A diagnostic method based on saccade analysis

Abstract

This invention provides a diagnostic method based on saccade analysis. The method includes: collecting eye movement displacement and time data of the subject to obtain a discrete eye movement velocity sequence; obtaining a denoised eye movement velocity sequence after denoising; selecting the velocity sequence to be processed using a moving window; determining the time length of the moving window based on the length of the velocity sequence and the calculation time requirements; determining an eye movement velocity threshold according to diagnostic needs; calculating local extrema of the velocity sequence within the moving window and determining the eye movement type and number of saccades based on the velocity threshold; calculating local extrema of the entire velocity sequence using this method to obtain the total number of saccades; diagnosing the subject's eye movement state based on the number of saccades and their time intervals, and determining whether detailed analysis of local data is needed to obtain a more accurate diagnostic result based on the diagnosis results. This invention can help doctors diagnose different types of saccades and their potential predictive conditions such as cognitive impairment and Parkinson's disease more quickly.

Description

Technical Field

[0001] This invention relates to the field of eye movement data analysis, and more particularly to a diagnostic method based on saccade analysis. Background Technology

[0002] With the development of optical technology, computer technology, and artificial intelligence, eye-tracking technology, as one of the most effective means of visual information processing, has received widespread attention from researchers at home and abroad in recent years. It has been applied to user experience and interaction research (web page usability, mobile device usability, software usability, game usability, gaze interaction research), market research and consumer research (shopping behavior research, packaging design research, advertising research), human performance research, psychology and neuroscience research (cognitive psychology research, neuroscience research, social psychology research, visual perception research, primate and canine animal research, etc.), automotive HMI interaction design, status monitoring of operators in aerospace / nuclear power plants, and medical research and medical applications.

[0003] Eye movements (saccades) are highly complex, encompassing three basic modes: fixation, saccades, and following movements. Fixation itself includes three types of movements: drifting, nystagmus, and subtle involuntary saccades. Saccades involve three processes: saccade latency, saccades, and saccade inhibition. Fixation and saccades, as two primary eye movements, are crucial for the study of eye movement behavior. For a long time, saccade behavior has been found to be associated with certain diseases. Especially in neurology and psychiatry, quantitative analysis of saccades has been an important diagnostic tool, particularly in cases of concussion, cognition, mild traumatic brain injury, autism, schizophrenia, macular degeneration-related functional deficits, and attention deficit disorder. In recent years, reverse saccade indicators have also become crucial for physicians in identifying known or suspected diseases involving the frontal cortex and / or basal ganglia, such as basal ganglia disorders, schizophrenia, attention deficit hyperactivity disorder, and dyslexia. Meanwhile, the role of saccades and retrosaccades as diagnostic tools in the early diagnosis of Alzheimer's and Parkinson's diseases is receiving increasing attention, and it is expected that saccade and retrosaccade technologies will achieve greater success in the diagnosis and treatment of Alzheimer's and Parkinson's diseases in the future.

[0004] Due to the enormous potential of eye-tracking analysis in medical research and diagnosis, high-precision eye-tracking devices have also seen significant development. For example, the Eyelink series of eye trackers, with their high sampling rate, high precision, and low noise, has always been the preferred choice in various research fields, especially in medical systems. The Eyelink 1000Plus can support binocular tracking up to 2000Hz, meaning that a single eye-tracking examination can generate a massive amount of eye-tracking data. On the other hand, when using eye-tracking analysis for the analysis of psychological and neurological diseases, subtle differences in eye movement phenomena, especially saccades, are often present, requiring coordination with other diagnostic equipment. Therefore, how to identify and analyze different types of saccades has become a pressing problem to be solved. Summary of the Invention

[0005] To overcome the problem that existing technologies cannot quickly identify and analyze saccade types, this invention provides a diagnostic method based on saccade analysis.

