Method, device and system for determining prediction parameters of abnormal eye movement

Abstract

The embodiments of the present application provide a method, device and system for determining predictive parameter items of abnormal eye movements, which relate to the field of brain function detection. The method includes: acquiring eye acquisition data; the eye acquisition data is information about the eye movements of multiple test subjects in response to a preset eye movement task, and the eye acquisition data includes a variety of eye movement parameter items and parameter values ​​corresponding to the eye movement parameter items; according to the multiple eye movement parameter items, the parameter values ​​corresponding to the eye movement parameter items and the design model, the predictive parameter items are determined from the multiple eye movement parameter items as independent predictive factors for early predetermined brain function diseases. The embodiments of the present application determine that eye movement abnormalities are related to early predetermined brain function diseases, and can determine key parameter items representing eye movement abnormalities as predictive parameter items, so that early prediction of predetermined brain function diseases can be achieved through eye movement behavior.

Description

Technical Field

[0001] The present application relates to the technical field of brain function detection, and in particular, to a method, device and system for determining prediction parameter items of abnormal eye movements. Background Art

[0002] At present, for some neurodegenerative diseases with motor symptoms as the main feature, early diagnosis and treatment can slow down the disease progression, reduce the damage to brain function, and improve the quality of life of patients to the greatest extent, such as Parkinson's disease (PD for short).

[0003] A large number of previous studies have pointed out that some brain functional diseases have different degrees of abnormalities when performing classic eye saccade tasks. However, most of these studies focus on patients with moderate to severe brain functional diseases (HY stage 2 and above), and few studies focus on early brain functional diseases. Summary of the invention

[0004] The embodiments of the present application provide a method, device and system for determining prediction parameter items of eye movement abnormalities, which are used to solve the technical problem of determining whether there are eye movement abnormalities in early brain function diseases or which eye movement parameter items can be used as prediction parameter items of eye movement abnormalities.

[0005] In a first aspect, an embodiment of the present application provides a method for determining a prediction parameter item of eye movement abnormality, comprising:

[0006] Acquire eye acquisition data; the eye acquisition data is information about the eye movements of multiple test subjects in response to a preset eye saccade task, and the eye acquisition data includes multiple eye saccade parameter items and parameter values ​​corresponding to the eye saccade parameter items;

[0007] According to a plurality of saccade parameter items, parameter values ​​corresponding to the saccade parameter items and a design model, a prediction parameter item is determined from the plurality of saccade parameter items as an independent predictive factor for early-stage predetermined brain function diseases; the design model includes at least one of a logistic regression model, a machine learning model and a neural network model.

[0008] In a possible implementation, the test subjects include a control group and an early disease group, the test subjects in the control group do not suffer from an early predetermined brain function disease, and the test subjects in the early disease group suffer from an early predetermined brain function disease;

[0009] According to the multiple saccadic parameter items, the parameter values ​​corresponding to the saccadic parameter items and the design model, a prediction parameter item is determined from the multiple saccadic parameter items, including:

[0010] Determining a prediction parameter item from a plurality of saccadic parameter items according to saccadic parameter items that meet predetermined conditions, parameter values ​​corresponding to the saccadic parameter items, and a design model;

[0011] Among them, meeting the predetermined conditions includes:

[0012] For each saccade parameter, there were significant differences between the parameter values ​​of multiple test subjects in the control group and the parameter values ​​of multiple test subjects in the early disease group;

[0013] There is no obvious collinearity between the parameter values ​​of each saccade parameter item.

[0014] In a possible implementation, the preset saccade task includes: a forward saccade task and / or a memory-guided saccade task;

[0015] The forward saccade task includes: displaying an initial fixation point at an initial position, detecting whether the test subject keeps fixating on the initial fixation point, stopping displaying the initial fixation point after a first predetermined time, and displaying a target point at a first predetermined position different from the initial position, detecting whether the test subject saccades to the first predetermined position within a second predetermined time;

[0016] The memory-guided saccade task includes: displaying an initial fixation point at an initial position and detecting whether the test subject is always fixating on the initial fixation point, displaying a target point at a first predetermined position different from the initial position and stopping displaying after a third predetermined time, controlling the initial fixation point to stop displaying after detecting that the test subject is fixating on the initial fixation point for a fourth predetermined time, and detecting whether the test subject saccades to the first predetermined position within a fifth predetermined time.

[0017] In a possible implementation, the saccade parameter item includes at least one of the following: the accuracy of the forward saccade, the reaction time of the forward saccade, the spatial error of the forward saccade, the saccade gain of the forward saccade, and the maximum saccade speed of the forward saccade;

[0018] And / or, the saccade parameter item includes at least one of the following: accuracy of memory-guided saccades, reaction time of memory-guided saccades, spatial error of memory-guided saccades, saccade gain of memory-guided saccades, and maximum saccade speed of memory-guided saccades;

[0019] Among them, the accuracy is used to indicate the accuracy of saccades, the reaction time is used to indicate the time interval from the disappearance of the initial fixation point to the start of the main saccade, the spatial error is used to indicate the straight-line distance from the saccade landing point to the target point, the saccade gain is used to indicate the ratio of the straight-line distance from the saccade starting point to the saccade target point to the straight-line distance from the saccade starting point to the target point, and the maximum saccade speed is used to indicate the peak speed of the saccade during the saccade process.

[0020] In one possible implementation, the design model is a binary logistic regression model;

[0021] According to the multiple saccadic parameter items, the parameter values ​​corresponding to the saccadic parameter items and the design model, a prediction parameter item is determined from the multiple saccadic parameter items, including:

[0022] Normalize the parameter values ​​corresponding to each saccade parameter item to obtain the data to be input;

[0023] Input the data to be input into the binary logistic regression model to obtain a candidate prediction model;

[0024] Determining the relevance of candidate prediction models for the prediction of early stage predetermined brain function disorders;

[0025] If the correlation between the candidate prediction model and the prediction of the early predetermined brain function disease exceeds the designed threshold, the eye movement parameter item used by the candidate prediction model is used as the prediction parameter item.

[0026] In a possible implementation, the prediction parameter item includes at least one of the following:

[0027] Maximum saccade velocity of memory-guided saccades, reaction time of memory-guided saccades, maximum saccade velocity of forward saccades, and spatial error of memory-guided saccades.

[0028] In one possible implementation, determining the relevance of a candidate prediction model to the prediction of an early stage predetermined brain function disease includes:

[0029] According to the parameter values ​​corresponding to the prediction parameter items, ROC analysis is performed on the candidate prediction model to obtain the area under the curve;

[0030] If the area under the curve is greater than a fourth predetermined value, it is determined that the correlation between the candidate prediction model and the prediction of the early predetermined brain function disease exceeds a designed threshold.

