Eyeball cornea treatment system based on near-infrared light recognition

Abstract

The invention relates to the technical field of ophthalmology medical treatment, in particular to an eyeball cornea treatment system based on near-infrared light recognition, which acquires optical information of eyeball cornea and choroid by acquiring near-infrared light signals, performs digital and filtering processing on original images and outputs processed image data. The choroidal defect or atrophy lesion point location is identified by combining edge detection and threshold segmentation, and the cornea thickness and curvature are measured at the same time to determine the energy intensity, irradiation time, wavelength and focusing position of the excimer light beam. According to the parameters, excimer light beams with the wavelength of 2,00048 nanometers are emitted to carry out irradiation treatment on a lesion area, the lesion positioning precision is improved, meanwhile, normal tissue structures are prevented from being damaged, and the safe and effective requirements of eye treatment are met.

Description

Technical Field

[0001] The present invention relates to the field of ophthalmic medical technology, and in particular to an eye cornea treatment system based on near-infrared light recognition. Background Art

[0002] Early diagnosis and treatment of corneal diseases are of great significance for maintaining visual health. Existing technologies mainly use direct naked eye observation and traditional white light inspection methods, as well as laser treatment devices that rely on rough parameter control. The image acquisition resolution is limited, the location of the lesion area is not accurate enough, and there is uncertainty in the adjustment of laser parameters, which can easily cause damage to normal tissues and reduce the safety and effectiveness of treatment. Summary of the invention

[0003] In view of the many problems existing in the above-mentioned prior art, the present invention provides an eye cornea treatment system based on near-infrared light recognition. The present invention uses near-infrared light to collect corneal and choroidal images of the eye, and uses edge detection and region segmentation technology to accurately locate the lesion area after digitization and filtering. The excimer beam emission parameters are determined based on the corneal thickness and curvature, and the lesion is accurately irradiated with a laser with a wavelength of 248 nanometers, thereby protecting normal tissues while ensuring the treatment effect and improving the safety of treatment.

[0004] like Figure 2 As shown, a corneal treatment system based on near-infrared light recognition includes: A near-infrared light collection module, used to obtain optical signals from the cornea and choroid of the eye and generate an initial image; An image processing module, used for digitizing and filtering the initial image and outputting processed image data; A lesion detection module, for identifying choroidal coloboma or atrophy lesion points from the processed image data and generating positioning information; A parameter determination module, used to determine the emission parameters of the excimer beam according to the corneal thickness and curvature and the choroidal image data, wherein the emission parameters include energy intensity, irradiation time, wavelength and focus position; The light beam emission module is used to output a quasi-molecular light beam with a wavelength of 248 nanometers according to the emission parameters to irradiate and treat the lesion site.

[0005] Preferably, the near-infrared light collection module acquires optical signals of the cornea and choroid of the eye in a frequency range of 26.565 to 27.405 MHz.

[0006] Preferably, the image processing module performs analog-to-digital conversion on the initial image and removes noise using frequency domain filtering to obtain processed image data.

[0007] Preferably, the lesion detection module extracts the eyeball tissue contour based on a gradient operator and uses a threshold segmentation method to identify the boundaries of choroidal coloboma or atrophy lesion points.

[0008] Preferably, the parameter determination module acquires corneal thickness and curvature data by emitting and receiving near-infrared light at at least two different angles and performing numerical inversion, and then determines the emission parameters of the excimer beam in combination with the choroidal image data.

[0009] Preferably, the energy intensity is graded and set based on the corneal thickness measurement value, and the ratio of the lesion area to the reference area is regulated in combination with the lesion depth and logarithmic function operation. When the lesion area is lower than the reference area, the logarithmic term is negative to reduce the energy output and the minimum energy threshold is used as the final output when the calculated value is lower than the minimum energy threshold. The irradiation time is segmented and set based on the degree of the lesion, the wavelength is selected based on the optical absorption coefficient of the choroid, and the focusing position is calibrated based on the coordinates of the lesion area.

