CN119157480B

CN119157480B - A system and method for monitoring the progression of thyroid-related eye disease

Info

Publication number: CN119157480B
Application number: CN202411247666.7A
Authority: CN
Inventors: 杨俊洁
Original assignee: Union Hospital Tongji Medical College Huazhong University of Science and Technology
Current assignee: Union Hospital Tongji Medical College Huazhong University of Science and Technology
Priority date: 2024-09-06
Filing date: 2024-09-06
Publication date: 2025-09-12
Anticipated expiration: 2044-09-06
Also published as: CN119157480A

Abstract

The application relates to the technical field of image analysis, in particular to a thyroid-related eye disease progress monitoring system. The progress monitoring method is applied to an eye detector and comprises the steps of obtaining eye monitoring videos shot by a camera device, analyzing the eye monitoring videos to determine eye characteristics, analyzing the eye characteristics to determine eyelid opening and closing degree and eyeball movement condition, determining pupil change condition according to eyeball movement condition, determining current eye disease state according to eyelid opening and closing degree, eyeball movement condition and pupil change condition, obtaining and analyzing historical eye disease diagnosis and treatment records to obtain historical eye disease state, comparing the current eye disease state with historical eye disease state, and determining whether thyroid related eye disease is improved or not.

Description

System and method for monitoring progress of thyroid-related eye diseases

Technical Field

The application relates to the technical field of image analysis, in particular to a system and a method for monitoring the progress of thyroid-related eye diseases.

Background

Thyroid-related eye disease is a specific autoimmune disease which involves organs such as thyroid and orbit simultaneously, and is one of the most common orbit diseases for adult.

At present, the treatment of the thyroid-related eye diseases is mostly comprehensive analysis by combining facial diagnosis with CT contrast scanning, and the severity of the thyroid-related eye diseases of patients is determined. The subsequent determination of the treatment progress of the thyroid-related eye disease needs to be carried out to a hospital for evaluation by a doctor, and the process is very troublesome for a patient, and for the doctor, the thyroid-related eye disease is more serious if any problem exists in the middle of the process due to a certain time interval between the first diagnosis and the second review, so that the diagnosis after the second examination is similar to the first diagnosis, and is time-consuming and labor-consuming, causes are not found, the illness state of the patient is delayed, and the health of the patient is endangered.

Disclosure of Invention

The application provides a system and a method for monitoring the progress of thyroid-related eye diseases, which are used for solving the problems.

In a first aspect, the application provides a method for monitoring the progress of thyroid-related eye diseases, which is applied to an eye detector, wherein the eye detector comprises an imaging device and a data processing chip, the method is applied to the data processing chip, and the method comprises the following steps:

Acquiring an eye monitoring video shot by the camera device, analyzing the eye monitoring video, and determining eye characteristics;

Analyzing the eye characteristics and determining the opening and closing degree of eyelids and the eyeball movement condition;

determining pupil change conditions according to the eyeball movement conditions;

determining a current eye disease state according to the eyelid opening and closing degree, the eyeball movement condition and the pupil change condition;

and comparing the current eye disease state with the historical eye disease state to determine whether the thyroid-related eye disease is improved.

Through this scheme, acquire the eye monitoring video that camera device shot in real time, realize the real-time capture of eye state change. The video analysis and image processing technology provides objective eye feature data, and reduces subjectivity of artificial judgment. The eyelid opening and closing degree, eyeball movement condition and pupil change are accurately measured by using an advanced image processing algorithm, so that the accuracy and reliability of diagnosis are improved. The comprehensive judgment is carried out by combining a plurality of eye features, so that the limitation of a single index is avoided, and comprehensive information is provided for the evaluation of the eye disease state. And acquiring and analyzing historical diagnosis and treatment records of the eye diseases, and comprehensively tracking the change of the illness state of the patient. By comparing the current eye disease state with the historical eye disease state, the treatment effect can be intuitively estimated, and a basis is provided for the adjustment of the treatment scheme. The situation that symptoms cannot be cured in time due to the fact that patients cannot be intuitively judged due to overlong time intervals when a doctor carries out re-diagnosis is avoided.

Optionally, the acquiring the eye monitoring video shot by the camera device includes:

acquiring and analyzing an initial eye image, and determining whether eye deviation exists according to an image analysis result;

if the eye deviation exists, determining an eye deviation angle according to an image analysis result;

and determining eye differences according to the eye deviation angles and the history eye disease diagnosis and treatment records, adjusting a picture grabbing mode according to the eye differences, and acquiring the eye monitoring video according to the adjusted picture grabbing mode.