[0006] In a first aspect, to solve the above-mentioned technical problems, the present invention provides a diagnostic method based on saccade analysis, comprising the following steps:

[0007] S1. Acquire eye-tracking image data of the subject;

[0008] S2. Perform image processing on the obtained eye-tracking image data to obtain eye movement displacement and time data;

[0009] S3. Convert the eye movement displacement and time data into a one-dimensional sequence to obtain the displacement sequence, and then perform the first derivative of the displacement sequence to obtain the eye movement velocity sequence;

[0010] S4. Denoise the displacement and eye-tracking velocity sequences;

[0011] S5. Determine the type of saccade diagnosed and the corresponding velocity threshold for that type of saccade;

[0012] S6. Determine the duration of the moving window based on the length of the eye-tracking velocity sequence;

[0013] S7. Place the moving window on the denoised eye-tracking velocity series;

[0014] S8. Find all extreme values ​​of the eye movement velocity sequence within the window and compare them with the set velocity threshold. If the extreme value exceeds the velocity threshold, it is determined to be the saccade type corresponding to the velocity threshold.

[0015] S9. Slide the moving window from the beginning of the eye movement velocity sequence to the end of the sequence to obtain the eye movement detection sequence corresponding to the entire eye movement velocity sequence and the total number of eye saccades detected;

[0016] S10. Compare the obtained eye movement detection sequence and the total number of each eye saccade with the expected data or normal eye saccade sequence and its time interval, and determine whether the obtained result matches the expected or normal eye saccade sequence. If it does, proceed to S11. If it does not, it is judged as abnormal data and proceed to S12.

[0017] S11. Diagnosis complete;

[0018] S12. Identify outlier data and, based on the time period corresponding to the outlier data, extract and amplify the outlier data for more detailed analysis;

[0019] S13. Analyze the abnormal data by moving the window to determine the cause of the abnormality and whether micro-saccade analysis is needed. If micro-saccade analysis is needed, repeat steps S5 and thereafter; if micro-saccade analysis is not needed, proceed to S14.

[0020] S14. Mark the abnormal data as outliers.

[0021] The beneficial effects of the diagnostic method based on saccade analysis provided by this invention are as follows: eye movement displacement and time data are converted into a displacement sequence, and then an eye movement velocity sequence is obtained by first-order differentiation. This allows us to obtain the offset velocity of the macula in the eye tracker. By moving the window across the eye movement velocity sequence, extreme values ​​are obtained, and these extreme values ​​are compared with velocity thresholds to determine the saccade type. For each saccade type, after the moving window has slid across the entire eye movement velocity sequence, the eye movement detection sequence and the total number of saccades are obtained. Finally, the obtained eye movement detection sequence and the total number of each saccade are compared with the expected data or the normal saccade sequence and its time interval to determine whether the saccade is normal. This solves the problem that existing technologies cannot quickly identify and analyze saccade types.

[0022] Based on the above technical solution, the diagnostic method based on saccade analysis of the present invention can be further improved as follows.

[0023] Furthermore, the length of the movement time window can be freely set according to the length of the eye-tracking velocity sequence and the requirements for calculation time.

[0024] The beneficial effect of adopting the above-mentioned further scheme is that the duration of the moving window can be freely set according to the length of the eye movement speed sequence and the requirements for calculation time, which can meet the requirements of any eye movement speed sequence.

[0025] Furthermore, speed thresholds are set for each type of saccade based on the type of saccade and diagnostic requirements.

[0026] The beneficial effect of adopting the above-mentioned further scheme is that: setting the speed threshold corresponding to each type of saccade according to the type of saccade and diagnostic requirements basically satisfies all types of saccade.

[0027] Furthermore, the moving window speed threshold algorithm for steps S5 and thereafter is as follows:

[0028] Determine the type of saccade to be diagnosed and the diagnostic requirements, and determine the corresponding oculomotor velocity threshold based on the type of saccade to be diagnosed and the requirements, in degrees per second (deg / s);

[0029] The moving window is a time window, so the time length of the moving window is determined based on the denoised sequence of all eye movement velocities and their corresponding time span and computation time requirements.

[0030] At the beginning of the algorithm, the moving window is placed on the denoised eye movement velocity sequence. The start time of the moving window is the start time of the time series corresponding to the eye movement velocity sequence, and the end time of the moving window is the sum of the start time of the time series and the length of the moving window. The sum of several moving windows completely covers the entire denoised eye movement velocity sequence.