[0010] Preferably, the excimer beam output by the beam emitting module has a power intensity not exceeding 1 mW / cm2 and can operate continuously for not less than one hour.

[0011] Preferably, the light beam emission module periodically monitors the choroidal reflection signal during the irradiation process and compares it with a power deviation threshold, and stops the light beam emission and outputs an alarm signal when the power deviation exceeds the threshold.

[0012] Preferably, the light beam emission module includes a short-circuit detection device, which immediately cuts off the power supply and records the fault information when the current exceeds a preset safety threshold.

[0013] Preferably, the image processing module and the lesion detection module transmit processed image data via a data bus, and the parameter determination module and the light beam emission module are coordinated and scheduled via control signals.

[0014] Compared with the prior art, the advantages and beneficial effects of the present invention are: High-quality images of the cornea and choroid of the eye are collected through near-infrared image recognition technology, and edge detection and region segmentation methods are used to accurately locate the lesion area. The excimer beam emission parameters are determined based on corneal thickness and curvature data to achieve precise irradiation of the lesion area, while effectively protecting normal tissues and significantly improving the safety and reliability of treatment. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 It is a schematic diagram of the near infrared image recognition principle in the present invention; Figure 2 It is a structural block diagram of the system of the present invention. DETAILED DESCRIPTION

[0016] Hereinafter, embodiments of the present disclosure will be described with reference to the accompanying drawings. However, it should be understood that these descriptions are exemplary only and are not intended to limit the scope of the present disclosure. In the following detailed description, for ease of explanation, many specific details are set forth to provide a comprehensive understanding of the embodiments of the present disclosure. However, it is obvious that one or more embodiments may also be implemented without these specific details. In addition, in the following description, descriptions of known structures and technologies are omitted to avoid unnecessary confusion of the concepts of the present disclosure.

[0017] The terms used herein are only for describing specific embodiments and are not intended to limit the present disclosure. The terms "include", "comprising", etc. used herein indicate the existence of the features, steps, operations and / or components, but do not exclude the existence or addition of one or more other features, steps, operations or components.

[0018] All terms (including technical and scientific terms) used herein have the meanings commonly understood by those skilled in the art unless otherwise defined. It should be noted that the terms used herein should be interpreted as having a meaning consistent with the context of this specification and should not be interpreted in an idealized or overly rigid manner.

[0019] Near-infrared image recognition technology is an inspection method that uses near-infrared technology to generate images and perform recognition. When the electrons in the near-infrared light emitter are accelerated or oscillated, continuous or intermittent near-infrared light will be generated in a directional manner. When near-infrared light of a specific wavelength is irradiated to the eyeball, it will penetrate transparent tissues such as the cornea and lens and reach the choroid of the fundus. The vascular network and other tissue structures of the choroid of the fundus will reflect, scatter or absorb these near-infrared lights. When encountering the choroid of the fundus of the eyeball, the carrier information carrying the refractive index information is reflected. Finally, the near-infrared light image formed by the amplification mechanism such as sign gain is obtained through mechanisms such as digital-to-analog conversion and modulation and demodulation, and the abnormal area is identified.

[0020] like Figure 1 As shown in the figure, near-infrared image recognition technology is an inspection method that uses near-infrared technology to generate images and perform recognition. When the electrons in the near-infrared light emitter are accelerated or oscillated, continuous or intermittent near-infrared light will be generated in a directional manner. When near-infrared light of a specific wavelength is irradiated to the eyeball, it will penetrate transparent tissues such as the cornea and lens and reach the choroid of the fundus. The vascular network and other tissue structures of the choroid of the fundus will reflect, scatter or absorb these near-infrared lights. When encountering the choroid of the fundus of the eyeball, the carrier information carrying the refractive index information is reflected. Finally, the near-infrared light image formed by the amplification mechanism such as the sign gain is matched through the mechanisms such as digital-to-analog conversion and modulation and demodulation, and the abnormal area is recognized.