[0031] In a second aspect, an embodiment of the present application provides a method for constructing a prediction model, comprising:

[0032] The prediction model is determined according to the design model and the prediction parameter items of the eye movement abnormality; the prediction parameter items are obtained based on the method for determining the prediction parameter items of the eye movement abnormality of the first aspect.

[0033] In a third aspect, an embodiment of the present application provides a device for determining a prediction parameter item of eye movement abnormality, using the method for determining a prediction parameter item of eye movement abnormality of the first aspect, the determination device includes:

[0034] An acquisition module is used to acquire eye acquisition data; the eye acquisition data is information about the eye movements of multiple test subjects in response to a preset eye saccade task, and the eye acquisition data includes multiple eye saccade parameter items and parameter values ​​corresponding to the eye saccade parameter items;

[0035] A determination module is used to determine a prediction parameter item from a plurality of saccadic parameter items according to a plurality of saccadic parameter items, parameter values ​​corresponding to the saccadic parameter items and a design model, as an independent predictive factor for early-stage predetermined brain function diseases; the design model includes at least one of a logistic regression model, a machine learning model and a neural network model.

[0036] In a fourth aspect, an embodiment of the present application provides a device for constructing a prediction model, including:

[0037] A construction module is used to determine a prediction model based on a design model and prediction parameter items of eye movement abnormalities; the prediction parameter items are obtained based on the method for determining the prediction parameter items of eye movement abnormalities in the first aspect.

[0038] In a fifth aspect, an embodiment of the present application provides a brain function detection system, including: an eye movement acquisition device and a processor electrically connected;

[0039] An eye movement acquisition device, used to sample the eye movement of the test subject in response to a preset eye saccade task to obtain eye acquisition data;

[0040] The processor is used to execute the steps of the method for determining the prediction parameter items of eye movement abnormalities in the first aspect; and / or, execute the steps of the method for constructing the prediction model in the second aspect.

[0041] In a possible implementation, the processor is further used to input parameter values ​​corresponding to the prediction parameter items of the target object into the prediction model to obtain a prediction result; the prediction result is used to indicate whether the target object suffers from an early-stage predetermined brain function disease.

[0042] In a possible implementation, the predetermined brain function disease includes any of the following diseases:

[0043] Alzheimer's disease of varying severity, amyotrophic lateral sclerosis, Huntington's disease, spinocerebellar atrophy, multiple sclerosis, Parkinson's disease.

[0044] In a sixth aspect, an embodiment of the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method for determining prediction parameter items for eye movement abnormalities of the first aspect; and / or, when executed, implements the steps of the method for constructing a prediction model of the second aspect.

[0045] The beneficial effects of the technical solution provided by the embodiment of the present application are:

[0046] The method for determining the prediction parameter items of eye movement abnormality provided in the embodiment of the present application can obtain the information of the eye movement of multiple test subjects in response to the preset eye movement task, and determine the prediction parameter items from the multiple eye movement parameter items according to the multiple eye movement parameter items, the parameter values ​​corresponding to the eye movement parameter items and the design model, as an independent prediction factor for the early predetermined brain function disease. Since each of these prediction parameter items has an independent strong correlation with the prediction of the early predetermined brain function disease, the embodiment of the present application determines that the eye movement abnormality is related to the early predetermined brain function disease, and determines the key parameter items representing the eye movement abnormality as the prediction parameter items, so that the early predetermined brain function disease can be predicted to a certain extent through the eye movement behavior. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in describing the embodiments of the present application are briefly introduced below.

[0048] Figure 1 A flowchart of a method for determining prediction parameter items of abnormal eye movements is provided for an embodiment of the present application;

[0049] Figure 2 A schematic diagram of the principle of a single test process of a forward saccade task provided in an embodiment of the present application;

[0050] Figure 3 A schematic diagram of the principle of a single test process of a memory-guided saccade task provided in an embodiment of the present application;

[0051] Figure 4 A schematic diagram showing comparison of various saccade parameters between early PD patients and a control group in a forward saccade task provided in an embodiment of the present application;

[0052] Figure 5 A schematic diagram showing comparison of various saccade parameters between early PD patients and a control group in a memory-guided saccade task provided in an embodiment of the present application;

[0053] Figure 6 A schematic diagram of a ROC analysis for distinguishing an early PD group from a control group according to a prediction parameter item determined by a binary logistic regression model provided in an embodiment of the present application;

[0054] Figure 7 A schematic diagram of the framework structure of a device for determining prediction parameters of abnormal eye movements provided in an embodiment of the present application;

[0055] Figure 8 A schematic diagram of the framework structure of a prediction model construction device provided in an embodiment of the present application;

[0056] Fig. 9A schematic diagram of the structure of a brain function detection system provided in an embodiment of the present application. DETAILED DESCRIPTION

[0057] The embodiments of the present application are described below in conjunction with the drawings in the present application. It should be understood that the implementation methods described below in conjunction with the drawings are exemplary descriptions for explaining the technical solutions of the embodiments of the present application and do not constitute a limitation on the technical solutions of the embodiments of the present application.

[0058] It will be understood by those skilled in the art that, unless specifically stated, the singular forms "one", "said", and "the" used herein may also include plural forms. It should be further understood that the terms "including" and "comprising" used in the embodiments of the present application refer to that the corresponding features can be implemented as the presented features, information, data, steps, operations, elements and / or components, but do not exclude the implementation as other features, information, data, steps, operations, elements, components and / or combinations thereof supported by the technical field. It should be understood that when we say that an element is "connected" or "coupled" to another element, the one element may be directly connected or coupled to the other element, or it may refer to that the one element and the other element establish a connection relationship through an intermediate element. In addition, the "connection" or "coupling" used herein may include wireless connection or wireless coupling. The term "and / or" used herein indicates at least one of the items defined by the term, for example, "A and / or B" indicates that it is implemented as "A", or is implemented as "A", or is implemented as "A and B".

[0059] Studies have found that PD is a neurodegenerative disease with motor symptoms as the main feature. Early diagnosis and treatment can slow down the disease process, reduce brain function damage, and improve the patient's quality of life to the greatest extent (Chetelat, 2018). A large number of previous studies have pointed out that PD patients have different degrees of abnormalities when performing classic eye saccade tasks. However, most of these studies focus on moderate to severe PD patients (HY stage 2 and above), and few studies focus on early PD. Therefore, it is unclear whether early PD patients have eye saccade abnormalities. The first goal of this application is to explore whether there are eye saccade abnormalities in early PD. If so, can the prediction model constructed based on these eye saccade parameters assist in the early diagnosis of PD.

[0060] In order to make the objectives, technical solutions and advantages of the present application clearer, the implementation methods of the present application will be further described in detail below with reference to the accompanying drawings.

[0061] See also Figure 1 As shown, an embodiment of the present application provides a method for determining prediction parameter items of abnormal eye movements, and the method includes: step S101 to step S102.

[0062] S101, acquiring eye acquisition data; the eye acquisition data is information about the eye movements of multiple test subjects in response to a preset saccade task, and the eye acquisition data includes multiple saccade parameter items and parameter values ​​corresponding to the saccade parameter items.

