gland; Move the treatment ultrasound energy from the current meibomian gland to the next meibomian gland.

8. The method according to claim 1, characterized in that After transmitting the treatment ultrasound energy to the target meibomian gland, the method further comprises: Real-time detection of the skin surface temperature of the eyelid area and output of the eyelid area temperature map; When the skin surface temperature exceeds a preset temperature threshold, the transmission of treatment ultrasound energy to the meibomian glands is stopped.

9. A method for treating meibomian glands, characterized in that: include: When receiving an instruction to image the meibomian glands without using ultrasound, based on the acquired eyelid thickness data, an output mode of processing ultrasound energy is selected to transmit processing ultrasound energy to the target meibomian glands; or When an instruction is received not to image the meibomian glands using ultrasound, ultrasound energy is emitted to the target meibomian glands based on the last image recognition result and the last three-dimensional coordinate data of the meibomian glands.

10. A device for meibomian gland imaging and processing, characterized in that: include: An ultrasonic probe, used to first transmit a first ultrasonic detection signal to the eyelid region, and then receive a first ultrasonic echo detection signal reflected from the eyelid region; An ultrasonic imaging module, used for preprocessing the first ultrasonic echo detection signal to obtain the first meibomian gland ultrasonic image and the first meibomian gland three-dimensional coordinate data; A preprocessing calculation module, used for preprocessing the first meibomian gland ultrasonic image to obtain a first image recognition result; An ultrasonic processing module, configured to transmit ultrasonic energy to a target meibomian gland based on the first image recognition result and the three-dimensional coordinate data of the first meibomian gland; The ultrasonic processing module is a flexible or non-flexible ultrasonic processing module, and the ultrasonic probe and the ultrasonic processing module are the same or different ultrasonic instruments.

Images