According to the scheme, the acquired initial eye image is analyzed through an image analysis technology, whether the eye is deviated or not is determined, the eye position and the wearing position of the eye detector are primarily judged, the eye deviation angle is determined, then differential determination is carried out on the image of the eye position based on the eye deviation angle and the history eye disease diagnosis and treatment record, and preparation is made for subsequent adjustment of an image picture. Through confirming the picture and snatching the mode, adjust shooting area, ensure that the patient is wearing when insufficient standard, also can carry out accurate snatching, reduce the error of video, improve the accuracy of follow-up eye condition state judgement.

Optionally, the eye detector comprises an imaging device, the acquiring and analyzing the initial eye image, and determining whether the eye deviation exists according to the image analysis result comprises the following steps:

Monitoring displacement changes of the eye detector in real time, and determining whether the eye detector has space position changes according to the displacement changes;

if the spatial position change is determined to exist, determining the horizontal position change amplitude according to the spatial position change;

determining whether to start the image pickup device according to the change amplitude of the horizontal position;

if the starting of the image pickup device is determined, acquiring and screening a shooting picture of the image pickup device to obtain the initial eye image;

analyzing the initial eye image, and extracting eye features according to an image analysis result;

And determining a structural boundary according to the eye features, analyzing the eye features in each structure according to the structural boundary, and determining whether eye deviation exists.

Through this scheme, through integrated high accuracy sensor and real-time data processing technique, the displacement that can real-time supervision eye detector changes, when eye detector takes place the spatial position change, according to the range of spatial position change, the instant decision whether starts camera shooting device and shoot, need not manual intervention just can realize automatic camera shooting device and start. And screening the pictures shot by the camera device to ensure that a clear and accurate initial eye image is acquired. And then, the eye features are extracted by using an advanced image recognition technology, so that the analysis accuracy is further improved. After extraction of ocular features, structural analysis helps to more accurately locate and analyze ocular deviation phenomena by determining individual structural boundaries of the eye from these features.

Optionally, the analyzing the eye feature in each structure according to the structure boundary to determine whether the eye deviation exists includes:

Determining an eyelid area from the structure boundaries and the ocular features within each structure;

analyzing the eye characteristics of the eyelid area and determining an eyelid shadow area;

determining eyelid shadow areas according to the eyelid shadow areas;

Acquiring the current illumination condition and the space position of the camera device;

determining an expected eyelid shadow area according to the current illumination condition and the spatial position;

comparing the eyelid shadow area with the desired eyelid shadow area, determining whether the eyelid shadow area is greater than the desired eyelid shadow area;

and if the eyelid shading area is larger than the expected eyelid shading area, determining that the eye deviation exists.

By the scheme, the eye characteristics in each structure are carefully analyzed, the eyelid area is accurately positioned, and the change of the eye characteristics can be accurately captured. The scheme not only focuses on the eye characteristics, but also considers the current illumination condition and the space position of the camera device. By incorporating these environmental factors into the analysis, the desired eyelid shadow area is more fully assessed. The actual eyelid shadow area is compared with the expected eyelid shadow area, and whether the eye is deviated or not is intelligently judged. The automatic judgment mode reduces errors and uncertainties caused by human factors.

Optionally, the analyzing the eye monitoring video to determine an eye feature includes:

analyzing the eye monitoring video to determine the brightness change;

determining a current detection mode according to the brightness change;

And according to the current detection mode, analyzing the eye monitoring video frame by frame to determine eye characteristics.

According to the scheme, the eye monitoring video is analyzed, the detection mode is determined by combining the brightness change, and the eye characteristics can be more effectively identified through the refined frame-by-frame analysis and the dynamically adjusted detection mode, so that the false alarm and the missing report are reduced. And the overall performance and the user experience are improved.

Optionally, the analyzing the eye feature to determine the opening and closing degree of the eyelid and the eyeball movement condition includes:

analyzing the eye features to determine eyelid contours and eyeball areas;

Determining a relative position between the eyelid contour and the eyeball area according to the feature vector of the eye feature;

According to the relative position, determining an eyelid opening and closing ratio, and according to the opening and closing ratio, determining the eyelid opening and closing degree;

determining the pupil position in each frame of video picture according to the eye characteristics;

Determining pupil change speed, pupil change direction and pupil change acceleration according to the pupil position in each frame of video picture;

and determining the eyeball movement condition according to the pupil change speed, the pupil change direction and the pupil change acceleration.

