Visual function grading method and system based on multi-channel visual electrophysiological signals

Abstract

The invention provides a visual function grading method and system based on multi-channel visual electrophysiological signals, and belongs to the technical field of medical signal processing. The method comprises the steps that the multi-channel visual electrophysiological signals of a patient are acquired; based on a preset classification model, performing visual function grade prediction processing on a multi-channel time sequence signal corresponding to the multi-channel visual electrophysiological signal to obtain a visual function grade prediction result of the patient; and performing data post-processing on the visual function grade prediction result based on the basic clinical information of the patient to obtain a final visual function grade evaluation result of the patient. By adopting the technical scheme provided by the invention, visual function grade evaluation can be carried out by utilizing the multi-channel visual electrophysiological signals, so that the accuracy and consistency of evaluation results of visual disability grades are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of medical signal processing, and particularly relates to a visual function grading method and system based on multi-channel visual electrophysiological signals. BACKGROUND

[0002] With the increasing attention to inclusive development in society, the needs of low-vision population in daily life and professional activities have gradually attracted widespread attention from society. Accurate visual disability grading for this population has become an important part of the social security system, which not only directly affects the support for this population by the government, but also provides a basis for the formulation of relevant social welfare policies.

[0003] However, in the related art, the visual disability grading method mainly relies on the use of visual acuity charts and perimeters to measure the visual acuity and visual field of low-vision patients, which is highly dependent on the subjective cooperation of patients and is easily affected by physiological and psychological factors of patients, thereby resulting in a lack of accuracy and consistency in the evaluation results of visual disability grading. SUMMARY

[0004] The main purpose of the embodiments of the present application is to propose a visual function grading method and system based on multi-channel visual electrophysiological signals, which aims to use multi-channel visual electrophysiological signals to grade visual function, thereby improving the accuracy and consistency of the evaluation results of visual disability grading.

[0005] To achieve the above purpose, a first aspect of the embodiments of the present application proposes a visual function grading method based on multi-channel visual electrophysiological signals, which comprises:

[0006] obtaining multi-channel visual electrophysiological signals of a patient;

[0007] performing visual function level prediction processing on multi-channel time sequence signals corresponding to the multi-channel visual electrophysiological signals based on a preset classification model, to obtain a visual function level prediction result of the patient;

[0008] performing data post-processing on the visual function level prediction result based on basic clinical information of the patient, to obtain a final visual function level evaluation result of the patient.

[0009] In some embodiments, the obtaining of the multi-channel visual electrophysiological signals of the patient comprises:

[0010] acquiring visual electrophysiological signals of the left eye and the right eye of the patient through a plurality of preset test modes, to obtain the multi-channel visual electrophysiological signals of the patient;

[0011] The plurality of test modes include a test mode of an electroretinogram and a test mode of a visual evoked potential.

[0012] In some embodiments, the test mode of the electroretinogram comprises a photopic single flash response test mode and a photopic flicker response test mode, and the test mode of the visual evoked potential comprises a flash visual evoked potential test mode;

[0013] The acquiring of the visual electrophysiological signals of the left eye and the right eye of the patient through the preset multiple test modes comprises:

[0014] Through the photopic single flash response test mode, the single flash response of the retinal cone cells of the left eye and the right eye of the patient to a preset standard flash intensity is acquired when the left eye and the right eye of the patient are in a photopic state; the multi-channel visual electrophysiological signal comprises the single flash response of the left eye and the right eye of the patient, respectively;

[0015] Through the photopic flicker response test mode, the changing light response of the retinal cone cells of the left eye and the right eye of the patient to a preset high-frequency flicker light is acquired when the left eye and the right eye of the patient are in a photopic state; the multi-channel visual electrophysiological signal comprises the changing light response of the left eye and the right eye of the patient, respectively;

[0016] Through the flash visual evoked potential test mode, the brain visual cortex response of the left eye and the right eye of the patient induced by full-field flash stimulation is acquired; the multi-channel visual electrophysiological signal comprises the brain visual cortex response of the left eye and the right eye of the patient, respectively.

[0017] In some embodiments, the visual function grade prediction processing of the multi-channel time sequence signal corresponding to the multi-channel visual electrophysiological signal based on the preset classification model comprises at least one of the following:

[0018] The visual function grade prediction processing of the multi-channel time sequence signal corresponding to the multi-channel visual electrophysiological signal based on the preset classification model is regression based on clinical indicators;

[0019] The visual function grade prediction processing of the multi-channel time sequence signal corresponding to the multi-channel visual electrophysiological signal based on the preset classification model is machine learning based;

[0020] The visual function grade prediction processing of the multi-channel time sequence signal corresponding to the multi-channel visual electrophysiological signal based on the preset classification model is deep neural network based.

[0021] In some embodiments, the multi-channel time sequence signal comprises a multi-channel one-dimensional time sequence signal;

[0022] Before the visual function grade prediction processing of the multi-channel time sequence signal corresponding to the multi-channel visual electrophysiological signal based on the preset classification model, the method further comprises at least one of the following:

[0023] inputting the multi-channel one-dimensional time sequence signal into a preset classification model;

[0024] mapping the multi-channel one-dimensional time sequence signal to a two-dimensional feature space to obtain a multi-channel two-dimensional time sequence signal, and inputting the multi-channel two-dimensional time sequence signal into the preset classification model.

[0025] In some embodiments, before the multi-channel visual electrophysiological signal corresponding multi-channel time sequence signal is processed by the preset classification model to predict the visual function level, the method further comprises:

[0026] standardizing the multi-channel visual electrophysiological signal, and performing local time domain corresponding feature analysis on the visual electrophysiological signals of the left eye and the right eye of the patient under different test modes in the multi-channel visual electrophysiological signal to obtain the multi-channel time sequence signal corresponding to the multi-channel visual electrophysiological signal.

[0027] In some embodiments, the data post-processing of the visual function level prediction result based on the basic clinical information of the patient comprises:

[0028] regression fine-tuning the visual function level prediction result based on the basic clinical information of the patient;

[0029] credibility evaluation of the visual function level prediction result based on the basic clinical information of the patient.

[0030] To achieve the above-mentioned purposes, a second aspect of the embodiments of the present application proposes a visual function grading system based on multi-channel visual electrophysiological signals, the system comprising:

[0031] a data acquisition and preprocessing module configured to acquire the multi-channel visual electrophysiological signals of the patient;

[0032] a feature modeling and prediction module configured to perform visual function level prediction processing on the multi-channel time sequence signal corresponding to the multi-channel visual electrophysiological signal based on a preset classification model to obtain the visual function level prediction result of the patient;

[0033] a grading and visualization analysis module configured to perform data post-processing on the visual function level prediction result based on the basic clinical information of the patient to obtain the final visual function level evaluation result of the patient.