[0031] The extreme values ​​of the eye movement velocity series within the moving window are calculated. All extreme values ​​of this segment of the eye movement velocity series are obtained and compared with the set velocity threshold. If the extreme value exceeds the velocity threshold, it is determined to be the saccade type corresponding to the velocity threshold. The value of the saccade type detection data at the corresponding time point is marked as 1. If no saccade type is detected, the value of the detection data of the undetected saccade type at the corresponding time point is marked as 0.

[0032] The moving window will slide from the start time to the end time of the eye movement velocity sequence. When the end time of the moving window exceeds the end time of the eye movement velocity sequence, the sliding stops, and the complete detection sequence of the entire eye movement velocity sequence and the total number of eye saccades and their time intervals for each eye saccade type are obtained.

[0033] The obtained complete test data and the total number of each saccade are compared with the expected data or the normal saccade data and the time interval. It is determined whether the obtained results meet the expected or normal saccade data. If they do, the diagnosis ends. If they do not, the data is judged as abnormal.

[0034] If not satisfied, it also includes:

[0035] Then identify the abnormal data, and extract and magnify the data within the time period corresponding to the abnormal data using a window to determine whether micro-eye movement analysis is needed. If micro-eye movement analysis is needed, set the micro-eye movement speed threshold and repeat the moving window speed threshold algorithm.

[0036] If micro-eye movement analysis is not required, mark the abnormal data as outliers.

[0037] The beneficial effects of adopting the above-mentioned further scheme are as follows: based on the eye movement velocity sequence slid across the moving window, the extreme values ​​are obtained, and the extreme values ​​are compared with the velocity threshold to obtain the saccade type. The total number of saccade types is increased by 1. For each saccade type, after the moving window has slid across the entire eye movement velocity sequence, the eye movement detection sequence and the total number of saccades corresponding to the saccade type can be obtained. Finally, the obtained eye movement detection sequence and the total number of each saccade are compared with the expected data or the normal saccade sequence and its time interval to determine whether the saccade is normal. If it is determined to be abnormal data, micro-eye movement analysis is required, and the abnormal data is determined to be a real anomaly based on the micro-eye movement analysis. Attached Figure Description

[0038] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the present invention will be further described below in conjunction with the accompanying drawings and embodiments.

[0039] Figure 1 This is a flowchart illustrating a diagnostic method based on saccade analysis according to an embodiment of the present invention. Detailed Implementation

[0040] The following embodiments are further explanations and supplements to the present invention and do not constitute any limitation on the present invention.

[0041] The following describes a diagnostic method based on saccade analysis according to an embodiment of the present invention, with reference to the accompanying drawings.

[0042] This invention discloses a diagnostic method based on saccade analysis, which is applied to a terminal device. The terminal device is used as the execution subject in this application. The terminal device can be a computer, server, etc., and is used to execute the steps of a diagnostic method based on saccade analysis. The terminal device is also connected to an eye tracker, which is used to acquire eye-tracking images.

[0043] like Figure 1 As shown, the present invention provides a diagnostic method based on saccade analysis, comprising the following steps:

[0044] S1. Acquire eye-tracking image data of the subject;

[0045] S2. Perform image processing on the obtained eye-tracking image data to obtain eye movement displacement and time data;

[0046] S3. Convert the eye movement displacement and time data into a one-dimensional sequence to obtain the displacement sequence, and then perform the first derivative of the displacement sequence to obtain the eye movement velocity sequence;

[0047] S4. Denoise the displacement and eye-tracking velocity sequences;

[0048] S5. Determine the type of saccade diagnosed and the corresponding velocity threshold for that type of saccade;

[0049] S6. Determine the duration of the moving window based on the length of the eye-tracking velocity sequence;

[0050] S7. Place the moving window on the denoised eye-tracking velocity series;

[0051] S8. Find all extreme values ​​of the eye movement velocity sequence within the window and compare them with the set velocity threshold. If the extreme value exceeds the velocity threshold, it is determined to be the saccade type corresponding to the velocity threshold.