[0063] Optionally, the test subjects may include normal subjects and subjects with early predetermined brain function diseases. The test subjects may be mainly elderly people, and the saccade task may select at least one basic classical task as needed, such as: forward saccade task, reverse saccade task, memory-guided saccade task, two-step saccade task, etc. The saccade parameter items are some parameter items representing the eye movement function obtained based on the saccade task.

[0064] S102. Determine a prediction parameter item from the multiple saccadic parameter items according to the multiple saccadic parameter items, the parameter values ​​corresponding to the saccadic parameter items and the design model as an independent predictive factor for early-stage predetermined brain function diseases; the design model includes at least one of a logistic regression model, a machine learning model and a neural network model.

[0065] Optionally, the prediction parameter items may include at least one eye saccade parameter item, each of which may be an independent predictive factor for early-stage predicted brain function diseases, and the independent predictive factor may independently predict early-stage predicted brain function diseases.

[0066] Based on the above content, the method for determining the prediction parameter items of eye movement abnormalities provided in the embodiment of the present application can obtain the information of the eye movements of multiple test subjects in response to the preset eye movement tasks, and determine the prediction parameter items from the multiple eye movement parameter items according to the multiple eye movement parameter items, the parameter values ​​corresponding to the eye movement parameter items and the design model, as an independent prediction factor for the early predetermined brain function disease. Since each of these prediction parameter items has an independent strong correlation with the prediction of the early predetermined brain function disease, the embodiment of the present application determines that eye movement abnormalities are related to the early predetermined brain function disease, and determines the key parameter items representing eye movement abnormalities as prediction parameter items, so that the early predetermined brain function disease can be predicted to a certain extent through eye movement behavior.

[0067] In some embodiments, the preset saccade task includes: a forward saccade task and / or a memory-guided saccade task.

[0068] The forward saccade task includes: displaying an initial fixation point at an initial position, detecting whether the test subject keeps fixating on the initial fixation point, stopping displaying the initial fixation point after a first predetermined time, and displaying a target point at a first predetermined position different from the initial position, detecting whether the test subject saccades to the first predetermined position within a second predetermined time.

[0069] The memory-guided saccade task includes: displaying an initial fixation point at an initial position and detecting whether the test subject is always fixating on the initial fixation point, displaying a target point at a first predetermined position different from the initial position and stopping displaying after a third predetermined time, controlling the initial fixation point to stop displaying after detecting that the test subject is fixating on the initial fixation point for a fourth predetermined time, and detecting whether the test subject saccades to the first predetermined position within a fifth predetermined time.

[0070] Specifically, the eye saccade to the first predetermined position refers to the test subject's gaze point jumping to the first predetermined position of the display screen.

[0071] The embodiment of the present application may use a forward saccade task and a memory-guided saccade task to perform saccade tests on the test subject.

[0072] In some embodiments, the test subjects include a control group and an early disease group, wherein the test subjects in the control group do not suffer from an early predetermined brain function disease, and the test subjects in the early disease group suffer from an early predetermined brain function disease.

[0073] According to the multiple saccadic parameter items, the parameter values ​​corresponding to the saccadic parameter items and the design model, a prediction parameter item is determined from the multiple saccadic parameter items, including:

[0074] Determining a prediction parameter item from a plurality of saccadic parameter items according to saccadic parameter items that meet predetermined conditions, parameter values ​​corresponding to the saccadic parameter items, and a design model;

[0075] Wherein, satisfying the predetermined condition includes: there is a significant difference between the parameter values ​​of the multiple test subjects in the control group and the parameter values ​​of the multiple test subjects in the early disease group for each saccade parameter item;

[0076] There is no obvious collinearity between the parameter values ​​of each saccade parameter item.

[0077] Optionally, there are significant differences between the parameter values ​​of multiple test subjects in the control group of each saccade parameter item and the parameter values ​​of multiple test subjects in the early disease group, including: performing a rank sum test on each saccade parameter item to determine the test value; if the test value is less than a first predetermined value, then there is a significant difference in the saccade parameter item between the control group and the early disease group.

[0078] Optionally, there is no obvious collinearity between the parameter values ​​of each saccade parameter item, including: performing collinearity analysis on each saccade parameter item to determine a tolerance and / or an expansion factor; if the tolerance is greater than a second predetermined value and / or the expansion factor is less than a third predetermined value, then there is no obvious collinearity between the saccade parameter items.

[0079] Specifically, the first predetermined value may be 0.05, ie, the test value p<0.05, the second predetermined value may be 0.1, and the third predetermined value may be 10, ie, the tolerance>0.1 and the expansion factor<10.

[0080] Among them, p < 0.05 indicates that there is a significant difference between the eye saccadic parameter items in the control group and the early disease group; tolerance > 0.1 and / or expansion factor < 10 indicate that there is no obvious collinearity between the eye saccadic parameter items.

[0081] As an example, each saccade parameter item can be subjected to a first round of screening, followed by a ranksum test, and the saccade parameter items with p < 0.05 can be subjected to a second round of screening. The saccade parameter items corresponding to tolerance > 0.1 and expansion factor < 10 can be considered as predictive factors for the logistic regression model.

[0082] In some embodiments, the saccade parameter items include at least one of the following: the accuracy of forward saccades, the reaction time of forward saccades, the spatial error of forward saccades, the saccade gain of forward saccades, and the maximum saccade speed of forward saccades; and / or, the saccade parameter items include at least one of the following: the accuracy of memory-guided saccades, the reaction time of memory-guided saccades, the spatial error of memory-guided saccades, the saccade gain of memory-guided saccades, and the maximum saccade speed of memory-guided saccades.

[0083] Among them, the accuracy is used to indicate the accuracy of saccades, the reaction time is used to indicate the time interval from the disappearance of the initial fixation point to the primary saccade, the spatial error is used to indicate the straight-line distance from the saccade landing point to the target point, the saccade gain is used to indicate the ratio of the straight-line distance from the saccade starting point to the saccade target point to the straight-line distance from the saccade starting point to the target point, and the maximum saccade velocity is used to indicate the peak velocity of the saccade during the saccade process.

[0084] Optionally, the saccade starting point corresponds to the test subject's current gaze point on the display screen, the saccade target point corresponds to the test subject's actual saccade to the gaze point on the display screen, and the target target point is the target stimulus point displayed on the display screen.

[0085] Optionally, the accuracy rate includes: 1. The direction of the saccade is correct; 2. The spatial error of the saccade is within the allowable error range; 3. The start time of the saccade is within the allowable time window; 4. There are no errors such as "fixation point fixation failure", "trial abandonment", "target point fixation failure" and so on.

[0086] Optionally, the maximum saccade speed includes: calculating the instantaneous speed of each sampling point (time interval is 1 ms), taking 5 sampling points as a window, performing data smoothing, and finding the maximum value of the instantaneous speed after smoothing in this saccade, which is the peak speed.