[0034] To achieve the above-mentioned purposes, a third aspect of the embodiments of the present application proposes a computer device, the computer device comprising a memory and a processor, the memory storing a computer program, and the processor implementing the method of the first aspect described above when executing the computer program.

[0035] To achieve the above object, a computer readable storage medium is provided in the fourth aspect of the embodiments of the present application, the computer readable storage medium stores a computer program, and the computer program is executed by a processor to implement the method in the first aspect.

[0036] To achieve the above object, a computer program product is provided in the fifth aspect of the embodiments of the present application, the computer program product stores a computer program, and the computer program is executed by a processor to implement the method in the first aspect.

[0037] The method and system for visual function grading based on multi-channel visual electrophysiological signals, the computer device, the computer readable storage medium and the computer program product provided in the embodiments of the present application, by acquiring the multi-channel visual electrophysiological signals of the patient, performing visual function grade prediction processing on the multi-channel time sequence signals corresponding to the multi-channel visual electrophysiological signals based on a preset classification model, obtaining the visual function grade prediction result of the patient, and performing data post-processing on the visual function grade prediction result based on the basic clinical information of the patient, obtaining the final visual function grade evaluation result of the patient.

[0038] In this way, compared with the traditional visual disability grading method, the embodiments of the present application acquire the multi-channel visual electrophysiological signals of the patient, then perform visual function grade prediction processing on the multi-channel time sequence signals corresponding to the multi-channel visual electrophysiological signals based on a classification model, and perform data post-processing on the visual function grade prediction result based on the basic clinical information of the patient, to obtain the final visual function grade evaluation result of the patient. In this way, the multi-channel visual electrophysiological signals of the patient are predicted by the classification model, and the model prediction result is post-processed in combination with the basic clinical information of the patient, which can effectively objectively and automatically evaluate the visual function state of the patient, break the limitation of the traditional method which depends on the subjective cooperation of the patient and is easily affected by the physiological and psychological factors of the patient, and improve the accuracy and consistency of the visual disability grading result.

[0039] In addition, the technical scheme provided in the embodiments of the present application does not need to rely too much on the subjective cooperation of the patient, and is particularly suitable for the auxiliary diagnosis and grading management of low vision subjects, infants and children, and patients with neuro-ophthalmic diseases, has good universality, scalability and clinical application potential, and has important application and popularization value.

[0040] Furthermore, the embodiments of the present application have stronger quantization ability and identification objectivity than the traditional method, which is conducive to the construction of a scientific visual function grading system. BRIEF DESCRIPTION OF DRAWINGS

[0041] Figure 1A flowchart illustrating the steps of the visual function grading method based on multi-channel visual electrophysiological signals provided in some embodiments of this application;

[0042] Figure 2 for Figure 1 A detailed flowchart of step S102;

[0043] Figure 3 A schematic diagram of the system structure involved in a complete embodiment of the visual function grading method based on multi-channel visual electrophysiological signals provided in this application embodiment;

[0044] Figure 4 A schematic diagram of the structure of a visual function grading system based on multi-channel visual electrophysiological signals provided in an embodiment of this application;

[0045] Figure 5 This is a schematic diagram of the hardware structure of a computer device provided in an embodiment of this application. Detailed Implementation

[0046] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0047] It should be noted that although functional modules are divided in the device schematic diagram and a logical order is shown in the flowchart, in some cases, the steps shown or described may be performed in a different order than the module division in the device or the order in the flowchart. The terms "first," "second," etc., in the specification, claims, and the aforementioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence.

[0048] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.

[0049] First, the overall concept of the embodiments of this application will be explained.

[0050] With increasing societal emphasis on inclusive development, the needs of people with low vision in their daily lives and professional activities are gradually attracting widespread attention. Accurate assessment of the visual disability level of this population has become an important component of the social security system, directly impacting the government's support for them and providing a basis for the formulation of relevant social welfare policies.

[0051] However, current methods for assessing visual impairment levels primarily rely on visual acuity charts and perimeters to measure the visual acuity and visual field of patients with low vision. This method heavily depends on the subjective cooperation of the patient (also known as the test subject) and is easily influenced by the patient's physiological and psychological factors, leading to a lack of accuracy and consistency in the assessment results. For example, traditional methods of assessing visual impairment levels require patients to make explicit responses to visual stimuli (such as identifying the direction of patterns on the visual acuity chart or the location of flashing stimuli when pressing buttons) during subjective behavioral tests like visual acuity chart and perimeter tests. This makes the test results highly dependent on the patient's subjective cooperation. Consequently, for patients with poor concentration, insufficient comprehension, or poor willingness to cooperate, assessment results that are lower than their true visual function level are easily obtained, leading to distortion in the visual function grading.

[0052] Furthermore, traditional methods of assessing visual impairment levels cannot directly evaluate the patient's retinal function from the perspective of the physiological response of the visual system, which is not conducive to establishing a scientific and continuous visual function grading standard.

[0053] Based on this, embodiments of this application propose a method, system, computer device, computer-readable storage medium, and computer program product for visual function grading based on multi-channel visual electrophysiological signals. The method involves acquiring a patient's multi-channel visual electrophysiological signals; performing visual function level prediction processing on the multi-channel time-series signals corresponding to the multi-channel visual electrophysiological signals based on a preset classification model to obtain the patient's visual function level prediction result; and performing data post-processing on the visual function level prediction result based on the patient's basic clinical information to obtain the patient's final visual function level assessment result.

[0054] Therefore, compared to traditional methods of assessing visual impairment levels, this embodiment acquires the patient's multi-channel visual electrophysiological signals, then uses a classification model to predict the visual function level of the corresponding multi-channel time-series signals, and performs post-processing on the predicted visual function level based on the patient's basic clinical information to obtain the patient's final visual function level assessment result. In this way, by predicting the patient's multi-channel visual electrophysiological signals using a classification model and combining the prediction results with the patient's basic clinical information for post-processing, an effective, objective, and automated assessment of the patient's visual function status can be achieved. This overcomes the limitations of traditional methods that rely on the patient's subjective cooperation and are easily influenced by the patient's physiological and psychological factors, thus improving the accuracy and consistency of visual impairment level assessment results.

[0055] Furthermore, the technical solutions provided in this application do not require excessive reliance on the patient's subjective cooperation, making them particularly suitable for the auxiliary diagnosis and triage management of low-vision subjects, infants, and people with neuro-ophthalmological diseases. They have good versatility, scalability, and clinical applicability potential, and possess significant application and promotion value.

[0056] Furthermore, the embodiments of this application have stronger quantitative capabilities and objectivity of identification compared with traditional methods, which is conducive to the construction of a scientific visual function classification system.