[0021] In clinical practice, doctors use a near-infrared light emitting probe, place it in the patient's eye, and then emit near-infrared light. Near-infrared light will attenuate to varying degrees when passing through the retinal melanin epithelium of the human eye. Therefore, when it encounters the choroid of the eyeball with different thicknesses, it will reflect and return to the probe. The reflected signal received by the near-infrared light probe is recorded and converted into an image. By measuring the intensity and time difference of the near-infrared light reflection, different images can be generated. These images can be used to help doctors understand the patient's fundus choroid problems and help locate choroidal defects or atrophic lesions.

[0022] In the present invention, the steps of signal collection, digital conversion, filtering, image reconstruction and abnormal area positioning after near-infrared light penetrates the transparent tissue of the eye are used: First, near-infrared light irradiates the eye tissue through the transmitter, and the receiver captures the signal after penetration and converts it into an electrical signal (the working frequency band of near-infrared light recognition is shown in Table 1); then, the analog-to-digital converter (ADC) digitizes the analog signal to ensure that the sampling frequency and quantization accuracy meet the requirements. Then, the digital filtering technology is used to remove noise and improve the signal quality. The filtered signal uses the image reconstruction algorithm to generate a two-dimensional or three-dimensional image of the eye tissue. Finally, the abnormal area (such as corneal lesions or lens opacity) is located by edge detection, region segmentation, feature extraction and machine learning. First, the contour information of the eye structure is extracted using edge detection algorithms such as Canny and Sobel, and the image is divided into regions such as cornea and lens through threshold segmentation, region growing or deep learning segmentation network. Then, texture, shape, color and other features are extracted from the segmented area to distinguish between normal and abnormal areas. Then, the support vector machine, random forest or deep learning classifier is used to classify the area to determine whether there is an abnormality. Finally, the target detection algorithm is directly used to accurately locate the abnormal area. Provide accurate basis for the early diagnosis and treatment of ophthalmic diseases. This process combines the properties of near-infrared light and advanced signal processing technology to achieve non-invasive eye tissue detection and analysis.

[0023] Table 1 Excimer beam is a 248nm beam generated by the emission of photons when the molecules formed by the mixed gas formed by the combination of inert gas and halogen gas excited by the excimer beam emitter transition to their ground state. It has strong penetration, strong directionality and high output power. Excimer beam mainly irradiates biological tissues to produce benign biological stimulation and photochemical effects, stimulates the nerve receptors and peripheral blood vessels around the eyes, effectively improves blood circulation and nerve regulation around the fundus, improves blood circulation and metabolism of fundus tissues, restores the physiological regulation function of the fundus choroid, and achieves the purpose of correcting and alleviating hyperopia or astigmatism. The comparison of retinal vascular density is shown in Table 2.

[0024] Table 2 Among them, compared with the mild myopia group, P<0.05; compared with the moderate myopia group, P<0.05; compared with the high myopia group, P<0.05.

[0025] The present invention combines near-infrared light image recognition with excimer light beams to identify and treat a variety of eye diseases. Through near-infrared light image recognition technology, detailed data on the thickness, curvature, and choroid of the patient's cornea can be obtained. These data are used to determine the energy intensity, irradiation time, wavelength, and focusing position of the excimer light beam to ensure that the light beam can accurately act on the target area while avoiding damage to healthy tissue. The excimer light beam stimulates the nerve receptors and peripheral blood vessels around the eye, effectively improving blood circulation and nerve regulation around the eye, improving blood circulation and metabolism of eye tissues, and restoring the physiological regulation function of the fundus choroid. The high precision and controllability of the excimer light beam is combined with the precise navigation of near-infrared light image recognition to treat myopia and reduce the risk of complications such as dry eyes and glare. The details of controlling the excimer light beam are as follows: 1. Energy intensity control: adjust the beam energy intensity according to corneal thickness and lesion depth.