[0087] Taking the prediction of early PD as an example, this application has found through research that PD is a neurodegenerative disease with motor symptoms as the main feature. Early diagnosis and treatment can slow down the disease progression, reduce brain function damage, and improve the patient's quality of life to the greatest extent. A large number of previous studies have pointed out that PD patients have varying degrees of abnormalities when performing classic saccade tasks. However, most of these studies focus on moderate to severe PD patients (HY stage 2 and above), and few studies focus on early PD. Therefore, it is unclear whether early PD patients have saccade abnormalities. The first goal of this application is to explore whether there are saccade abnormalities in early PD. If so, the logistic regression model constructed based on these abnormal saccade parameters assists in the early diagnosis of PD.

[0088] During the study, this application included 2 groups of subjects, 39 elderly control group (control) and 39 early elderly PD group (early PD). This experiment used GPower (version number 3.1.9.7) software to determine the sample size, effect size = 0.4, α = 0.05, β = 0.1, and the test power was 1-β = 0.943. The demographic data of the subjects are shown in Table 1.

[0089] Table 1:

[0090]

[0091] Among them, early PD patients are elderly people with Hoehn-Yahr stage 1-1.5 (including stage 1 and stage 1.5), over 60 years old, and without other chronic diseases such as psychiatric, neurological, and metabolic diseases. This group of subjects was recruited from the outpatient department of neurology of the General Hospital of the People's Liberation Army; the elderly control group is elderly people who match the early PD group in age, gender, and education level and do not have other chronic diseases such as psychiatric, neurological, and metabolic diseases. This group of subjects was recruited from the community.

[0092] This experiment used two classic saccade tasks, the Pro Saccade task (PS) and the Memory Guided Saccade task (MGS). The PS task is a classic reflexive saccade task, and the MGS task is a classic autonomous saccade task.

[0093] The specific test process is as follows: Figure 2 As shown in FIG, a schematic diagram of the principle of a single test process of a forward saccade task is shown. Figure 2As shown in the figure, FP on means FP appears, Target on means the target appears, Saccade and holdtarget means keeping gaze on the target, 300ms means keeping gaze on the target for 300ms, Target off (trialend) means the target disappears and the test ends.

[0094] See also Figure 2 As shown in the figure, in the PS task, at the beginning of each trial, a fixation point (FP) appears in the center of the screen, which is the initial fixation point. The fixation point is a white cross with a size of 1 visual angle horizontally × 1 visual angle vertically. When FP appears, we ask the subjects to look at the stimulus as quickly as possible. If the subject's eyes enter the check window with a radius of 4 visual angles and stay in the window for 800 milliseconds (ms), the FP will disappear. At the same time, a white solid stimulus point (Stimulus, Sti), that is, the target point, will randomly appear in one of the four peripheral positions of the top, bottom, left, and right. The size of the stimulus point is 1 visual angle and it is 10 visual angles away from the center of the screen. When the target point appears, the subject needs to look at the point as accurately and quickly as possible and keep fixating. At this time, if the subject's eyes enter the check window with a radius of 4 visual angles and stay in the window for 300ms, the target stimulus point disappears and the trial ends. If the subject still does not look at the FP after 1000ms after the FP appears, or does not make a saccade after 2000ms after the target stimulus point appears, the current trial is terminated. After each trial, there is an 800-ms inter-trial interval (ITI), during which no visual stimulus is displayed on the screen. Among them, SRT (Saccade Reaction Time) is the saccade reaction time.

[0095] See also Figure 3 As shown in FIG, a schematic diagram showing the principle of a single test process of a memory-guided eye-saccade task is shown. Figure 3 As shown in the figure, FP on means the FP appears, Cue on means the target appears, Cue off means the target disappears, FP off (go signal) means the FP disappears, Saccade and hold means to keep gazing at the position where the target disappears, and Trialend means the end of the test.

[0096] See also Figure 3As shown in the figure, in the MGS task, at the beginning of each trial, a FP appears in the center of the screen. The physical properties of the FP are the same as those in the PS task. When the FP appears, the subject is required to look at the stimulus as quickly as possible. If the subject's eyes enter the inspection window with a radius of 4 visual angles and stay for 800 ms, a white solid cue point cue, that is, the target point, will randomly appear in one of the four peripheral positions of the top, bottom, left, and right. The size of the cue point is 1 visual angle, and it is 10 visual angles away from the center of the screen. The cue point is presented for 500ms. When the cue point appears, the subject needs to continue to look at the FP point and use the peripheral vision of the eye to see and remember the position of the cue point. After 500ms, the cue point will disappear, followed by a delay period, the length of which is 1000 ms. During the delay period, the subject still needs to keep looking at the FP. After the delay period, the FP point will disappear. At this time, the subject needs to look at the position where the cue point appeared as accurately and quickly as possible. If the subject's eyes enter and stay in the detection window of the cue point (radius of 4 visual angles) and maintain fixation for 300 ms, the trial ends. If the subject still does not look at the FP after 1000 ms after the FP appears, the current trial ends. Similarly, if the subject still does not make an eye saccade 2000 ms after the FP disappears, the current trial ends. If the subject moves the eye position out of the inspection window of the FP before the FP disappears, the current trial ends. After each trial, there is an ITI of 800ms, during which no visual stimulus is displayed on the screen. Among them, SRT (saccade reaction time) is the eye saccade reaction time.

[0097] As an example, the present application adopts an adaptive speed threshold method to determine eye saccades.

[0098] The first step is to find the possible starting and ending points of saccades with an initial speed of 15 visual degrees per second, and divide the eye movement trajectory into possible saccade segments and non-saccade segments.

[0099] In the second step, the average speed of the non-saccade segment adjacent to the starting point of each saccade plus 2.58 times the standard deviation of the average speed is used as the new speed threshold, and the starting point of the saccade is re-searched. If there is no segment between the original starting point and the end point with a speed stably greater than the new speed threshold, the saccade is deleted.

[0100] In the third step, if the saccade is not deleted in the second step, the average speed of the non-saccade segment adjacent to the end point of each saccade plus 2.58 times the standard deviation of the average speed is used as the new speed threshold to re-find the end point of the saccade.

[0101] Finally, repeat steps 1 to 3 to find and record the new starting and ending points of all saccades.

[0102] Among them, the calculation of reaction time, spatial error and saccade gain: reaction time is the time interval from the disappearance of FP to the start of the first saccade. The first saccade is the first step of the saccade with an amplitude greater than 1 degree (in order to exclude microsaccades) and a direction within 90 degrees toward the saccade target position. The spatial error of the first saccade is the straight-line distance from the first saccade landing point to the target point. The saccade gain is the ratio of the straight-line distance from the saccade start point to the saccade end point to the straight-line distance from the saccade start point to the target point. For each experimental session, the above parameters are calculated only when the number of correct trials is greater than or equal to 10%.