[0057] It should be noted that the inventors of this application considered that visual electrophysiological testing, by recording the bioelectrical signals of the retina and visual conduction pathways, can more objectively and quantitatively reflect the functional state of the visual system. Furthermore, through extensive research, the inventors discovered that as visual dysfunction worsens, patients' visual electrophysiological signals (such as electroretinograms (ERG)) exhibit significant waveform delays and amplitude reductions. Therefore, the inventors of this application conceived of a method for classifying visual function using visual electrophysiological signals, thereby reducing the reliance on patient subjective cooperation and providing a new technical approach for achieving accurate and automated assessment of visual disability levels.

[0058] In this embodiment, the temporal characteristics of multi-channel electroretinography (ERG) and visual evoked potential (VEP) signals can be collected and analyzed, and combined with models such as clinical indicator regression and deep learning to achieve objective and automated grading of patients' visual impairment. Thus, compared to traditional grading methods relying on visual acuity charts or perimeters, the solution provided in this embodiment does not require subjective cooperation from the subject, avoiding assessment bias caused by differences in compliance and improving the accuracy and consistency of the grading results.

[0059] It should be noted that electroretinography (ERG) is a type of visual electrophysiological examination, primarily used to assess a patient's retinal function (such as the function of rod / cone cells). Visual evoked potentials (VEP) are also a type of visual electrophysiological examination, primarily used to assess a patient's optic nerve and central conduction pathways.

[0060] Based on the overall concept of the embodiments of this application described above, specific embodiments of the visual function grading method, apparatus, computer device, computer-readable storage medium, and computer program product based on multi-channel visual electrophysiological signals provided in the embodiments of this application are proposed. First, the various specific embodiments of the visual function grading method based on multi-channel visual electrophysiological signals in the embodiments of this application are described in detail.

[0061] It should be noted that the embodiments of this application can acquire and process relevant data based on artificial intelligence technology. Artificial intelligence (AI) refers to the theories, methods, technologies, and application systems that use digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to obtain optimal results.

[0062] Foundational technologies for artificial intelligence generally include sensors, dedicated AI chips, cloud computing, distributed storage, big data processing, operating / interactive systems, and mechatronics. AI software technologies mainly encompass computer vision, robotics, biometrics, speech processing, natural language processing, and machine learning / deep learning.

[0063] Furthermore, in various specific embodiments of this application, when processing data related to user identity or characteristics, such as user information, user behavior data, user historical data, and user location information, user permission or consent is obtained first. Moreover, the collection, use, and processing of this data comply with relevant laws, regulations, and standards. Additionally, when embodiments of this application require access to sensitive personal information of users, separate permission or consent from the user is obtained through pop-ups or redirects to confirmation pages. Only after explicitly obtaining the user's separate permission or consent is the necessary user-related data for the proper functioning of these embodiments acquired.

[0064] Furthermore, the visual function grading method based on multi-channel visual electrophysiological signals provided in this application can be applied to a terminal, a server, or software running on either a terminal or a server. In some embodiments, the terminal can be a medical imaging device (e.g., a medical image acquisition device), a control and management device for a medical imaging device, a smartphone, tablet, laptop, desktop computer, etc.; the server can be configured as an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms; the software can be an application that implements the visual function grading method based on multi-channel visual electrophysiological signals, etc., but is not limited to the above forms.

[0065] Alternatively, the visual function grading method based on multi-channel visual electrophysiological signals provided in this application embodiment can also be used in numerous general-purpose or special-purpose computer system environments or configurations. Examples include: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer computer devices, network PCs, minicomputers, mainframe computers, and distributed computing environments including any of the above systems or devices, etc. This application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform specific tasks or implement specific abstract data types. This application can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communication network. In a distributed computing environment, program modules can reside in local and remote computer storage media, including storage devices.

[0066] For ease of understanding and explanation, the following description will use the visual function classification method based on multi-channel visual electrophysiological signals provided in the embodiments of this application as an example. The implementation of the visual function classification method based on multi-channel visual electrophysiological signals provided in the embodiments of this application in any of the above-described forms can refer to the process of applying the visual function classification method based on multi-channel visual electrophysiological signals in a terminal device as described below.

[0067] Please refer to Figure 1 , Figure 1 The flowchart illustrates the steps of the visual function grading method based on multi-channel visual electrophysiological signals provided in some embodiments of this application. It should be understood that, although... Figure 1 The figure shows the execution order of some method steps, but based on the different design needs of practical applications, the visual function grading method based on multi-channel visual electrophysiological signals provided in this application embodiment can of course adopt a different execution order of method steps than that shown in the figure. That is, Figure 1 The order of the method steps shown does not constitute a limitation on the execution logic order of the visual function grading method based on multi-channel visual electrophysiological signals provided in the embodiments of this application. Any other method based on... Figure 1 Reasonable changes to the sequence of steps shown should be included within the protection scope of the visual function grading method based on multi-channel visual electrophysiological signals provided in the embodiments of this application.

[0068] like Figure 1 As shown, in some embodiments, the visual function grading method based on multi-channel visual electrophysiological signals provided in this application may include, but is not limited to, steps S101 to S103.

[0069] Step S101: Acquire the patient's multi-channel visual electrophysiological signals.

[0070] Before formally classifying visual function, the terminal device first performs a visual electrophysiological examination on the patient who needs to be classified to obtain the patient's multi-channel visual electrophysiological signals.

[0071] In some embodiments, the terminal device can acquire the patient's multi-channel visual electrophysiological signals by performing visual electrophysiological examinations on the patient's electroretinography (ERG) and visual evoked potentials (VEP).

[0072] Step S102: Based on a preset classification model, perform visual function level prediction processing on the multi-channel time-series signals corresponding to the multi-channel visual electrophysiological signals to obtain the visual function level prediction results of the patient.

[0073] It should be noted that the pre-set classification model can be obtained by training the terminal device with large-scale clinical data. In some embodiments, model training can employ a supervised learning mechanism, and training labels can be assigned by clinicians based on patients' previous examination results. In some embodiments, to improve the robustness and clinical adaptability of the classification model, several optimization strategies can be introduced during model training. For example, using the FocalLoss loss function to alleviate class imbalance, employing signal perturbation and time shifting for data augmentation, and applying K-fold cross-validation to improve generalization ability. In some embodiments, the classification model can also embed an interpretability analysis module (such as visualization technology (Gradient-weighted Class Activation Mapping, Grad-CAM)) to provide visualized results of the prediction basis, enhancing clinicians' understanding and trust in the model output. In some embodiments, the classification model can include four functional sub-modules: signal feature extraction, time-dependent modeling, global feature fusion, and multi-class discrimination. Among these, the feature extraction submodule is used to identify local waveform changes in ERG; the timing modeling part can capture the dynamic evolution of the signal through a network structure with remote dependency modeling capabilities; the global feature fusion submodule is a functional submodule that introduces a fusion mechanism to enhance the integration capability of information between different channels in order to improve the global semantic representation.