According to the method and the device, the relative position between the eyelid outline and the eyeball area is calculated, the opening and closing ratio of the eyelid is accurately determined, the opening and closing degree of the eyelid is accurately judged, the pupil position in each frame of video picture is tracked in real time, and the possibility is provided for calculation of the pupil change speed, direction and acceleration. Comprehensively considering the change speed, direction and acceleration of the pupil, realizing comprehensive evaluation of the scaling condition of the eyeball, obtaining more accurate eyeball movement condition, and being more convenient for subsequent detection and diagnosis of the eye disease of the patient.

Optionally, the determining the pupil change condition according to the eyeball motion condition includes:

Acquiring an environment change condition in real time based on the eyeball movement condition;

Determining a pupil inspection mode according to the environment change condition;

determining the real-time position of the pupil according to the eyeball movement condition;

Determining a pupil real-time diameter based on the pupil real-time location and the eye feature;

and determining pupil change conditions according to the pupil inspection mode and the real-time pupil diameter.

According to the technical scheme, through real-time monitoring of the eyeball movement condition, the light of the surrounding environment under the condition of eye scaling is rapidly captured, and the pupil inspection mode is determined based on the determined surrounding environment change condition, so that analysis of the pupil change condition can be closely related to different inspection modes, and follow-up patient disease analysis is facilitated. In addition, the eye examination requirements of the patient are met. The change of the pupil diameter can directly reflect the response degree of the pupil to different environmental stimuli, and is helpful for evaluating the vision adjustment capability of a patient, so that the condition of the patient can be analyzed more conveniently.

Optionally, the eyeball motion condition is determined according to the pupil change speed, the pupil change direction and the pupil change acceleration, and the following formula is referred to:

;

Wherein, the Representing an eyeball movement index; representing the pupil change rate; a weight coefficient representing the pupil change speed; representing the average value of the pupil change speed in a preset detection period; a standard deviation representing the pupil change rate; A weight coefficient representing the pupil variation direction; Representing the quantified pupil variation direction; A weight coefficient representing the pupil change acceleration; Representing the pupil change acceleration; Indicating the average value of the pupil variation acceleration in a preset detection period; A standard deviation representing the pupil variation acceleration;

And determining the eyeball motion according to the eyeball motion index.

According to the scheme, pupil change speed, pupil change direction and pupil change acceleration are considered, the eyeball scaling index is calculated by the design formula, so that the eyeball scaling index is utilized to reflect the eyeball movement condition, and on one hand, the evaluation of the eyeball movement condition can be more comprehensive and accurate. On the other hand, the design formula can digitize complex content, so that the eyeball movement condition is reflected more simply.

Optionally, the determining the current eye disease state according to the eyelid opening and closing degree, the eyeball motion condition and the pupil change condition includes:

determining last check data according to the current detection mode;

determining the opening and closing degree of last eyelid, the scaling degree of last eyeball and the change condition of last pupil according to the last examination data;

determining an eye change trend according to the eyelid opening and closing degree, the eyeball movement condition, the pupil change condition, the last eyelid opening and closing degree, the last eyeball scaling degree and the last pupil change condition;

And determining the current eye disease state according to the eye change trend.

According to the scheme, the data of multiple dimensions such as eyelid opening and closing degree, eyeball movement condition, pupil change condition and the like are integrated, and a solid foundation is provided for comprehensive evaluation of eye states. And comparing the current detection data with the last detection data, and accurately capturing the change trend of the eye state. The current eye disease state can be determined by analyzing the eye change trend, and a basis is provided for the follow-up judgment of whether the eye disease is improved.

In a second aspect, the present application provides a system for monitoring the progression of thyroid-related eye disease, the system comprising:

the characteristic determining module is used for acquiring eye monitoring videos shot by the camera device, analyzing the eye monitoring videos and determining eye characteristics;

The feature analysis module is used for analyzing the eye features and determining the opening and closing degree of eyelids and the eyeball movement condition;

The pupil analysis module is used for determining pupil change conditions according to the eyeball movement conditions;

the state analysis module is used for determining the current eye disease state according to the eyelid opening and closing degree, the eyeball movement condition and the pupil change condition;

the disease analysis module is used for acquiring and analyzing the history eye disease diagnosis and treatment record to obtain the history eye disease state, comparing the current eye disease state with the history eye disease state and determining whether the thyroid-related eye disease is improved or not.