[0103] Determination of trial errors and error types: Trial errors are divided into the following four categories: 1. Abandoned trials refer to trials in which the subject did not move his eyes to the FP detection window within 1000ms after the FP appeared; 2. Missing trials refer to trials in which the subject did not make a saccade within 2000ms after the FP disappeared; 3. Spatial error trials refer to trials in which the distance from the main saccade landing point to the target position was greater than the average spatial error of the control group plus 1.96 times the standard deviation of the average spatial error; 4. Unsuppressed reflexive saccade trials refer to trials in the MGS task in which the subject did not maintain fixation on the FP before the disappearance of the FP, during the appearance of the cue, and after the disappearance of the cue.

[0104] Based on the above test content, this application compares and analyzes the eye movement parameters of early PD patients and the control group.

[0105] See also Figure 4 As shown in FIG, a schematic diagram of the comparison of various saccade parameters between early PD patients (PD patients) and the control group (control) in a forward saccade task is shown. Figure 4 As shown, A, B, C, D, and E correspond to the accuracy of positive saccades, the reaction time of positive saccades, the spatial error of positive saccades, the saccade gain of positive saccades, and the maximum saccade speed of positive saccades, respectively. Among them, the p value is calculated by the rank sum test, and the error bar is the standard error of the mean (Standard Error of Mean, SEM). The detection value P of the rank sum test can be divided into three categories, namely: *p<0.05, **p<0.01, ***p<0.001.

[0106] See also Figure 5 As shown in FIG, a schematic diagram of comparing various saccade parameters between early PD patients and the control group in a memory-guided saccade task is shown. Figure 5 As shown, A, B, C, D, and E correspond to the accuracy of memory-guided saccades, the reaction time of memory-guided saccades, the spatial error of memory-guided saccades, the saccade gain of memory-guided saccades, and the maximum saccade speed of memory-guided saccades, respectively. Among them, the detection value P of the rank sum test can be divided into three categories, namely: the p value is calculated by the rank sum test, and the error bar is the standard error of mean (SEM). *p<0.05, **p<0.01, ***p<0.001.

[0107] In PS tasks, such as Figure 4 As shown in the figure, compared with the control group, the early PD group had significant differences in all parameters. The differences were mainly manifested in: the PD group had low accuracy (p<0.01), reaction time (p<0.001), large spatial error (p<0.01), small saccadic gain (p<0.01) and small maximum saccadic velocity (p<0.01); in the MGS task, Figure 5 As shown in the figure, compared with the control group, the early PD group had significant differences in all parameters. The differences were mainly manifested in: the PD group had low accuracy (p<0.001), long reaction time (p<0.001), large spatial error (p<0.001), small saccade gain (p<0.01) and small maximum speed (p<0.001). Although there were differences in the eye movement parameters of the two tasks, the differences in the various parameters of the memory-guided saccade task MGS were greater.

[0108] Therefore, based on the above test content and test results, we can know that: 1. Early PD has significant saccade abnormalities in PS tasks and MGS tasks; 2. Early PD patients also have significant saccade impairment in the PS task, a reflex saccade task with less cognitive function involvement.

[0109] In some embodiments, the designed model is a binary logistic regression model.

[0110] According to a plurality of saccade parameter items, parameter values ​​corresponding to the saccade parameter items and a design model, a prediction parameter item is determined from the plurality of saccade parameter items, including: normalizing the parameter values ​​corresponding to each saccade parameter item to obtain data to be input; inputting the data to be input into a binary logistic regression model to obtain a candidate prediction model; determining the correlation between the candidate prediction model and the prediction of an early predetermined brain function disease; if the correlation between the candidate prediction model and the prediction of an early predetermined brain function disease exceeds a design threshold, using the saccade parameter item used by the candidate prediction model as the prediction parameter item.

[0111] Specifically, the parameter values ​​corresponding to each saccade parameter item are normalized respectively, and all parameter values ​​of each saccade parameter item can be distributed in [0, 1]. For example, when the reaction time is between 100ms and 500ms, 100ms can be corresponded to 0, and 500ms can be corresponded to 1. The parameter values ​​of the reaction time of memory-guided saccades and the reaction time of forward saccades are mapped to the interval [0, 1] to obtain the data to be processed corresponding to the reaction time. Then, according to each saccade parameter item, the data to be processed are input into a binary logistic regression model to obtain a candidate prediction model.

[0112] Optionally, the candidate prediction model may also be a logistic regression model, for example, a regression equation, including various prediction parameter items and corresponding regression coefficients.

[0113] In some embodiments, the prediction parameter items include at least one of the following: maximum eye speed of memory-guided saccades, reaction time of memory-guided saccades, maximum eye speed of forward saccades, and spatial error of memory-guided saccades.

[0114] Optionally, any one of the maximum saccade velocity of memory-guided saccades, the reaction time of memory-guided saccades, the maximum saccade velocity of forward saccades, and the spatial error of memory-guided saccades can serve as independent predictors of early-stage predetermined brain function diseases.

[0115] During the research process, in order to further improve the predictive effectiveness of saccade parameter items in early PD, this application constructs a prediction model through a binary logistic regression model. Before the prediction model is constructed, the saccade parameter items to be included in the prediction model must first be confirmed. This application incorporates all saccade parameter items in the PS task and the MGS task that are significantly different between the early PD group and the control group into the binary logistic regression model. The parameters to be included are shown in Table 2.

[0116] Table 2: Saccadic parameters for early PD diagnosis to be included in the binary logistic regression model analysis

[0117]

[0118] As shown in Table 2 , the p values ​​were calculated by the rank sum test. The accuracy of the 9 early PD patients when performing memory saccades was too low, resulting in the number of correct trials being less than 5. Therefore, they could not be included in the analysis of reaction time, accuracy, saccade gain, and maximum saccade velocity of memory-guided saccades.

[0119] Based on the contents of Table 2, according to the test value P < 0.05, the following 10 saccade parameter items can be included in the logistic regression model: the accuracy of forward saccades, the reaction time of forward saccades, the spatial error of forward saccades, the saccade gain of forward saccades, the maximum saccade velocity of forward saccades, the accuracy of memory-guided saccades, the reaction time of memory-guided saccades, the spatial error of memory-guided saccades, the saccade gain of memory-guided saccades, and the maximum saccade velocity of memory-guided saccades.

[0120] Furthermore, the present application uses two parameters, tolerance and expansion factor, to evaluate whether there is obvious collinearity in each saccade parameter item to be included in the logistic regression model, as shown in Table 3.