[0074] After acquiring the patient's multi-channel visual electrophysiological signals, the terminal device can input the multi-channel time-series signals corresponding to the multi-channel visual electrophysiological signals into a preset classification model. The classification model then performs visual function level prediction processing on the multi-channel time-series signals to obtain the visual function level prediction result of the patient output by the classification model.

[0075] In some embodiments, the terminal device can output a multi-class prediction result obtained by processing multi-channel temporal signals for visual function level prediction using a classification model based on a multi-layer perceptual structure or a regressor. In this way, the terminal device can acquire the multi-class prediction result and use it as the patient's visual function level prediction result.

[0076] Step S103: Based on the patient's basic clinical information, perform data post-processing on the visual function level prediction results to obtain the patient's final visual function level assessment result.

[0077] After obtaining the predicted visual function level of the patient based on the classification model, the terminal device further performs data post-processing on the predicted visual function level based on the patient's basic clinical information (such as age, gender, past medical history, etc.) to obtain the patient's final visual function level assessment result.

[0078] In some embodiments, the terminal device can further fuse the patient's visual function level prediction results output by the classification model with multi-channel response features, and then combine the patient's basic clinical information to perform post-processing on the fused visual function level prediction results, thereby generating the patient's final visual function level assessment result.

[0079] In some embodiments, the terminal device can generate a two-dimensional heat map to visually present the contribution of different channels or frequency bands to the visual function level prediction results, and automatically generate a structured grading report for doctors to refer to, archive and follow up.

[0080] For example, the terminal device can first establish standards for visual impairment levels, compatible with the current "Standards for the Assessment of Disability Levels" and the visual function classification criteria in the "ICF International Classification of Functional Disability and Health," establishing five levels of visual function: Level I – Severe visual impairment (near total blindness), corresponding to Level 1 blindness; Level II – High visual impairment, corresponding to Level 2 blindness; Level III – Moderate visual impairment, corresponding to Level 3 low vision; Level IV – Mild visual impairment, corresponding to Level 4 low vision; Level V – Normal visual function. Then, the terminal device generates and outputs the grading results, mapping the visual function level prediction results output by the classification model to the above five-level standards and generating an individualized visual function grading report for clinicians to use as a diagnostic reference or rehabilitation assessment basis for patients' visual function.

[0081] In this embodiment, a visual electrophysiological examination is first performed on a patient requiring visual function grading using a terminal device to obtain the patient's multi-channel visual electrophysiological signals. Then, the multi-channel time-series signals corresponding to the multi-channel visual electrophysiological signals are input into a preset classification model. The classification model then performs visual function level prediction processing on the multi-channel time-series signals to obtain the visual function level prediction result of the patient output by the classification model. Finally, the visual function level prediction result is further post-processed based on the patient's basic clinical information to obtain the patient's final visual function level assessment result.

[0082] Therefore, compared to traditional methods of assessing visual impairment levels, this embodiment acquires the patient's multi-channel visual electrophysiological signals, then uses a classification model to predict the visual function level of the corresponding multi-channel time-series signals, and performs post-processing on the predicted visual function level based on the patient's basic clinical information to obtain the patient's final visual function level assessment result. In this way, by predicting the patient's multi-channel visual electrophysiological signals using a classification model and combining the prediction results with the patient's basic clinical information for post-processing, an effective, objective, and automated assessment of the patient's visual function status can be achieved. This overcomes the limitations of traditional methods that rely on the patient's subjective cooperation and are easily influenced by the patient's physiological and psychological factors, thus improving the accuracy and consistency of visual impairment level assessment results.

[0083] Furthermore, the technical solutions provided in this application do not require excessive reliance on the patient's subjective cooperation, making them particularly suitable for the auxiliary diagnosis and triage management of low-vision subjects, infants, and people with neuro-ophthalmological diseases. They have good versatility, scalability, and clinical applicability potential, and possess significant application and promotion value.

[0084] Furthermore, the embodiments of this application have stronger quantitative capabilities and objectivity of identification compared with traditional methods, which is conducive to the construction of a scientific visual function classification system.

[0085] In some embodiments, step S101 above: acquiring the patient's multi-channel visual electrophysiological signals may include the following steps:

[0086] By collecting visual electrophysiological signals from the patient's left and right eyes using multiple preset test modes, multi-channel visual electrophysiological signals of the patient were obtained.

[0087] It should be noted that multiple testing modes are available, including electroretinography (ERG) and visual evoked potentials (VEP).

[0088] When the terminal device performs visual electrophysiological examinations on the patient using electroretinography (ERG) and visual evoked potentials (VEP), and obtains the patient's multi-channel visual electrophysiological signals, it can acquire visual electrophysiological signals from the patient's left and right eyes respectively through the ERG test mode and the VEP test mode, thereby obtaining the patient's multi-channel visual electrophysiological signals.

[0089] In some embodiments, when the terminal device acquires visual electrophysiological signals from the patient's left and right eyes using the electroretinography (ERG) test mode, it can do so after the patient's left and right eyes have been adapted to a certain intensity of light (typically background light 3.0 cd·s / m²). 2 In a state of light adaptation (LA), visual electrophysiological tests were performed on the patient's left and right eyes to collect visual electrophysiological signals from the corresponding channels of the patient's left and right eyes under this test mode.

[0090] In some embodiments, the above-described electroretinography (ERG) test modes include a light-adapted single flash response test mode (e.g., LA 3.0) and a light-adapted flicker response test mode (e.g., LA 30Hz). Furthermore, the above-described visual evoked potential (VEP) test modes include a flash visual evoked potential test mode (e.g., Flash VEP).

[0091] In this case, the aforementioned step of "collecting visual electrophysiological signals from the patient's left and right eyes through multiple preset test modes" can include the following steps:

[0092] Using the light-adapted single-flash response test mode, with the patient's left and right eyes in a light-adapted state, the single-flash response of the retinal cone cells of the left and right eyes to a preset standard flash intensity is collected; the multi-channel visual electrophysiological signal includes the single-flash response of the patient's left and right eyes respectively;

[0093] Using the light-adapted flicker response test mode, with the patient's left and right eyes in a light-adapted state, the changes in the light responses of the retinal cone cells of the left and right eyes to a preset high-frequency flicker light are collected; the multi-channel visual electrophysiological signals include the changes in the light responses of the patient's left and right eyes respectively.