Optionally, the feature determining module is specifically configured to:

Optionally, the eye detector comprises an imaging device, and the characteristic determining module is specifically configured to:

Optionally, the feature determining module is specifically configured to:

determining eyelid shadow areas according to the eyelid shadow areas;

Optionally, the feature determining module is specifically configured to:

analyzing the eye monitoring video to determine the brightness change;

determining a current detection mode according to the brightness change;

Optionally, the feature analysis module is specifically configured to:

analyzing the eye features to determine eyelid contours and eyeball areas;

Optionally, the pupil analysis module is specifically configured to:

Optionally, the feature analysis module is specifically configured to:

;

And determining the eyeball motion according to the eyeball motion index.

Optionally, the state analysis module is specifically configured to, when determining the current eye disease state according to the eyelid opening and closing degree, the eyeball motion condition, and the pupil change condition:

determining last check data according to the current detection mode;

Drawings

In order to more clearly illustrate the embodiments of the present application or the technical solutions of the prior art, the drawings that are needed in the embodiments or the description of the prior art will be briefly described below, it will be obvious that the drawings in the following description are some embodiments of the present application, and that other drawings can be obtained according to these drawings without inventive effort to a person skilled in the art.

Fig. 1 is a schematic view of an application scenario provided in an embodiment of the present application;

FIG. 2 is a flow chart of a method for monitoring the progression of thyroid-related eye disease according to an embodiment of the present application;

Fig. 3 is a schematic structural diagram of a system for monitoring the progress of thyroid-related eye diseases according to an embodiment of the present application.

Detailed Description

In order to make the objects, technical solutions and advantages of the embodiments of the present application more clear, the technical solutions of the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application. It will be apparent that the described embodiments are some, but not all, embodiments of the application. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to be within the scope of the application.

In addition, the term "and/or" is merely an association relation describing the association object, and means that three kinds of relations may exist, for example, a and/or B, and that three kinds of cases where a exists alone, while a and B exist alone, exist alone. In this context, unless otherwise specified, the term "/" generally indicates that the associated object is an "or" relationship.

Embodiments of the application are described in further detail below with reference to the drawings.

The subsequent determination of the treatment progress of the thyroid-related eye disease needs to be carried out to a hospital for evaluation by a doctor, and the process is very troublesome for a patient, and for the doctor, the thyroid-related eye disease is more serious if any problem exists in the middle of the process due to a certain time interval between the first diagnosis and the second review, so that the diagnosis after the second examination is similar to the first diagnosis, and is time-consuming and labor-consuming, causes are not found, the illness state of the patient is delayed, and the health of the patient is endangered.

Based on the above, the application provides a system and a method for monitoring the progress of thyroid-related eye diseases, which are used for acquiring an eye monitoring video shot by a camera device in real time and capturing eye state changes in real time. The video analysis and image processing technology provides objective eye feature data, and reduces subjectivity of artificial judgment. The eyelid opening and closing degree, eyeball movement condition and pupil change are accurately measured by using an advanced image processing algorithm, so that the accuracy and reliability of diagnosis are improved. The comprehensive judgment is carried out by combining a plurality of eye features, so that the limitation of a single index is avoided, and comprehensive information is provided for the evaluation of the eye disease state. And acquiring and analyzing historical diagnosis and treatment records of the eye diseases, and comprehensively tracking the change of the illness state of the patient. By comparing the current eye disease state with the historical eye disease state, the treatment effect can be intuitively estimated, and a basis is provided for the adjustment of the treatment scheme. The situation that symptoms cannot be cured in time due to the fact that patients cannot be intuitively judged due to overlong time intervals when a doctor carries out re-diagnosis is avoided.

Fig. 1 is a schematic diagram of an application scenario provided by the present application, where the method provided by the present application may be applied when it is required to monitor the treatment progress of a thyroid-related eye disease.