[0026]

[0027] in Refers to the reference area. When , it means that the lesion area is smaller than the reference value, and the system needs to reduce the energy intensity. Small lesions require more concentrated low-energy stimulation to avoid excessive treatment and damage to surrounding healthy tissues. The negative value is actually the reverse regulation of energy compensation. It is the depth-dominant term that determines the energy baseline value and reflects the core influence of lesion depth on energy penetration. It is an area correction item. Based on the depth, fine-tuning is performed according to the area of ​​the lesion (large areas require energy dispersion, small areas require energy concentration). Depth coefficient: Converts physical depth into energy demand, reflecting the optical properties of tissue. The effective treatment energy threshold at different depths is measured through in vitro tissue experiments, and the slope is obtained by fitting the data. . Depth: directly quantifies the vertical position of the lesion, which is the core input of energy regulation. The two form the linear body of the formula, ensuring that the energy increases monotonically with depth, providing a benchmark reference for the logarithmic term, and the staff can adjust Adapt to the light transmittance of different patient tissues to achieve personalized treatment.

[0028] Synergistic relationship with other parameters: (1) If Large (depth sensitive), then the area correction term It needs to be appropriately reduced to avoid excessive dependence of energy on area. (2) b affects the sensitivity of the area term and needs to be Calibrate together to ensure that depth and area are weighted appropriately.

[0029] generally , so a negative logarithmic term will reduce the total energy , which is consistent with the clinical logic that small lesion areas require low energy. ,ensure Monotonically increasing, in line with clinical logic. When the result is greater than 0, it means that the energy is increasing positively; when the result is less than 0, the energy is negatively fed back, thus satisfying .

[0030] Practical application: Considering It is impossible to output a negative value, so the present invention sets a minimum energy threshold to ensure that negative values ​​do not lead to unphysical energy output:

[0031] For example: Let the reference area , lesion area , :set up =5 mJ / mm 2 , then the logarithmic term contribution is 5×(−1)=−5 mJ / mm 2 , indicating that the energy needs to be reduced by 5 mJ / mm 2 To avoid overtreatment.

[0032] In the present invention, in order to ensure The unit is mJ / mm 2 and The unit is mm, then The units must be:

[0033] in, Indicates the energy density absorbed per unit volume of tissue (mJ / mm 3 ), reflecting the energy attenuation rate in the depth direction. The logarithmic term is dimensionless, so The unit needs to be Same, i.e. mJ / mm 2 .

[0034] 2. Irradiation time regulation: Adjust the irradiation time according to the type of lesion and nerve sensitivity.

[0035]

[0036] in is the irradiation time; is the nerve sensitivity coefficient; is the lesion type weight factor; is the beam energy intensity.

[0037] 3. Wavelength selection: Select the most effective wavelength based on the optical properties of the lesion area (such as absorption spectrum). For example, corneal lesions may be more sensitive to a certain wavelength of light, and the system will automatically select the best wavelength based on the patient's data.

[0038] 4. Focus position adjustment: adjust the focus point of the beam according to the location and depth of the lesion area. Dynamically calibrate the focus position of the beam through real-time image processing and patient data.

[0039] Near-infrared image recognition technology replaces traditional eye examination methods. The non-contact and non-invasive characteristics of near-infrared image recognition technology completely avoid direct contact with the patient's eyes, thereby reducing the patient's discomfort or potential infection risk caused by contact. Near-infrared image recognition technology can capture and analyze fundus choroidal images more quickly. Compared with traditional methods, the present invention can more quickly and accurately identify and locate choroidal defects or atrophic lesions.

[0040] The information of choroidal defect or atrophy lesion point located by near-infrared light image recognition, among which, near-infrared light image can effectively identify and locate the information of choroidal defect or atrophy lesion point through its deep tissue penetration ability and high contrast imaging characteristics, combined with image processing, feature extraction and machine learning technology. First, the image is pre-processed by denoising, contrast enhancement and registration, and then the choroid area is segmented by edge detection, and features such as texture, shape and intensity are extracted. Next, machine learning is used to identify the lesion area, and the lesion point is accurately located by edge detection, target detection algorithm or heat map generation technology. Finally, the bounding box and center point coordinates of the lesion area are finally output through morphological operation and connected region analysis optimization results, and the lesion point is transmitted to this product to analyze the lesion point, control the excimer beam device to produce the excimer beam, and the excimer beam provides extremely precise energy delivery, stimulates the nerve receptors and peripheral blood vessels around the choroid, and ensures that the choroidal defect or atrophy lesion point area is treated, thereby minimizing the damage to the surrounding healthy choroid tissue, and then playing a therapeutic role. The average axial length of adolescents of different ages before and after treatment is shown in Table 3.