[0094] The visual cortical responses of the patient's left and right eyes induced by the flash visual evoked potential test mode were collected using the aforementioned flash visual evoked potential test mode. The multi-channel visual electrophysiological signals included the visual cortical responses of the patient's left and right eyes respectively.

[0095] When the terminal device acquires visual electrophysiological signals from the patient's left and right eyes using the electroretinography (ERG) and visual evoked potential (VEP) testing modes, it can also use the light-adapted single flash response test mode LA 3.0. This mode allows for the acquisition of retinal cone cell responses to a preset standard flash intensity (e.g., 3.0 cd·s / m²) from the patient's left and right eyes, either separately or simultaneously, while both eyes are in a light-adapted state. 2 The single flash response of the patient's left and right eyes was collected and used as the visual electrophysiological signals of two channels in the multi-channel visual electrophysiological signal.

[0096] In addition, the terminal device can also collect the changes in light response of retinal cone cells to a preset high-frequency flicker light (e.g., 30Hz high-frequency flicker light) in the light-adapted state of the patient's left and right eyes, either separately or simultaneously, through the light-adapted flicker response test mode LA 30Hz. Then, the collected changes in light response of the patient's left and right eyes are used as the visual electrophysiological signals of the other two channels of the multi-channel visual electrophysiological signal.

[0097] Furthermore, the terminal device can also use the Flash Visual Evoked Potential (FPV) test mode to induce the patient's visual cortex with full-field flash stimulation, thereby collecting the patient's left and right eye visual cortex responses induced under full-field flash stimulation, and using the corresponding visual cortex responses of the patient's left and right eyes as the visual electrophysiological signals of the other two channels of the multi-channel visual electrophysiological signal.

[0098] In some embodiments, the terminal device can collect the patient's visual electrophysiological signals under the above-mentioned multiple test modes through the electrophysiological detection module, that is, it includes a total of six channels, namely the visual electrophysiological signals of the left and right eyes in the LA 30Hz ERG mode, the visual electrophysiological signals of the left and right eyes in the LA 3.0ERG mode, and the visual electrophysiological signals of the left and right eyes in the Flash VEP mode.

[0099] In some embodiments, after acquiring the patient's multi-channel visual electrophysiological signals, the terminal device can further preprocess the multi-channel visual electrophysiological signals to obtain the multi-channel time-series signals corresponding to the multi-channel visual electrophysiological signals, and then input the multi-channel time-series signals into a preset classification model for visual function level prediction processing.

[0100] Based on this, before step S102 above: performing visual function level prediction processing on the multi-channel temporal signals corresponding to the multi-channel visual electrophysiological signals based on a preset classification model, the visual function classification method based on multi-channel visual electrophysiological signals provided in this application embodiment may further include the following steps:

[0101] The multi-channel visual electrophysiological signals are standardized and preprocessed, and local temporal correspondence feature analysis is performed on the visual electrophysiological signals of the patient's left and right eyes under different test modes in the multi-channel visual electrophysiological signals to obtain the multi-channel temporal signals corresponding to the multi-channel visual electrophysiological signals.

[0102] To ensure the accuracy and stability of assessing patients' visual function levels based on their multi-channel visual electrophysiological signals, the terminal device can further perform a series of standardized preprocessing steps on the acquired multi-channel visual electrophysiological signals. These preprocessing steps may include noise reduction through stacking, bandpass filtering, artifact removal, artifact segment compensation, and baseline correction. Furthermore, the terminal device can combine experimental task annotation information to accurately extract specified time windows and perform multi-channel data alignment and normalization on the multi-channel visual electrophysiological signals, forming standardized time-series data suitable for input into the classification model.

[0103] Subsequently, the terminal device can perform local temporal correspondence feature analysis on the visual electrophysiological signals of the left and right eyes in each test mode based on the synergy of the visual electrophysiological signals of the patient's left and right eyes.

[0104] For example, taking the visual electrophysiological signals of the left and right eyes in the LA 30Hz ERG mode as an example, the terminal device can first use a Gaussian derivative filter to calculate the gradient of the ERG signals of the left and right eyes respectively, so as to reduce high-frequency noise interference in the signal. The specific formula for gradient calculation can be as shown in formula (1) below.

[0105]

[0106] Among these, g σ (t) is a Gaussian kernel function with standard deviation σ, E(t) represents the ERG signals of the left and right eyes at time t, and * represents the convolution operation.

[0107] Then, the terminal device can use the normalized cross-correlation coefficient to calculate the similarity between the gradients of the left and right eye signals. The specific calculation formula is shown in formula (2) below.

[0108]

[0109] Among them, E f,L and E f,R These are the gradient-filtered ERG signals for the left and right eyes, respectively. This represents the corresponding time average.

[0110] Finally, the terminal device can perform local temporal correspondence feature analysis between the left and right eye ERG signals using the following formula (3):

[0111]

[0112] Thus, the local temporal correspondence features between the left and right eye ERG signals ultimately obtained by the terminal device are also one-dimensional temporal signals. By performing the same operation on the visual electrophysiological information of the six channels in the above three test modes using the terminal device, three local temporal correspondence feature signals can be obtained and used as three additional signal channels.

[0113] Please refer to Figure 2 , Figure 2 for Figure 1 A detailed flowchart of step S102.

[0114] In some embodiments, such as Figure 2 As shown, step S102 above: performing visual function level prediction processing on the multi-channel time-series signal corresponding to the multi-channel visual electrophysiological signal based on a preset classification model, may include at least one of steps S201 to S203 as shown below.

[0115] Step S201: Based on clinical index regression, the multi-channel time-series signals corresponding to the multi-channel visual electrophysiological signals are processed to predict visual function levels using a preset classification model.

[0116] When the terminal device performs visual function level prediction processing on the multi-channel time-series signals corresponding to the patient's multi-channel visual electrophysiological signals based on the classification model, it can use the classification model to regress clinical indicators such as α and β wave amplitude and phase, and automatically predict the patient's visual disability level based on the preprocessed multi-channel time-series signals, thereby obtaining the patient's visual function level prediction result.

[0117] Step S202: Using a preset classification model and machine learning, perform visual function level prediction processing on the multi-channel time-series signals corresponding to the multi-channel visual electrophysiological signals;

[0118] When the terminal device performs visual function level prediction processing on the multi-channel time-series signals corresponding to the patient's multi-channel visual electrophysiological signals based on the classification model, it can also use machine learning methods such as decision tree and support vector machine (SVM) supervised learning model to automatically predict the patient's visual disability level based on the preprocessed multi-channel time-series signals, thereby obtaining the patient's visual function level prediction result.

[0119] Step S203: Based on a deep neural network, a preset classification model is used to perform visual function level prediction processing on the multi-channel time-series signals corresponding to the multi-channel visual electrophysiological signals.