[0052] S9. Slide the moving window from the beginning of the eye movement velocity sequence to the end of the sequence to obtain the eye movement detection sequence corresponding to the entire eye movement velocity sequence and the total number of eye saccades detected;

[0053] S10. Compare the obtained eye movement detection sequence and the total number of each saccade with the expected data or normal saccade sequence and its time interval, and determine whether the obtained result matches the expected or normal saccade sequence. If it matches, proceed to S11. If it does not match, it is judged as abnormal data and proceed to S12. S11. Diagnosis ends.

[0054] S12. Identify outlier data and, based on the time period corresponding to the outlier data, extract and amplify the outlier data for more detailed analysis;

[0055] S13. Analyze the abnormal data by moving the window to determine the cause of the abnormality and whether micro-saccade analysis is needed. If micro-saccade analysis is needed, repeat steps S5 and thereafter; if micro-saccade analysis is not needed, proceed to S14.

[0056] S14. Mark the abnormal data as outliers.

[0057] Optionally, the eye tracker can be an Eyelink series eye tracker, which has excellent characteristics such as high sampling rate, high accuracy and low noise. The Eyelink 1000Plus can support binocular tracking up to 2000Hz, which can generate a large amount of eye movement data in a single eye movement examination.

[0058] Optionally, in S1, eye-tracking image data of the subject is collected at each time point within a preset time period, for example, eye-tracking image data of the subject is collected once every second within an hour.

[0059] Optionally, the eye-tracking image data is an image of the macula in the fundus of the eye shifting in the eye tracker.

[0060] Optionally, eye movement displacement is the displacement distance corresponding to the shift of the macula in the fundus of the eye in the eye tracker.

[0061] Optionally, the time data can be each point in time within a preset time period.

[0062] Optionally, since eye movement velocity refers to the velocity corresponding to the offset angle of the macula in the fundus when it is shifted in the eye tracker, the eye movement velocity sequence can be obtained by taking the first derivative of the displacement sequence based on the fact that the offset velocity is equal to the displacement divided by time.

[0063] Optionally, in S4, the method further includes:

[0064] The sampling rate of the eye tracker when acquiring eye-tracking image data captured within a preset time period;

[0065] Denoising the displacement sequence includes:

[0066] Based on the sampling rate, the displacement sequence is denoised using the first formula, which is:

[0067]

[0068]

[0069]

[0070]

[0071] Where y represents the displacement sequence, x represents the denoised displacement sequence, J(x) is the preset objective equation, α and β are preset regularization parameters, D1 and D3 are preset first-order difference matrices and second-order difference matrices respectively, f represents the sampling rate, σ represents the preset noise signal, A represents the preset average saccade amplitude, and D is the preset average saccade duration.

[0072] Optionally, in S4, the eye-tracking velocity sequence is denoised, including:

[0073] Obtain the preset moving average value corresponding to eye saccades;

[0074] The displacement array is constructed by using the product of the moving average and the sampling rate as its size.

[0075] Determine the target eye movement velocity sequence based on the displacement array and the eye movement velocity sequence.

[0076] Optionally, when constructing the displacement array, the product of the moving average and the sampling rate is used as the size of the displacement array, and any column in the displacement array must be completely replaced with 1.

[0077] Optionally, the displacement array and the eye movement velocity sequence can be convolved to obtain the denoised target eye movement velocity sequence.

[0078] Optionally, in S6, the time length of the moving window is determined based on the length of the eye-tracking velocity sequence, including:

[0079] The length of the moving time window can be freely set according to the length of the eye movement velocity sequence and the requirements for calculation time. In this embodiment, the calculation time refers to the estimated time for the terminal device to complete one diagnostic method based on saccade analysis.

[0080] Optionally, all extreme values ​​of the eye movement velocity series within the window are obtained and compared with a set velocity threshold. If an extreme value exceeds the velocity threshold, it is determined to be the saccade type corresponding to the velocity threshold, including:

[0081] According to the type of saccade and the diagnostic requirements, a velocity threshold is set for each type of saccade. In this embodiment, the diagnostic requirements are the consensus reached by experts in the field of eye movement. For example, there are three main types of saccade: fixation, saccade, and smoothing. The above three types of saccade are distinguished according to different velocity thresholds. When the extreme value exceeds any velocity threshold, it indicates that the saccade type is the saccade type corresponding to the velocity threshold.