Specifically, the method provided by the application is applied to any eye detector, the eye detector interacts with a server of a main doctor, and an imaging device and a data processing chip are arranged in the eye detector. The method comprises the steps of acquiring an eye monitoring video shot by a camera device, transmitting the eye monitoring video to a data processing chip for analysis, determining and analyzing eye characteristics, determining eyelid opening and closing degree and eyeball movement condition, further analyzing eyeball movement condition, determining pupil change condition, combining eyelid opening and closing degree, eyeball movement condition and pupil change condition, determining the current eye disease state, retrieving and analyzing the history eye disease diagnosis and treatment record of the patient in a server mainly used for doctors to obtain the history eye disease state, comparing the current eye disease state with the history eye disease state, and determining whether the thyroid-related eye disease is improved or not. The video analysis and image processing technology provides objective eye feature data, and reduces subjectivity of artificial judgment. The comprehensive judgment is carried out by combining a plurality of eye features, so that the limitation of a single index is avoided, and comprehensive information is provided for the evaluation of the eye disease state.

Reference may be made to the following examples for specific implementation.

Fig. 2 is a flowchart of a method for monitoring the progress of thyroid-related eye diseases according to an embodiment of the present application, where the method of the present embodiment may be applied to an eye detector in the above scenario. As shown in fig. 2, the method includes:

s201, acquiring an eye monitoring video shot by the camera device, analyzing the eye monitoring video, and determining eye characteristics.

The eye detector may be understood as an instrument worn in the eye area, having a physical activation button, and the patient performing eye detection with the eye detector may be able to select at his own node to activate, for example after wearing is completed. When the physical button is pressed to send out a starting signal, the image pickup device is automatically started to pick up the images, and all videos generated in the shooting process can be filtered to remove useless frame images, such as images of non-eye areas shot by the patient when the patient is not wearing the device. And obtaining the reddened eye monitoring video after screening, analyzing the eye monitoring video by utilizing an image analysis technology, determining an eye area and extracting eye characteristics.

In some implementation means, a strain sensor and a gyroscope can be installed in the bandage of the eye detector, the strain sensor senses stretching or compression of the bandage, so that the tightness degree of the bandage is determined, in general, a patient is changed from wearing the bandage to wearing the bandage from an original state, the strain sensor receives the stretching data of the bandage and can be considered as wearing the bandage by the patient, and at the moment, the strain sensor is matched with the gyroscope in the eye detector to sense whether the eye detector is in a stable state or not, namely, the strain sensor is changed from a shaking state to a static state, so that the process of wearing the bandage by a person is described. If the video camera is in a stable state, a starting signal can be sent to the camera device to acquire the video.

S202, analyzing the eye characteristics and determining the opening and closing degree of eyelids and the eyeball movement condition.

Analyzing the obtained eye characteristics, determining eyelid areas and eyeball areas, distinguishing the eye characteristics according to frame images, determining the change of the eyelid area characteristics through the analysis of the pixel change of the eye characteristics of each frame image, and calculating the distance from the eyelid edge to the center of the eyeball in each frame image so as to obtain the eyelid opening and closing degree. In a specific implementation, the opening and closing degree of the eyelid may also be determined by using an index such as an eye aspect ratio.

Similarly, after the frame image is distinguished from the eye feature in the above manner, the eye movement condition is determined by tracking the position and the range formed by the pixel points of the eye region of each frame image.

S203, determining pupil change conditions according to eyeball movement conditions.

After the eyeball movement condition is obtained, further characteristic pixel distinction is carried out on the eyeball area in each frame of picture, eyeball pixels and pupil pixels are determined, the size of an exit pupil is determined through the pupil pixels, and then the change of the size of each frame is analyzed frame by frame, so that the change condition of the pupil is determined.

S204, determining the current eye disease state according to the eyelid opening and closing degree, eyeball movement condition and pupil change condition.

After the eyelid opening and closing degree, eyeball movement condition and pupil change condition are obtained through the above mode, firstly, according to the basic information of the patient, the corresponding age range and the eye characteristics of the patient, such as the single eyelid, the eye distance, the eye length and the like, are determined, according to the characteristics, the normal eye state corresponding to the single eyelid, the eye distance and the like is called from the preset database, and then the eyelid opening and closing degree, the eyeball movement condition and the pupil change condition corresponding to the normal eye state are respectively compared with the current eyelid opening and closing degree, the eyeball movement condition and the pupil change condition one by one, so as to determine whether the eye disease and the current eye disease state exist at present.

In a specific implementation manner, the preset database can also store eye characteristics of a plurality of known eye diseases under different degrees, when the difference from the normal eye state is determined through the steps, the current eyelid opening and closing degree, eyeball movement condition and pupil change condition can be utilized to select data matched with the state of the known eye diseases from the eye characteristics of the known eye diseases under different degrees, and the eye diseases and the illness degree which are more corresponding to the state are used as the current eye state of the patient.