[0041] Table 3 The present invention combines the method of stimulating the eyes with an excimer beam, and finds that it can relieve eye muscle fatigue, reduce eye muscle spasm, and blood stasis. At the same time, it can produce biological stimulation inside the eye tissue, improve local blood circulation, and promote the repair of the fundus choroid. By penetrating the excimer beam into the eye tissue, it can not only exert the due biological thermal effect, but also avoid damaging the normal biological tissue of the human body, which is in line with the basic concept of this project.

[0042] The integrated architecture of the system of the present invention includes: The near-infrared light image acquisition and processing module uses near-infrared light to acquire images of the cornea and choroid of the eyeball, and performs digitization and filtering on the acquired image signals to generate processed image data; The image analysis module performs edge detection and region segmentation on the processed image data to identify the choroidal coloboma or atrophy lesion points and outputs the positioning information; The emission parameter determination module determines the emission parameters of the excimer beam according to the corneal thickness, curvature and choroidal image data, and the emission parameters include energy intensity, irradiation time, wavelength and focus position; The excimer beam emission module emits an excimer beam with a wavelength of 248 nanometers according to the emission parameters to irradiate the lesion site for treatment; The communication control unit is used to realize data transmission and collaborative work between modules.

[0043] The modules communicate with each other using CAN bus or SPI protocol, and the data exchange uses a unified frame structure to ensure the stability and real-time performance of the transmission. The emission parameter determination module is the core to coordinate the work of each module: based on the determined emission parameters, it is distributed to the excimer beam emission module to perform treatment, and interacts with the image analysis module to adjust the emission parameters of the excimer beam according to the eye tissue image; the battery charging and voltage stabilization module monitors the power supply status in real time to ensure the stable operation of the equipment; the present invention also includes an alarm prompt module, which issues a warning and records a log in abnormal situations (such as laser failure or low battery). The overall architecture is divided into a hardware layer, a control layer, and an application layer. The hardware layer is connected through a standardized interface, the control layer runs a real-time operating system (RTOS) to ensure task scheduling, and the application layer provides a user interface and algorithm support. This layered design and modular collaboration ensures the efficiency, stability, and accuracy of the equipment, and provides a reliable technical guarantee for eye treatment.

[0044] The emission power of the excimer beam is an important parameter of the excimer laser beam eye training device. Each country has a limit on the maximum emission power: China and European countries do not exceed 1mW / cm 2 ; The maximum power in North America does not exceed 1.2mW / cm 2 ; Japan's maximum power does not exceed 1.1mW / cm 2For example, in practical applications, the shortest distance between the excimer laser beam eye training device and the human eye is considered, and 0.8mW / cm 2 The excimer beam emitter's emission power can achieve maximum prevention and treatment effectiveness without harming human eyes.

[0045] The continuous working time of the excimer laser beam eye training device under normal working conditions should be no less than 1 hour.

[0046] The excimer laser beam eye training device mainly uses near-infrared light in the ISM band between 26.565 and 27.405 MHz. The near-infrared light in this frequency band has good penetrability, can identify and distinguish different reflections from different objects, and is widely used in medicine.

[0047] The wavelength of the excimer beam is an important parameter of the excimer beam output by the excimer laser eye training device. The excimer laser eye training device uses an excimer beam with a wavelength of 248nm. On the one hand, it has a strong penetrating ability and does not harm human tissue; on the other hand, the excimer beam can have a good therapeutic effect on the choroid.

[0048] Based on the analysis of user needs, the excimer laser beam eye training device of the present invention integrates treatment, image recognition and intelligent control through excimer beam technology, near-infrared light image recognition technology and ATmega328P single-chip microcomputer control technology.