[0082] Optionally, the moving window speed threshold algorithm for S5 and later versions is as follows:

[0083] Determine the type of saccade to be diagnosed and the diagnostic requirements, and determine the corresponding oculomotor velocity threshold based on the type of saccade to be diagnosed and the requirements, in degrees per second (deg / s);

[0084] The moving window is a time window, so the time length of the moving window is determined based on the denoised sequence of all eye movement velocities and their corresponding time span and computation time requirements.

[0085] At the beginning of the algorithm, the moving window is placed on the denoised eye movement velocity sequence. The start time of the moving window is the start time of the time series corresponding to the eye movement velocity sequence, and the end time of the moving window is the sum of the start time of the time series and the length of the moving window. The sum of several moving windows completely covers the entire denoised eye movement velocity sequence.

[0086] The extreme values ​​of the eye movement velocity series within the moving window are calculated. All extreme values ​​of this segment of the eye movement velocity series are obtained and compared with the set velocity threshold. If the extreme value exceeds the velocity threshold, it is determined to be the saccade type corresponding to the velocity threshold. The value of the saccade type detection data at the corresponding time point is marked as 1. If no saccade type is detected, the value of the detection data of the undetected saccade type at the corresponding time point is marked as 0.

[0087] The moving window will slide from the start time to the end time of the eye movement velocity sequence. When the end time of the moving window exceeds the end time of the eye movement velocity sequence, the sliding stops, and the complete detection sequence of the entire eye movement velocity sequence and the total number of eye saccades and their time intervals for each eye saccade type are obtained.

[0088] The obtained complete test data and the total number of each saccade are compared with the expected data or the normal saccade data and the time interval. It is determined whether the obtained results meet the expected or normal saccade data. If they do, the diagnosis ends. If they do not, the data is judged as abnormal.

[0089] If not satisfied, it also includes:

[0090] Then identify the abnormal data, and extract and magnify the data within the time period corresponding to the abnormal data using a window to determine whether micro-eye movement analysis is needed. If micro-eye movement analysis is needed, set the micro-eye movement speed threshold and repeat the moving window speed threshold algorithm.

[0091] If micro-eye movement analysis is not required, mark the abnormal data as outliers.

[0092] Optionally, the obtained complete detection sequence and the total number of each saccade are compared with the expected data or normal saccade sequence and its time intervals to determine whether the obtained results match the expected or normal saccade sequence, including:

[0093] The complete detection sequence is compared with the normal saccade sequence. If the complete detection sequence is almost identical to the normal saccade sequence, it indicates that the complete detection sequence meets expectations. If the complete detection sequence differs significantly from the normal saccade sequence, it indicates that the complete detection sequence does not meet expectations. The specific judgment of whether the complete detection sequence is almost identical to the normal saccade sequence is set according to the actual situation. For example, if the complete detection sequence is almost identical to the normal saccade sequence in more than n corresponding positions (n≥1), it is considered to not meet expectations.

[0094] Compare the total number of saccades with the total number of saccades in a normal person. If the difference is small, the total number of saccades is in line with expectations. If the difference is large, the total number of saccades is not in line with expectations.

[0095] Compare the time interval with the normal saccade time interval. If the difference between the time interval and the normal saccade time interval is moderate, it indicates that the time interval meets expectations. If the difference between the time interval and the normal saccade time interval is large or small, it indicates that the time interval does not meet expectations. In addition, the time interval refers to the time interval between two adjacent saccade types in the total number of saccades. A value range can also be preset. If the difference between the time interval and the normal saccade time interval falls within the value range, it indicates that the time interval meets expectations. If the difference between the time interval and the normal saccade time interval falls outside the value range, it indicates that the time interval does not meet expectations.

[0096] This invention also provides a diagnostic system based on saccade analysis, comprising:

[0097] The eye-tracking image data module is used to collect eye-tracking image data of the subject being tested;

[0098] The first data acquisition module is used to perform image processing on the acquired eye-tracking image data to obtain eye movement displacement and time data;

[0099] The eye-tracking velocity sequence module is used to convert eye-tracking displacement and time data into a one-dimensional sequence, obtain the displacement sequence, and perform first-order differentiation on the displacement sequence to obtain the eye-tracking velocity sequence.