S205, acquiring and analyzing a history eye disease diagnosis and treatment record to obtain a history eye disease state, and comparing the current eye disease state with the history eye disease state to determine whether the thyroid-related eye disease is improved.

And calling the history eye disease diagnosis and treatment record of the patient through a server of a communicated main doctor. In a specific implementation, when the attending doctor diagnoses the patient and confirms that the patient needs to use the eye detector to monitor the continuous progress of the eye disease, the record of the diagnosis is directly transmitted to the corresponding eye detector. The diagnosis records can be used as historical eye disease diagnosis and treatment records and stored in the preset database, and the historical eye disease diagnosis and treatment records can be directly retrieved from the preset database at the moment.

And performing natural language processing on the obtained history eye disease diagnosis and treatment record, and determining the history eye disease state described in the record. Comparing the historical eye disease state with the current eye disease state, and determining whether the eye disease state is improved, for example, the eyelid opening and closing range is enlarged, which indicates that the eye disease state is improved.

Through the scheme provided by the embodiment, the eye monitoring video shot by the camera device is obtained in real time, so that the real-time capturing of the eye state change is realized. The video analysis and image processing technology provides objective eye feature data, and reduces subjectivity of artificial judgment. The eyelid opening and closing degree, eyeball movement condition and pupil change are accurately measured by using an advanced image processing algorithm, so that the accuracy and reliability of diagnosis are improved. The comprehensive judgment is carried out by combining a plurality of eye features, so that the limitation of a single index is avoided, and comprehensive information is provided for the evaluation of the eye disease state. And acquiring and analyzing historical diagnosis and treatment records of the eye diseases, and comprehensively tracking the change of the illness state of the patient. By comparing the current eye disease state with the historical eye disease state, the treatment effect can be intuitively estimated, and a basis is provided for the adjustment of the treatment scheme. The situation that symptoms cannot be cured in time due to the fact that patients cannot be intuitively judged due to overlong time intervals when a doctor carries out re-diagnosis is avoided.

In some embodiments, an initial eye image is acquired and analyzed, whether eye deviation exists is determined according to an image analysis result, if the eye deviation exists, an eye deviation angle is determined according to an image analysis result, eye differences are determined according to the eye deviation angle and a history of eye disease diagnosis and treatment record, a picture grabbing mode is adjusted according to the eye differences, and an eye monitoring video shot by a camera device is acquired according to the adjusted picture grabbing mode.

The initial eye image may be considered as the first frame of picture taken after the camera is started.

The picture capturing mode can be considered as that a self-correction picture technology in the equipment is utilized to carry out layout adjustment on pictures with offset, so that video capturing is realized without error when an eye detector is not required to be adjusted.

Specifically, after the image capturing device is started in the above manner, the image capturing device may be set to capture an image preferentially, that is, an initial eye image, which is used to analyze the initial eye image to determine whether there is a positional deviation in wearing of the patient. At this time, the position of the eye corresponding to the wearing without position deviation in the image can be determined according to the history wearing record, and then under the condition of not correcting, if the eye is not completely at the position, the eye deviation can be considered to exist. And when the deviation is determined, a plane rectangular coordinate system is established according to the relation between the two corresponding positions, and the angle between the two is determined to obtain the eye deviation angle. In a specific implementation, the position of the eye in the image corresponding to when the wearing has no positional deviation may be obtained by the attending physician wearing the eye before the first wearing. This location may be stored in a historical eye disease diagnosis and treatment record.

When the eye deviation exists, the eye position in the initial eye image can be compared with the position of the eye in the image corresponding to the position when the wearer does not wear the eye deviation, so that the eye difference is determined, automatic correction is carried out based on the eye difference, after correction, the eye position in the initial eye image is determined to be consistent with the position of the eye in the image corresponding to the position when the wearer does not wear the eye deviation, and shooting and obtaining of an eye monitoring video can be carried out.

In a specific implementation manner, if the eye detector is not worn for the first time, that is, after the patient uses the usage data for multiple times, the eye difference determination can be performed based on the angle of the previous wearing time.