[0120] When the terminal device performs visual function level prediction processing on the multi-channel time-series signals corresponding to the patient's multi-channel visual electrophysiological signals based on the classification model, it can also use the classification model to adopt deep neural networks such as Convolutional Neural Network (CNN), Transformer, and Mamba to automatically predict the patient's visual disability level based on the pre-processed multi-channel time-series signals, thereby obtaining the patient's visual function level prediction result.

[0121] In some embodiments, when the multi-channel time-series signal includes a multi-channel one-dimensional time-series signal, before step S102 above: performing visual function level prediction processing on the multi-channel time-series signal corresponding to the multi-channel visual electrophysiological signal based on a preset classification model, the terminal device, when executing the visual function classification method based on multi-channel visual electrophysiological signals provided in this application embodiment, can also input the multi-channel time-series signal into the classification model through the following methods 1 and / or 2, so that the classification model can perform visual function level prediction processing on the multi-channel time-series signal.

[0122] Method 1: Input the multi-channel one-dimensional time-series signal into a preset classification model.

[0123] After obtaining the multi-channel time-series signal corresponding to the patient's multi-channel visual electrophysiological signal, the terminal device can directly input the multi-channel one-dimensional time-series signal into the classification model, since the multi-channel time-series signal is already a one-dimensional time-series signal suitable for model input. This allows the classification model to perform visual function level prediction processing on the multi-channel one-dimensional time-series signal and obtain the patient's visual function level prediction result.

[0124] Method 2: Map the multi-channel one-dimensional time-series signal to a two-dimensional feature space to obtain a multi-channel two-dimensional time-series signal, and input the multi-channel two-dimensional time-series signal into a preset classification model.

[0125] After obtaining the multi-channel time-series signal corresponding to the patient's multi-channel visual electrophysiological signal, the terminal device can also use wavelet transform, short-time Fourier transform and other methods to map the one-dimensional visual electrophysiological signal (i.e., multi-channel one-dimensional time-series signal) into a two-dimensional feature space to obtain a multi-channel two-dimensional time-series signal. Then, the multi-channel two-dimensional time-series signal is input into the classification model, so that the classification model performs visual function level prediction processing on the multi-channel two-dimensional time-series signal to obtain the patient's visual function level prediction result.

[0126] In some embodiments, the terminal device may first input a multi-channel one-dimensional time-series signal into a classification model to obtain a first visual function level prediction result output by the classification model after performing visual function level prediction processing on the multi-channel one-dimensional time-series signal. Then, the terminal device inputs a multi-channel two-dimensional time-series signal into the classification model to obtain a second visual function level prediction result output by the classification model after performing visual function level prediction processing on the multi-channel two-dimensional time-series signal. Finally, the terminal device fuses the first and second visual function level prediction results and uses the fused visual function level prediction result as the patient's visual function level prediction result.

[0127] In this embodiment, the classification model used by the terminal device is not limited to learning features from one-dimensional visual electrophysiological signals. That is, the terminal device uses wavelet transform, short-time Fourier transform and other methods to map one-dimensional visual electrophysiological signals into a two-dimensional feature space, which makes the selection of classification models more extensive and the feature patterns easier to visualize, thereby bringing stronger interpretability to the visual function level prediction results.

[0128] In some embodiments, step S103 above, the step of "post-processing the visual function level prediction result based on the patient's basic clinical information" may include the following steps:

[0129] The visual function level prediction results were fine-tuned based on the patient's basic clinical information;

[0130] The reliability of the visual function level prediction results is assessed based on the patient's basic clinical information.

[0131] When performing post-processing of visual function level prediction results based on the patient's basic clinical information, the terminal device can further integrate the visual function level prediction results with multi-channel response characteristics. Then, it can combine the patient's basic clinical information such as age, gender, and past medical history to perform regression fine-tuning on the visual function level prediction results and conduct a reliability assessment of the visual function level prediction results. Finally, based on the fine-tuned function level prediction results and the reliability assessment results, the terminal device generates the patient's final visual function level assessment result.

[0132] In some embodiments, the terminal device may further perform a reliability assessment on the refined visual function level prediction results based on the patient's basic clinical information after performing regression fine-tuning on the prediction results based on the basic clinical information, and then generate the patient's final visual function level assessment result based on the reliability assessment result.

[0133] In some embodiments, the terminal device may first assess the reliability of the visual function level prediction results based on the patient's basic clinical information, and then decide whether to further fine-tune the visual function level prediction results based on the reliability assessment results. Then, after deciding to fine-tune the visual function level prediction results based on the reliability assessment results and obtaining the fine-tuned functional level prediction results, the terminal device uses the fine-tuned functional level prediction results as the patient's final visual function level assessment result.

[0134] Next, a complete embodiment of the visual function grading method based on multi-channel visual electrophysiological signals provided in this application is presented.

[0135] In a complete embodiment of the visual function grading method based on multi-channel visual electrophysiological signals provided in this application, the terminal device can assess the visual function level of the patient through a visual function grading system based on multi-channel visual electrophysiological signals (also known as a multi-channel electrophysiological signal analysis system).

[0136] Please refer to Figure 3 , Figure 3 A schematic diagram of the system structure involved in a complete embodiment of the visual function classification method based on multi-channel visual electrophysiological signals provided in this application.

[0137] like Figure 3 As shown, the visual function classification system mainly includes three functional modules: data acquisition and preprocessing module, feature modeling and prediction module, and classification and visualization analysis module.

[0138] In the data acquisition and preprocessing module, the system can acquire visual electrophysiological signals from patients in multiple modes through the electrophysiological detection module, including six channels (LA 30Hz ERG mode, LA 3.0ERG mode, and Flash VEP mode, left and right eyes). To ensure the accuracy and stability of subsequent analysis, the system first performs a series of standardized preprocessing steps on the acquired data, including superposition noise reduction, bandpass filtering, artifact removal, artifact segment compensation, and baseline correction. Combined with experimental task annotation information, the system can accurately extract specified time windows, complete multi-channel data alignment and normalization, and form standardized time-series data suitable for model input. Subsequently, the system performs local temporal correspondence feature analysis on the left and right eye signals in each mode based on the synergy of the visual electrophysiological signals of the left and right eyes. The specific implementation process is described above using the two-channel visual electrophysiological signals of the left and right eyes in the LA 30Hz ERG mode as an example to illustrate the local temporal correspondence feature analysis of the left and right eye signals. Further details on the same content will not be elaborated here.

[0139] By performing the same operation on all six signal channels in three test modes, the system can obtain three local time-domain corresponding characteristic signals, which can be used as three additional signal channels.