[0100] The noise reduction module is used to denoise the displacement and eye-tracking velocity sequences.

[0101] The velocity threshold module is used to determine the type of saccade diagnosed and the velocity threshold corresponding to the saccade type.

[0102] The moving window module is used to determine the duration of the moving window based on the length of the eye-tracking velocity sequence.

[0103] The sliding module is used to place the moving window on the denoised eye-tracking velocity sequence;

[0104] The extreme value module is used to find all extreme values ​​of the eye movement velocity series within the window and compare them with the set velocity threshold. If the extreme value exceeds the velocity threshold, it is determined to be the eye saccade type corresponding to the velocity threshold.

[0105] The second data acquisition module is used to slide the moving window from the beginning to the end of the eye movement velocity sequence to obtain the eye movement detection sequence corresponding to the entire eye movement velocity sequence and the total number of eye saccades detected.

[0106] The first judgment module is used to compare the obtained eye movement detection sequence and the total number of each eye saccade with the expected data or normal eye saccade sequence and its time interval, and to determine whether the obtained result matches the expected or normal eye saccade sequence. If it does, the function corresponding to the second judgment module is executed. If it does not, it is judged as abnormal data and the function corresponding to the third judgment module is executed.

[0107] The second judgment module is used to end the diagnosis;

[0108] The third judgment module is used to identify abnormal data and, based on the time period corresponding to the abnormal data, extract and amplify the abnormal data for more detailed analysis.

[0109] The micro-eye movement analysis module is used to analyze abnormal data by moving the window, determine the cause of the abnormality and whether micro-eye movement analysis is needed. If micro-eye movement analysis is needed, the corresponding functions between the speed threshold module and the micro-eye movement analysis module are repeated; if micro-eye movement analysis is not needed, the corresponding functions of the anomaly point module are executed.

[0110] The anomaly point module is used to mark abnormal data as anomalies.

[0111] Optional, a mobile window module, specifically used for:

[0112] The length of the movement time window can be freely set according to the length of the eye movement speed sequence and the requirements for calculation time.

[0113] Optional, speed threshold module, specifically used for:

[0114] Set the speed threshold corresponding to each type of saccade based on the type of saccade and diagnostic requirements.

[0115] Optional, sliding module, specifically used for:

[0116] First, place the moving window on the denoised eye movement velocity series. The start time of the moving window is the start time of the corresponding time series of the eye movement velocity series, and the end time of the moving window is the sum of the start time of the time series and the length of the moving window. The sum of several moving windows completely covers the entire denoised eye movement velocity series.

[0117] The extremum module is specifically used for:

[0118] The extreme values ​​of the eye movement velocity series within the moving window are calculated. All extreme values ​​of this segment of the eye movement velocity series are obtained and compared with the set velocity threshold. If the extreme value exceeds the velocity threshold, it is determined to be the saccade type corresponding to the velocity threshold. The value of the saccade type detection data at the corresponding time point is marked as 1. If no saccade type is detected, the value of the detection data of the undetected saccade type at the corresponding time point is marked as 0.

[0119] The second data acquisition module is specifically used for:

[0120] The moving window will slide from the start time to the end time of the eye movement velocity sequence. When the end time of the moving window exceeds the end time of the eye movement velocity sequence, the sliding stops, and the complete detection sequence of the entire eye movement velocity sequence is obtained, as well as the total number of eye saccades and their time intervals for each eye saccade type.

[0121] The first judgment module is specifically used for:

[0122] The obtained complete test data and the total number of each saccade are compared with the expected data or the normal saccade data and their time intervals to determine whether the results meet the expected or normal saccade data. If they do, the diagnosis ends; otherwise, the data is judged as abnormal.

[0123] The micro-eye-tracking analysis module is specifically used for:

[0124] Then, identify the abnormal data, and based on the time period corresponding to the abnormal data, use a window to extract and magnify the data within that time period to determine whether micro-eye movement analysis is needed. If micro-eye movement analysis is needed, set a micro-eye movement speed threshold and repeat the moving window speed threshold algorithm.

[0125] The exception point module is specifically used for:

[0126] If micro-eye movement analysis is not required, mark the abnormal data as outliers.