According to the scheme provided by the embodiment, the acquired initial eye image is analyzed through an image analysis technology, whether the eye is deviated or not is determined, the eye position and the wearing position of the eye detector are primarily judged, the eye deviation angle is determined, and then the difference determination is carried out on the image of the eye position based on the eye deviation angle and the history eye disease diagnosis and treatment record, so that preparation is made for the adjustment of a subsequent image picture. Through confirming the picture and snatching the mode, adjust shooting area, ensure that the patient is wearing when insufficient standard, also can carry out accurate snatching, reduce the error of video, improve the accuracy of follow-up eye condition state judgement.

In some embodiments, the displacement change of the eye detector is monitored in real time, whether the eye detector has a spatial position change is determined according to the displacement change, if the spatial position change is determined, the horizontal position change amplitude is determined according to the spatial position change, whether the camera is started or not is determined according to the horizontal position change amplitude, if the camera is started, a shooting picture of the camera is acquired and screened to obtain an initial eye image, the initial eye image is analyzed, eye features are extracted according to an image analysis result, a structure boundary is determined according to the eye features, the eye features in each structure are analyzed according to the structure boundary, and whether eye deviation exists or not is determined.

A high-precision gyroscope can be integrated in the eye detector, so that the position change of the eye detector is analyzed in real time to obtain the displacement change of the eye detector, and whether the eye detector has the spatial position change is judged, at the moment, a person possibly wants to pick up and observe, but does not want to carry out eye examination, therefore, the change amplitude of the horizontal position needs to be determined through the change of the spatial position, for example, a spatial rectangular coordinate system is established according to the position of the eye detector before the change, and then the change of the spatial coordinate is determined according to the continuous change of the spatial position, so that the final change amplitude of the horizontal position is obtained according to the change of the spatial coordinate.

When a certain level of position variation amplitude is reached, further analysis of the variation amplitude can be performed, such as whether there is an up-and-down fluctuation of the position, or other position variation. If the horizontal position rises straight and becomes stable at a certain point in time, it is considered that an eye examination may be necessary at this time, and when this occurs, it is determined to automatically start the image pickup apparatus.

After the camera device is started, the camera device can perform continuous image shooting, at the moment, the shot images can be screened to obtain one or more complete eye images related to eyes, and the complete eye images are further screened to obtain the complete eye image with the highest definition, and the complete eye image is used as an initial eye image.

After the initial eye image is obtained, the eye features are extracted by utilizing an image analysis technology, so that each structural boundary of the eye is determined according to the spatial distribution and morphological features of the features, more detailed eye feature analysis is performed, and whether the eye deviation exists or not is determined.

Through the scheme that this embodiment provided, through integrated high accuracy sensor and real-time data processing technique, the displacement that can real-time supervision eye detector changes, when eye detector takes place the spatial position change, according to the range of spatial position change, the instant decision whether starts camera shooting, need not manual intervention just can realize automatic camera shooting device and start. And screening the pictures shot by the camera device to ensure that a clear and accurate initial eye image is acquired. And then, the eye features are extracted by using an advanced image recognition technology, so that the analysis accuracy is further improved. After extraction of ocular features, structural analysis helps to more accurately locate and analyze ocular deviation phenomena by determining individual structural boundaries of the eye from these features.

In some embodiments, an eyelid area is determined from a structure boundary and eye features within each structure, an eye feature of the eyelid area is analyzed, an eyelid shadow area is determined from the eyelid shadow area, a current illumination condition and a spatial position of the camera device are obtained, an expected eyelid shadow area is determined from the current illumination condition and the spatial position, the eyelid shadow area is compared with the expected eyelid shadow area, whether the eyelid shadow area is greater than the expected eyelid shadow area is determined, and if the eyelid shadow area is greater than the expected eyelid shadow area, an eye shift is determined to exist.

The desired eyelid shading area may be an area of eyelid shading of a patient who performs an eye examination using the eye detector described above when the patient is wearing correctly and the environmental conditions are consistent, and may be determined from a screen obtained when the patient is wearing the eye detector by the doctor in the above embodiment.

Using the structural boundaries and the ocular features within each structure obtained in the above embodiments, it is determined what each structure belongs to and from that which part belongs to the eyelid area. Eye characteristics of the eyelid area are analyzed, including eyelid position and eye-see shape characteristics, to calculate a shadow area of the eyelid area using pixel counting.

The method comprises the steps of obtaining the current illumination condition of an initial eye image when photographing and imaging is carried out by arranging an illumination sensor in an eye detector, and then determining eyelid shielding and an expected eyelid shadow area if the eyelid is in normal wearing when photographing the initial eye image by combining the space position of an imaging device. This can be predicted by deep learning in combination with the situation when the patient is wearing normally for the first time.