[0140] The feature modeling and prediction module is the core component of the system. This module is responsible for automatically predicting the level of visual impairment based on preprocessed multi-channel time-series signals. The classification model used is not limited to learning features from one-dimensional visual electrophysiological signals; it can also map one-dimensional visual electrophysiological signals to a two-dimensional feature space using methods such as wavelet transform and short-time Fourier transform. This allows for a wider range of classification model choices and makes feature patterns easier to visualize, resulting in stronger interpretability.

[0141] In the grading and visualization analysis module, the system further integrates the output of the classification model with multi-channel response features, and combines the subjects' basic clinical information (such as age, gender, and past medical history) for regression fine-tuning and reliability assessment to generate the final visual function level judgment result. Furthermore, the system supports the generation of two-dimensional heatmaps to visually present the contribution of different channels or frequency bands to the prediction results, and automatically generates structured grading reports for easy reference, archiving, and follow-up by doctors.

[0142] In this embodiment, the visual function grading system based on multi-channel visual electrophysiological signals, through modeling mechanisms such as deep neural networks or clinical index regression and global feature fusion strategies, can effectively achieve objective and automated assessment of the visual function status of subjects, breaking through the limitations of traditional reliance on subjective cooperation. It is particularly suitable for the auxiliary diagnosis and grading management of subjects with low vision, infants, and individuals with neuro-ophthalmological diseases. Furthermore, the visual function grading system possesses good versatility, scalability, and clinical applicability potential, making it of significant value for application and promotion.

[0143] Based on the same technical concept as the above-mentioned visual function grading method based on multi-channel visual electrophysiological signals, this application embodiment also provides a visual function grading system based on multi-channel visual electrophysiological signals, which can realize the above-mentioned visual function grading method based on multi-channel visual electrophysiological signals.

[0144] Please see Figure 4 The visual function grading system based on multi-channel visual electrophysiological signals provided in this application embodiment may include:

[0145] The data acquisition and preprocessing module is used to acquire multi-channel visual electrophysiological signals from patients;

[0146] The feature modeling and prediction module is used to perform visual function level prediction processing on the multi-channel time-series signals corresponding to the multi-channel visual electrophysiological signals based on a preset classification model, so as to obtain the visual function level prediction result of the patient.

[0147] The grading and visualization analysis module is used to perform data post-processing on the visual function level prediction results based on the patient's basic clinical information to obtain the patient's final visual function level assessment result.

[0148] In some embodiments, the data acquisition and preprocessing module is further configured to acquire visual electrophysiological signals from the patient's left and right eyes through a variety of preset test modes to obtain the patient's multi-channel visual electrophysiological signals; wherein, the variety of test modes includes an electroretinogram test mode and a visual evoked potential test mode.

[0149] In some embodiments, the electroretinogram testing modes include a light-adapted single flash response testing mode and a light-adapted flicker response testing mode, and the visual evoked potential testing modes include a flash visual evoked potential testing mode.

[0150] The data acquisition and preprocessing module is further configured to, through the light-adapted single-flash response test mode, acquire the single-flash response of the retinal cone cells of the left and right eyes to a preset standard flash intensity in a light-adapted state; the multi-channel visual electrophysiological signals include the single-flash response of the patient's left and right eyes; through the light-adapted flicker response test mode, acquire the changing light response of the retinal cone cells of the left and right eyes to a preset high-frequency flicker light in a light-adapted state; the multi-channel visual electrophysiological signals include the changing light response of the patient's left and right eyes; and through the flash visual evoked potential test mode, acquire the induced visual cortical response of the patient's left and right eyes under full-field flash stimulation; the multi-channel visual electrophysiological signals include the corresponding visual cortical response of the patient's left and right eyes.

[0151] In some embodiments, the feature modeling and prediction module is further configured to perform visual function level prediction processing on the multi-channel time-series signal corresponding to the multi-channel visual electrophysiological signal based on clinical index regression using a preset classification model; perform visual function level prediction processing on the multi-channel time-series signal corresponding to the multi-channel visual electrophysiological signal based on machine learning using a preset classification model; and perform visual function level prediction processing on the multi-channel time-series signal corresponding to the multi-channel visual electrophysiological signal based on a deep neural network using a preset classification model.

[0152] In some embodiments, the multi-channel time series signal includes a multi-channel one-dimensional time series signal; the feature modeling and prediction module is further configured to input the multi-channel one-dimensional time series signal into a preset classification model; and to map the multi-channel one-dimensional time series signal to a two-dimensional feature space to obtain a multi-channel two-dimensional time series signal, and input the multi-channel two-dimensional time series signal into the preset classification model.

[0153] In some embodiments, the data acquisition and preprocessing module is further configured to perform standardized preprocessing on the multi-channel visual electrophysiological signals and to perform local temporal domain correspondence feature analysis on the visual electrophysiological signals of the patient's left and right eyes under different test modes in the multi-channel visual electrophysiological signals to obtain the multi-channel time-series signals corresponding to the multi-channel visual electrophysiological signals.

[0154] In some embodiments, the grading and visualization analysis module is further configured to perform regression fine-tuning of the visual function level prediction results based on the patient's basic clinical information; and to perform a credibility assessment of the visual function level prediction results based on the patient's basic clinical information.

[0155] It should be noted that the specific implementation of the visual function grading device based on multi-channel visual electrophysiological signals provided in this application is basically the same as the specific implementation of the visual function grading method based on multi-channel visual electrophysiological signals described above, and will not be repeated here.

[0156] This application also provides a computer device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the above-described visual function grading method based on multi-channel visual electrophysiological signals. This computer device can be a medical imaging device, a control and management device for medical imaging devices, a smartphone, tablet computer, laptop computer, desktop computer, or other terminal device.

[0157] Please see Figure 5 , Figure 5 This illustration shows the hardware structure of a computer device according to one embodiment. The computer device includes:

[0158] The processor 501 can be implemented using a general-purpose CPU (Central Processing Unit), microprocessor, application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this application.

[0159] The memory 502 can be implemented as a read-only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM). The memory 502 can store the operating system and other application programs. When the technical solutions provided in the embodiments of this specification are implemented through software or firmware, the relevant program code is stored in the memory 502 and is called and executed by the processor 501 to execute the visual function grading method based on multi-channel visual electrophysiological signals according to the embodiments of this application.

[0160] The input / output interface 503 is used to implement information input and output;

[0161] The communication interface 504 is used to enable communication and interaction between this device and other devices. Communication can be achieved through wired means (such as USB, network cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.).

[0162] Bus 505 transmits information between various components of the device (e.g., processor 501, memory 502, input / output interface 503, and communication interface 504);

[0163] The processor 501, memory 502, input / output interface 503, and communication interface 504 are connected to each other within the device via bus 505.