[0121] Table 3: Collinearity analysis of saccade parameters to be included in the binary logistic regression model

[0122]

[0123] As shown in Table 3, if the tolerance of the predictor is > 0.1 and the inflation factor is < 10, it is considered that there is no obvious collinearity. Then, the following 10 saccade parameter items can be included in the logistic regression model: the accuracy of forward saccades, the reaction time of forward saccades, the spatial error of forward saccades, the saccade gain of forward saccades, the maximum saccade speed of forward saccades, the accuracy of memory-guided saccades, the reaction time of memory-guided saccades, the spatial error of memory-guided saccades, the saccade gain of memory-guided saccades, and the maximum saccade speed of memory-guided saccades.

[0124] The analysis results show that there is no obvious collinearity between the eye saccade parameter items. In order to make the coefficients of the binary logistic regression model comparable, the present application standardizes the eye saccade parameter items to be included in the binary logistic regression model, so that all parameter values ​​of each eye saccade parameter item are distributed in [0, 1]. Finally, the present application determines the prediction model through the conditional entry method, and the prediction model is shown in Table 4.

[0125] Table 4: Related eye-saccade parameters and prediction models for early PD diagnosis based on binary logistic regression model analysis

[0126]

[0127] As shown in Table 4, B represents regression coefficient; SE represents standard error of regression coefficient; Wald represents Wald test value of regression coefficient; OR represents odds ratio; 95% CI represents 95% confidence interval of odds ratio; p value represents p value of Wald test of regression coefficient.

[0128] As shown in Table 4, according to the regression coefficient from large to small, the predictive parameters of early PD are the maximum speed in MGS (B = -2.083, OR = 0.125); reaction time in MGS (B = 2.018, OR = 7.521); maximum speed in PS (B = -1.479, OR = 6.152) and spatial error in MGS (B = 1.269, OR = 3.556). The predictive parameters are mainly the maximum speed, spatial error and reaction time that directly reflect the eye movement function.

[0129] Optionally, the prediction model can be constructed based on each prediction parameter item and the regression coefficient corresponding to the prediction parameter item, and through repeated iterations and verifications, a final target prediction model can be established, and the target prediction model can effectively predict early PD.

[0130] In some embodiments, determining the correlation between the candidate prediction model and the prediction of early predetermined brain function diseases also includes: performing ROC analysis on the candidate prediction model according to the parameter values ​​corresponding to the prediction parameter items to obtain the area under the curve; if the area under the curve is greater than a fourth predetermined value, it is determined that the correlation between the candidate prediction model and the prediction of early predetermined brain function diseases exceeds a designed threshold.

[0131] Optionally, the fourth predetermined value may be 0.75.

[0132] See also Figure 6 , a schematic diagram of ROC analysis for distinguishing the early PD group from the control group based on the prediction parameter items determined by the binary logistic regression model is shown. AUC stands for area under the curve; p-value is based on the Z test; sensitivity stands for sensitivity, and 100-Specificity stands for specificity.

[0133] like Figure 6 As shown, the present application uses ROC analysis to analyze the prediction model and obtains an area under the curve AUC = 0.852. In practice, if the area under the curve AUC>0.75, it has diagnostic efficacy, and it can be determined that the correlation between the prediction model and the prediction of early PD exceeds the design threshold.

[0134] This application can better assist the prediction of early PD through the binary logistic regression model (AUC = 0.852, zstatistic = 6.829, p (Area = 0.5) < 0.0001). Therefore, the prediction model constructed based on PS and MGS tasks can better assist the early prediction of PD.

[0135] In summary, this application found through research that early PD patients have obvious eye movement abnormalities in both PS and MGS tasks, and the binary logistic regression model constructed based on traditional eye movement parameter items in PS and MGS tasks has a high diagnostic efficacy.

[0136] Based on the same inventive concept, the present application embodiment provides a method for constructing a prediction model, including:

[0137] A prediction model is determined based on the design model and the prediction parameter items of eye movement abnormalities; the prediction parameter items are obtained based on the method for determining the prediction parameter items of eye movement abnormalities in the embodiment of the present application.

[0138] The prediction model of the embodiment of the present application can predict whether the target object suffers from an early-stage predicted brain function disease. The parameter values ​​corresponding to the predicted parameter items of the target object are input into the prediction model to obtain the prediction results. According to the prediction results, it can be known whether the target object suffers from an early-stage predicted brain function disease.

[0139] Based on the same inventive concept, the present application embodiment provides a device for determining a prediction parameter item of eye movement abnormality, and adopts the method for determining a prediction parameter item of eye movement abnormality of the present application embodiment. Figure 7 As shown, a schematic diagram of the framework structure of a device for determining prediction parameter items of eye movement abnormalities is shown. The device 70 for determining prediction parameter items of eye movement abnormalities includes: an acquisition module 701 and a determination module 702.

[0140] The acquisition module 701 is used to acquire eye acquisition data; the eye acquisition data is information about the eye movements of multiple test subjects in response to a preset eye saccade task, and the eye acquisition data includes multiple eye saccade parameter items and parameter values ​​corresponding to the eye saccade parameter items.

[0141] The determination module 702 is used to determine the prediction parameter item from the multiple saccadic parameter items according to the multiple saccadic parameter items, the parameter values ​​corresponding to the saccadic parameter items and the design model, as an independent predictive factor for early-stage predetermined brain function diseases; the design model includes at least one of a logistic regression model, a machine learning model and a neural network model.

[0142] Optionally, the determination module 702 is also used to normalize the parameter values ​​corresponding to each eye-saccade parameter item to obtain the data to be input; input the data to be input into a binary logistic regression model to obtain a candidate prediction model; determine the correlation between the candidate prediction model and the prediction of early predicted brain function diseases; if the correlation between the candidate prediction model and the prediction of early predicted brain function diseases exceeds a designed threshold, the eye-saccade parameter item used by the candidate prediction model is used as the prediction parameter item.

[0143] Optionally, the determination module 702 is also used to perform ROC analysis on the candidate prediction model according to the parameter value corresponding to the prediction parameter item to obtain the area under the curve; if the area under the curve is greater than the fourth predetermined value, it is determined that the correlation between the candidate prediction model and the prediction of the early predetermined brain function disease exceeds the design threshold.

[0144] The device for determining the prediction parameter items of eye movement abnormalities in the embodiment of the present application can execute the method for determining the prediction parameter items of eye movement abnormalities provided in the embodiment of the present application. The implementation principles are similar. The actions performed by each module in the device of each embodiment of the present application correspond to the steps in the method of each embodiment of the present application. For the detailed functional description of each module of the device, please refer to the description in the corresponding method shown in the previous text, which will not be repeated here.

[0145] Based on the same inventive concept, see Figure 8 , a schematic diagram of a structure of a prediction model construction device is shown. The prediction model construction device 80 includes: a construction module 801.

[0146] Construction module 801 is used to determine the prediction model according to the design model and the prediction parameter items of eye movement abnormality; the prediction parameter items are obtained based on the method for determining the prediction parameter items of eye movement abnormality in the embodiment of the present application.