[0127] An electronic device according to an embodiment of the present invention includes a memory, a processor, and a program stored in the memory and running on the processor. When the processor executes the program, it implements some or all of the steps of the diagnostic method based on saccade analysis described above.

[0128] The electronic device can be a computer, and its program is computer software. The parameters and steps of the electronic device of the present invention can be referred to the parameters and steps in the embodiment of the diagnostic method based on saccade analysis above, and will not be repeated here.

[0129] Those skilled in the art will recognize that this invention can be implemented as a system, method, or computer program product. Therefore, this disclosure can be embodied in the following forms: it can be entirely hardware, entirely software (including firmware, resident software, microcode, etc.), or a combination of hardware and software, generally referred to herein as a "circuit," "module," or "system." Furthermore, in some embodiments, the invention can also be implemented as a computer program product contained in one or more computer-readable media, which contains computer-readable program code. Computer-readable storage media can be, for example, but not limited to—electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatuses, or devices, or any combination thereof.

[0130] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0131] Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention.

Claims

1. An electronic device comprising a memory, a processor, and a program stored in the memory and running on the processor, wherein the processor, when executing the program, implements all steps of a diagnostic method based on saccade analysis, characterized in that, Includes the following steps: S1. Acquire eye-tracking image data of the subject; S2. Perform image processing on the obtained eye-tracking image data to obtain eye movement displacement and time data; S3. Convert the eye movement displacement and the time data into a one-dimensional sequence to obtain the displacement sequence, and perform the first derivative of the displacement sequence to obtain the eye movement velocity sequence; S4. Denoise the displacement sequence and the eye movement velocity sequence; S5. Determine the type of saccade diagnosed and the velocity threshold corresponding to the saccade type; S6. Determine the duration of the moving window based on the length of the eye movement velocity sequence; S7. Place the moving window on the denoised eye movement velocity sequence. The moving window is a time window, whose start time is the start time of the eye movement velocity sequence and whose end time is the sum of the start time and the window time length. S8. Obtain all extreme values ​​of the eye movement velocity sequence within the window and compare them with the set velocity threshold. If the extreme value exceeds the velocity threshold, it is determined to be the saccade type corresponding to the velocity threshold. The detection data of the saccade type is marked as 1 at the corresponding time point. If no saccade type is detected, the detection data of the undetected saccade type is marked as 0 at the corresponding time point. S9. Slide the moving window from the beginning of the eye movement velocity sequence to the end of the sequence, with a sliding step size of a preset time unit. Stop sliding when the window ends after the sequence ends, and obtain the complete detection sequence of the entire eye movement velocity sequence and the total number of saccades and their time intervals corresponding to each saccade type. S10. Compare the obtained complete detection sequence and the total number of each eye saccade and its time interval with normal data to determine whether the obtained result meets expectations. If it does, proceed to S11; if it does not, it is determined to be abnormal data and proceed to S12. S11. Diagnosis complete; S12. Identify outlier data and, based on the time period corresponding to the outlier data, extract and amplify the outlier data for more detailed analysis; S13. Analyze the abnormal data by moving the window to determine the cause of the abnormality and whether micro-saccade analysis is needed. If micro-saccade analysis is needed, repeat steps S5 and thereafter. If microsaccade analysis is not required, proceed to S14; S14. Mark the abnormal data as outliers; Additionally, in S4, the method also includes: The sampling rate of the eye tracker is used to acquire eye-tracking image data captured within a preset time period; the displacement sequence is denoised. Obtain the preset moving average corresponding to eye saccades; construct the displacement array by using the product of the moving average and the sampling rate as the size of the displacement array; determine the target eye movement velocity sequence based on the displacement array and the eye movement velocity sequence, and then denoise the eye movement velocity sequence.

2. The electronic device according to claim 1, characterized in that, The steps in the diagnostic method based on saccade analysis also include: The duration of the moving window can be freely set according to the length of the eye movement velocity sequence and the requirements for calculation time.

3. An electronic device according to claim 1, characterized in that, The steps in the diagnostic method based on saccade analysis also include: Set the speed threshold corresponding to each of the saccade types according to the saccade type and diagnostic requirements.

Images