[0049] The present invention transmits near-infrared light to the eye through a transmitter. When the near-infrared light encounters the choroid of the fundus of the eye, reflection, scattering, transmission and other phenomena will occur. The receiver receives the electromagnetic waves after being acted upon by the target, processes and analyzes the received signals, reconstructs the image of the choroid of the fundus of the eye, and further identifies the choroidal defects or atrophic lesions.

[0050] In the practical application of the present invention, biological tissues are not only not damaged, but also have the effects of repairing and activating blood circulation, so that biological stimulation is generated inside the eye tissues, local blood circulation is improved, and the fundus choroid repairs the lesion tissues to restore to a normal state.

[0051] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems or computer program products. Therefore, the present application may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware.

[0052] The above are only embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.

Claims

1. An eye cornea treatment system based on near-infrared light recognition, characterized in that: include: A near-infrared light collection module, used to obtain optical signals from the cornea and choroid of the eye and generate an initial image; An image processing module, used for digitizing and filtering the initial image and outputting processed image data; A lesion detection module, for identifying choroidal coloboma or atrophy lesion points from the processed image data and generating positioning information; A parameter determination module, used to determine the emission parameters of the excimer beam according to the corneal thickness and curvature and the choroidal image data, wherein the emission parameters include energy intensity, irradiation time, wavelength and focus position; Used to output a quasi-molecular beam with a wavelength of 248 nanometers according to the emission parameters to irradiate and treat the lesion site.

2. The eye cornea treatment system based on near-infrared light recognition according to claim 1 is characterized in that: The near-infrared light collection module acquires optical signals of the cornea and choroid of the eye in the frequency range of 26.565~27.405 MHz.

3. The eye cornea treatment system based on near-infrared light recognition according to claim 1 is characterized in that: The image processing module performs analog-to-digital conversion on the initial image and removes noise using frequency domain filtering to obtain processed image data.

4. The eye cornea treatment system based on near-infrared light recognition according to claim 1 is characterized in that: The lesion detection module extracts the eyeball tissue contour based on the gradient operator and uses threshold segmentation to identify the boundaries of choroidal coloboma or atrophy lesion points.

5. The eye cornea treatment system based on near-infrared light recognition according to claim 1 is characterized in that: The parameter determination module acquires corneal thickness and curvature data by emitting and receiving near-infrared light at at least two different angles and performing numerical inversion, and then determines the emission parameters of the excimer beam in combination with the choroidal image data.

6. The eye cornea treatment system based on near-infrared light recognition according to claim 5, characterized in that: The energy intensity is graded and set according to the corneal thickness measurement value, and the ratio of the lesion area to the reference area is adjusted in combination with the lesion depth and the logarithmic function operation. When the lesion area is lower than the reference area, the logarithmic term is negative to reduce the energy output, and when the calculated value is lower than the minimum energy threshold, the minimum energy threshold is used as the final output. The irradiation time is segmented and set according to the degree of the lesion, the wavelength is selected in combination with the optical absorption coefficient of the choroid, and the focusing position is calibrated according to the coordinates of the lesion area.

7. The eye cornea treatment system based on near-infrared light recognition according to claim 1 is characterized in that: The excimer beam output by the beam emission module has a power intensity not exceeding 1mW / cm 2 And be able to work continuously for not less than one hour.

8. The eye cornea treatment system based on near-infrared light recognition according to claim 1 is characterized in that: During the irradiation process, the light beam emission module periodically monitors the choroidal reflection signal and compares it with the power deviation threshold. When the power deviation exceeds the threshold, the light beam emission is stopped and an alarm signal is output.

9. The eye cornea treatment system based on near-infrared light recognition according to claim 1, characterized in that: The beam emission module contains a short-circuit detection device, which immediately cuts off the power supply and records the fault information when the current exceeds the preset safety threshold.

10. The eye cornea treatment system based on near-infrared light recognition according to claim 1, characterized in that: The image processing module and the lesion detection module transmit processed image data through the data bus, and the parameter determination module and the light beam emission module are coordinated and scheduled through control signals.

Images