[0147] The prediction model construction device of the embodiment of the present application can execute the prediction model construction method provided by the embodiment of the present application, and the implementation principle is similar. The actions performed by each module in the device of each embodiment of the present application correspond to the steps in the method of each embodiment of the present application. For the detailed functional description of each module of the device, please refer to the description in the corresponding method shown in the previous text, which will not be repeated here.

[0148] Based on the same inventive concept, see Fig. 9 , which shows a schematic diagram of the structure of a brain function detection system. The brain function detection system 90 includes: an eye movement acquisition device 901 and a processor 902.

[0149] The eye movement acquisition device 901 is used to sample the eye movement of the test subject in response to a preset eye saccade task to obtain eye acquisition data.

[0150] Processor 902 is used to execute the steps of the method for determining the prediction parameter items of eye movement abnormalities in the embodiment of the present application; and / or, execute the steps of the method for constructing the prediction model in the embodiment of the present application.

[0151] Applied to the embodiment of the present application, at least the following technical effects can be achieved: the brain function detection system 90 of the embodiment of the present application can obtain the information of the eye movements of multiple test subjects in response to the preset eye movement task, and then determine the prediction parameter items from the multiple eye movement parameter items according to the multiple eye movement parameter items, the parameter values ​​corresponding to the eye movement parameter items and the logistic regression model. Since each of these prediction parameter items has an independent strong correlation with the prediction of the early predetermined brain function disease, the brain function detection system 90 of the embodiment of the present application determines that the eye movement abnormality is related to the early predetermined brain function disease, and determines the key parameter items representing the eye movement abnormality as the prediction parameter items, so that the early predetermined brain function disease can be predicted to a certain extent through the eye movement behavior.

[0152] Optionally, the processor 902 may be a CPU (Central Processing Unit), a general-purpose processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It may implement or execute various exemplary logic blocks, modules, and circuits described in conjunction with the disclosure of this application. The processor 902 may also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, and the like.

[0153] Optionally, the processor 902 is further used to input parameter values ​​corresponding to the prediction parameter items of the target object into the prediction model to obtain a prediction result; the prediction result is used to indicate whether the target object suffers from an early-stage predetermined brain function disease.

[0154] Optionally, the candidate prediction model is used as an initial prediction model, and the candidate prediction model is continuously trained, iterated and verified to obtain a target prediction model, which can be used to predict whether the target subject suffers from an early-stage predetermined brain function disease. In practice, the candidate prediction model can also be used to predict an early-stage predetermined brain function disease.

[0155] In some embodiments, the predetermined brain function disease includes any one of the following diseases: Alzheimer's disease, amyotrophic lateral sclerosis, Huntington's disease, cerebellar atrophy, multiple sclerosis, Parkinson's disease of varying severity.

[0156] Optionally, the brain function detection system 90 may further include a memory; the memory stores a computer program, and the processor 902 executes the computer program to implement the steps of the method of the embodiment of the present application.

[0157] The memory may be a ROM (Read Only Memory) or other types of static storage devices that can store static information and instructions, a RAM (Random Access Memory) or other types of dynamic storage devices that can store information and instructions, or an EEPROM (Electrically Erasable Programmable Read Only Memory), a CD-ROM (Compact Disc Read Only Memory) or other optical disk storage, optical disk storage (including compact discs, laser discs, optical discs, digital versatile discs, Blu-ray discs, etc.), magnetic disk storage media, other magnetic storage devices, or any other medium that can be used to carry or store computer programs and can be read by a computer, without limitation herein.

[0158] The memory is used to store the computer program for executing the embodiment of the present application, and the execution is controlled by the processor 902. The processor 902 is used to execute the computer program stored in the memory to implement the steps shown in the above method embodiment.

[0159] The processor 902 can be applied to devices such as smart phones, desktop computers, laptop computers, and servers.

[0160] Based on the same inventive concept, an embodiment of the present application provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps of the method of the embodiment of the present application are implemented; and / or, when executed, the steps of the method for constructing a prediction model of the embodiment of the present application are implemented.

[0161] The computer-readable medium of the present application may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, a system, device or device of electricity, magnetism, light, electromagnetic, infrared, or semiconductor, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0162] It should be understood that, although each operation step is indicated by arrows in the flowchart of the embodiment of the present application, the implementation order of these steps is not limited to the order indicated by the arrows. Unless clearly stated herein, in some implementation scenarios of the embodiment of the present application, the implementation steps in each flowchart can be performed in other orders according to demand. In addition, some or all of the steps in each flowchart may include multiple sub-steps or multiple stages based on actual implementation scenarios. Some or all of these sub-steps or stages may be executed at the same time, and each sub-step or stage in these sub-steps or stages may also be executed at different times respectively. In different scenarios of execution time, the execution order of these sub-steps or stages may be flexibly configured according to demand, and the embodiment of the present application does not limit this.

[0163] In the embodiments of the present application, the term "module" or "unit" refers to a computer program or a part of a computer program that has a predetermined function and works together with other related parts to achieve a predetermined goal, and can be implemented in whole or in part by using software, hardware (such as processing circuits or memories), or a combination thereof. Similarly, a processor (or multiple processors or memories) can be used to implement one or more modules or units. In addition, each module or unit can be part of an overall module or unit that includes the function of the module or unit.

[0164] The above is only an optional implementation method for some implementation scenarios of the present application. It should be pointed out that for ordinary technicians in this technical field, without departing from the technical concept of the solution of the present application, other similar implementation methods based on the technical ideas of the present application are also within the protection scope of the embodiments of the present application.

Claims

1. A method for determining a prediction parameter item of eye movement abnormality, characterized in that: include: Obtain eye collection data; The eye acquisition data is information about the eye movements of multiple test subjects in response to a preset eye saccade task, and the eye acquisition data includes multiple eye saccade parameter items and parameter values ​​corresponding to the eye saccade parameter items; The preset saccade tasks include: a forward saccade task and a memory-guided saccade task; the test subjects include a control group and an early-stage disease group, the test subjects of the control group do not suffer from an early-stage predetermined brain function disease, and the test subjects of the early-stage disease group suffer from early-stage Parkinson's disease; According to the plurality of saccade parameter items, the parameter values ​​corresponding to the saccade parameter items and the design model, a prediction parameter item is determined from the plurality of saccade parameter items as an independent prediction factor for early-stage predetermined brain function diseases; the prediction parameter item includes at least one of the following: the maximum saccade velocity of memory-guided saccades, the reaction time of memory-guided saccades, the maximum saccade velocity of forward saccades, and the spatial error of memory-guided saccades; Wherein, the design model is a binary logistic regression model; The method of determining a prediction parameter item from the multiple saccadic parameter items according to the multiple saccadic parameter items, the parameter values ​​corresponding to the saccadic parameter items and the designed model comprises: normalizing the parameter values ​​corresponding to each saccadic parameter item to obtain data to be input; inputting the data to be input into the binary logistic regression model to obtain a candidate prediction model; determining the correlation between the candidate prediction model and the prediction of an early predetermined brain function disease; if the correlation between the candidate prediction model and the prediction of an early predetermined brain function disease exceeds a designed threshold, using the saccadic parameter item used by the candidate prediction model as the prediction parameter item.