[0164] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described visual function grading method based on multi-channel visual electrophysiological signals.

[0165] Memory, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs and non-transitory computer-executable programs. Furthermore, memory may include high-speed random access memory, and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, memory may optionally include memory remotely located relative to the processor, and these remote memories can be connected to the processor via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0166] This application also provides a computer program product that stores a computer program that, when executed by a processor, implements the above-described visual function grading method based on multi-channel visual electrophysiological signals.

[0167] The embodiments described in this application are for the purpose of more clearly illustrating the technical solutions of the embodiments of this application, and do not constitute a limitation on the technical solutions provided by the embodiments of this application. As those skilled in the art will know, with the evolution of technology and the emergence of new application scenarios, the technical solutions provided by the embodiments of this application are also applicable to similar technical problems.

[0168] Those skilled in the art will understand that the technical solutions shown in the figures do not constitute a limitation on the embodiments of this application, and may include more or fewer steps than shown, or combine certain steps, or different steps.

[0169] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0170] Those skilled in the art will understand that all or some of the steps in the methods disclosed above, as well as the functional modules / units in the systems and devices, can be implemented as software, firmware, hardware, or suitable combinations thereof.

[0171] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms “comprising” and “having,” and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0172] It should be understood that in this application, "at least one (item)" means one or more, and "more than" means two or more. "And / or" is used to describe the relationship between related objects, indicating that three relationships can exist. For example, "A and / or B" can represent three cases: only A exists, only B exists, and both A and B exist simultaneously, where A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one (item) of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one (item) of a, b, or c can represent: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be single or multiple.

[0173] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of the units described above is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0174] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0175] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0176] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes multiple instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing programs, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0177] The preferred embodiments of the present application have been described above with reference to the accompanying drawings, but this does not limit the scope of the claims of the present application. Any modifications, equivalent substitutions, and improvements made by those skilled in the art without departing from the scope and substance of the embodiments of the present application shall be within the scope of the claims of the present application.

Claims

1. A visual function grading method based on multi-channel visual electrophysiological signals, characterized in that, The method includes: Acquire multi-channel visual electrophysiological signals from patients; Based on a preset classification model, the visual function level prediction process is performed on the multi-channel time-series signal corresponding to the multi-channel visual electrophysiological signal to obtain the visual function level prediction result of the patient. Based on the patient's basic clinical information, the predicted visual function level is post-processed to obtain the patient's final visual function level assessment result.

2. The method according to claim 1, characterized in that, The acquisition of the patient's multi-channel visual electrophysiological signals includes: Visual electrophysiological signals of the patient's left and right eyes were collected using multiple preset test modes to obtain the patient's multi-channel visual electrophysiological signals. The various testing modes include electroretinography (ERG) and visual evoked potential (VAP) testing modes.

3. The method according to claim 2, characterized in that, The electroretinogram test modes include a light-adapted single flash response test mode and a light-adapted flicker response test mode, and the visual evoked potential test modes include a flash visual evoked potential test mode. The process involves collecting visual electrophysiological signals from the patient's left and right eyes using multiple preset test modes, including: Using the light-adapted single-flash response test mode, with the patient's left and right eyes in a light-adapted state, the single-flash response of the retinal cone cells of the left and right eyes to a preset standard flash intensity is collected; the multi-channel visual electrophysiological signal includes the single-flash response of the patient's left and right eyes respectively; Using the light-adapted flicker response test mode, with the patient's left and right eyes in a light-adapted state, the changes in the light responses of the retinal cone cells of the left and right eyes to a preset high-frequency flicker light are collected; the multi-channel visual electrophysiological signals include the changes in the light responses of the patient's left and right eyes respectively. The visual cortical responses of the patient's left and right eyes induced by the flash visual evoked potential test mode were collected using the aforementioned flash visual evoked potential test mode. The multi-channel visual electrophysiological signals included the visual cortical responses of the patient's left and right eyes respectively.

4. The method according to claim 1, characterized in that, The visual function level prediction processing of the multi-channel temporal signals corresponding to the multi-channel visual electrophysiological signals based on the preset classification model includes at least one of the following: The visual function level is predicted by performing multi-channel time-series signals corresponding to the multi-channel visual electrophysiological signals based on clinical index regression using a preset classification model. Visual function level prediction is performed on the multi-channel time-series signals corresponding to the multi-channel visual electrophysiological signals using a preset classification model based on machine learning. The visual function level is predicted by using a pre-defined classification model based on a deep neural network to perform multi-channel time-series signals corresponding to the multi-channel visual electrophysiological signals.

5. The method according to claim 1, characterized in that, The multi-channel timing signal includes a multi-channel one-dimensional timing signal; Before performing visual function level prediction processing on the multi-channel temporal signals corresponding to the multi-channel visual electrophysiological signals based on the preset classification model, the method further includes at least one of the following: The multi-channel one-dimensional time-series signal is input into a preset classification model; The multi-channel one-dimensional time-series signal is mapped to a two-dimensional feature space to obtain a multi-channel two-dimensional time-series signal, and the multi-channel two-dimensional time-series signal is input into a preset classification model.

6. The method according to claim 1, characterized in that, Before performing visual function level prediction processing on the multi-channel temporal signals corresponding to the multi-channel visual electrophysiological signals based on the preset classification model, the method further includes: The multi-channel visual electrophysiological signals are standardized and preprocessed, and local temporal correspondence feature analysis is performed on the visual electrophysiological signals of the patient's left and right eyes under different test modes in the multi-channel visual electrophysiological signals to obtain the multi-channel temporal signals corresponding to the multi-channel visual electrophysiological signals.

7. The method according to any one of claims 1 to 6, characterized in that, The post-processing of the visual function level prediction results based on the patient's basic clinical information includes: The visual function level prediction results were fine-tuned based on the patient's basic clinical information; The reliability of the visual function level prediction results is assessed based on the patient's basic clinical information.

8. A visual function grading system based on multi-channel visual electrophysiological signals, characterized in that, The system includes: The data acquisition and preprocessing module is used to acquire multi-channel visual electrophysiological signals from patients; The feature modeling and prediction module is used to perform visual function level prediction processing on the multi-channel time-series signals corresponding to the multi-channel visual electrophysiological signals based on a preset classification model, so as to obtain the visual function level prediction result of the patient. The grading and visualization analysis module is used to perform data post-processing on the visual function level prediction results based on the patient's basic clinical information to obtain the patient's final visual function level assessment result.

9. A computer device, characterized in that, The computer device includes a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, it implements the visual function grading method based on multi-channel visual electrophysiological signals as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the visual function classification method based on multi-channel visual electrophysiological signals as described in any one of claims 1 to 7.

Images