2. The method for determining the prediction parameter item of eye movement abnormality according to claim 1, characterized in that: Determining a prediction parameter item from the plurality of eye saccade parameter items according to the plurality of eye saccade parameter items, parameter values ​​corresponding to the eye saccade parameter items and a design model comprises: Determining a prediction parameter item from a plurality of saccadic parameter items according to the saccadic parameter items that meet a predetermined condition, parameter values ​​corresponding to the saccadic parameter items, and a design model; Wherein, satisfying the predetermined condition includes: There are significant differences between the parameter values ​​of the multiple test subjects in the control group and the parameter values ​​of the multiple test subjects in the early disease group for each of the saccadic parameter items; There is no obvious collinearity between the parameter values ​​of the eye-saccade parameter items.

3. The method for determining the prediction parameter item of eye movement abnormality according to claim 1, characterized in that: The forward saccade task includes: displaying an initial fixation point at an initial position, detecting whether the test subject has been fixating on the initial fixation point, stopping displaying the initial fixation point after a first predetermined time, and displaying a target point at a first predetermined position different from the initial position, detecting whether the test subject saccades to the first predetermined position within a second predetermined time; The memory-guided saccade task includes: displaying an initial fixation point at an initial position, detecting whether the test subject has been gazing at the initial fixation point, displaying a target point at a first predetermined position different from the initial position and stopping displaying the target point after a third predetermined time, controlling the initial fixation point to stop displaying after detecting that the test subject has been gazing at the initial fixation point for a fourth predetermined time, and detecting whether the test subject saccades to the first predetermined position within a fifth predetermined time.

4. The method for determining the prediction parameter item of eye movement abnormality according to claim 3, characterized in that: The saccade parameter items include: the accuracy of positive saccade, the reaction time of positive saccade, the spatial error of positive saccade, the saccade gain of positive saccade and the maximum saccade speed of positive saccade; The saccade parameter items also include: the accuracy of memory-guided saccades, the reaction time of memory-guided saccades, the spatial error of memory-guided saccades, the saccade gain of memory-guided saccades and the maximum saccade speed of memory-guided saccades; Among them, the accuracy is used to indicate the accuracy of saccades, the reaction time is used to indicate the time interval from the disappearance of the initial fixation point to the start of the main saccade, the spatial error is used to indicate the straight-line distance from the saccade landing point to the target point, the saccade gain is used to indicate the ratio of the straight-line distance from the saccade starting point to the saccade target point to the straight-line distance from the saccade starting point to the target point, and the maximum saccade speed is used to indicate the peak speed of the saccade during the saccade process.

5. The method for determining the prediction parameter item of eye movement abnormality according to claim 1, characterized in that: Determining the correlation between the candidate prediction model and the prediction of the early stage predetermined brain function disease includes: According to the parameter values ​​corresponding to the prediction parameter items, ROC analysis is performed on the candidate prediction model to obtain the area under the curve; If the area under the curve is greater than a fourth predetermined value, it is determined that the correlation between the candidate prediction model and the prediction of the early predetermined brain function disease exceeds a designed threshold.

6. A method for constructing a prediction model, characterized in that: include: Obtain eye collection data; The eye acquisition data is information about the eye movements of multiple test subjects in response to a preset eye saccade task, and the eye acquisition data includes multiple eye saccade parameter items and parameter values ​​corresponding to the eye saccade parameter items; A prediction model is determined according to the plurality of saccadic parameter items, parameter values ​​corresponding to the saccadic parameter items and a design model; the prediction model comprises the prediction parameter items as described in any one of claims 1-5.

7. A device for determining prediction parameters of eye movement abnormalities, characterized in that: The method for determining the prediction parameter item of eye movement abnormality according to any one of claims 1 to 5 is adopted, wherein the determination device comprises: An acquisition module is used to acquire eye acquisition data; the eye acquisition data is information about eye movements of multiple test subjects in response to a preset eye saccade task, and the eye acquisition data includes multiple eye saccade parameter items and parameter values ​​corresponding to the eye saccade parameter items; the preset eye saccade task includes: a forward eye saccade task and a memory-guided eye saccade task; the test subjects include a control group and an early-stage disease group, the test subjects of the control group do not suffer from an early-stage predetermined brain function disease, and the test subjects of the early-stage disease group suffer from early-stage Parkinson's disease; A determination module is used to determine a prediction parameter item from the plurality of saccadic parameter items according to the plurality of saccadic parameter items, parameter values ​​corresponding to the saccadic parameter items and a design model, as an independent prediction factor for early-stage predetermined brain function diseases; the prediction parameter item includes at least one of the following: a maximum saccadic velocity of memory-guided saccadic saccades, a reaction time of memory-guided saccadic saccades, a maximum saccadic velocity of forward saccadic saccades, and a spatial error of memory-guided saccadic saccades; Wherein, the design model is a binary logistic regression model; The method of determining a prediction parameter item from the multiple saccadic parameter items according to the multiple saccadic parameter items, the parameter values ​​corresponding to the saccadic parameter items and the designed model comprises: normalizing the parameter values ​​corresponding to each saccadic parameter item to obtain data to be input; inputting the data to be input into the binary logistic regression model to obtain a candidate prediction model; determining the correlation between the candidate prediction model and the prediction of an early predetermined brain function disease; if the correlation between the candidate prediction model and the prediction of an early predetermined brain function disease exceeds a designed threshold, using the saccadic parameter item used by the candidate prediction model as the prediction parameter item.

8. A device for constructing a prediction model, characterized in that: include: A construction module is used to determine a prediction model based on a design model and prediction parameter items of eye movement abnormalities; the prediction parameter items are obtained based on the method for determining the prediction parameter items of eye movement abnormalities described in any one of claims 1-5.

9. A brain function detection system, characterized in that: include: An electrically connected eye movement acquisition device and a processor; The eye movement acquisition device is used to sample the eye movement of the test subject in response to a preset eye saccade task to obtain eye acquisition data; The processor is used to execute the steps of the method for determining the prediction parameter items of eye movement abnormalities as described in any one of claims 1-5; and / or, execute the steps of the method for constructing the prediction model as described in claim 6.

10. The brain function detection system according to claim 9, characterized in that: The processor is further used to input the parameter value corresponding to the prediction parameter item of the target object into a prediction model to obtain a prediction result; the prediction result is used to indicate whether the target object suffers from early Parkinson's disease.

11. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method for determining the prediction parameter items of eye movement abnormalities as described in any one of claims 1-5 are implemented; and / or, when executed, the steps of the method for constructing the prediction model as described in claim 6 are implemented.

Images