In a specific implementation, the illumination intensity and illumination direction in the initial eye image frame may also be determined using image processing techniques.

After the expected eyelid shading area is determined, the eyelid shading area is compared with the eyelid shading area obtained in practice, and if the eyelid shading area is larger than the expected eyelid shading area, the eye deviation is determined.

In a specific implementation, an area threshold may also be set, and if the eyelid shading area is greater than the desired eyelid shading area and the difference between the two is greater than this area threshold, then it may be considered that an eye shift is present.

By the scheme provided by the embodiment, the eye characteristics in each structure are carefully analyzed, the eyelid area is accurately positioned, and the change of the eye characteristics can be accurately captured. The scheme not only focuses on the eye characteristics, but also considers the current illumination condition and the space position of the camera device. By incorporating these environmental factors into the analysis, the desired eyelid shadow area is more fully assessed. The actual eyelid shadow area is compared with the expected eyelid shadow area, and whether the eye is deviated or not is intelligently judged. The automatic judgment mode reduces errors and uncertainties caused by human factors.

In some embodiments, an eye monitoring video is analyzed to determine a change in shading, a current detection mode is determined based on the change in shading, and an eye feature is determined based on the current detection mode by analyzing the eye monitoring video from frame to frame.

The current detection mode may be an eye detection mode in which the patient is currently in progress. In order to achieve a more comprehensive detection, several detection modes may be preset. Different detection modes have different characteristics, such as obvious light and shade alternation when pupil detection is performed.

Specifically, after the eye monitoring video is obtained, each frame of image of the eye monitoring video is calculated, and the global average brightness or histogram distribution of each frame of image is determined so as to evaluate the overall brightness change. A threshold value for the change of brightness is set according to the whole brightness range of the video. When the luminance difference between adjacent frames exceeds the threshold, a significant change in brightness is considered to occur. And determining the current detection mode according to the possible sensitive changes of a plurality of preset detection modes and the brightness change in each current frame of image. After the current detection mode is determined, according to the characteristics of the current detection mode, frame-by-frame analysis is carried out, and the characteristics of eyes in the current detection mode are determined.

According to the scheme provided by the embodiment, the eye monitoring video is analyzed, the detection mode is determined by combining the brightness change, and the eye characteristics can be more effectively identified through the refined frame-by-frame analysis and the dynamically adjusted detection mode, so that the false alarm and the false alarm condition are reduced. And the overall performance and the user experience are improved.

In some embodiments, the eye features are analyzed to determine eyelid contours and eye regions, relative positions between the eyelid contours and the eye regions are determined according to feature vectors of the eye features, opening and closing ratios of the eyelids are determined according to the relative positions, opening and closing degrees of the eyelids are determined according to the opening and closing ratios, pupil positions in each frame of video frame are determined according to the eye features, pupil change speeds, pupil change directions and pupil change accelerations are determined according to pupil positions in each frame of video frame, and eye movement conditions are determined according to the pupil change speeds, pupil change directions and pupil change accelerations.

The opening/closing ratio of the eyelid can be understood as the ratio of the open portion of the eyelid to the fully open state of the eyelid with the eye in the open state.

The eyelid opening and closing degree can be understood as the current eyelid opening and closing degree compared with the eyelid opening and closing degree when the patient is in the normal eye state.

Specifically, the eye features obtained in the above embodiments are processed first, so as to reduce errors. The method comprises the steps of identifying eyelid removing areas by utilizing an edge detection algorithm, defining eyelid outlines, classifying all eye features by utilizing a color separation mode, carrying out texture analysis on the classified eye features, determining relevant eyeball features, and determining eyeball removing areas according to the colors and textures of the features.

Based on the determined eyelid outline and eyeball area, converting the feature vector of the eye feature at the corresponding position, comparing the feature vector of the eyelid outline with the feature vector of the eyeball area, and calculating to obtain the relative position relationship between the eyelid outline and the eyeball area, such as distance, angle and the like.

In order to better judge the eyelid opening and closing degree, an opening and closing ratio threshold value can be set in advance according to the eye condition of a patient, and a plurality of threshold values can be set to reflect the eyelid opening and closing degree under different eyelid opening and closing ratio conditions. Such as fully open, semi-open, closed, etc.

According to the eye characteristics corresponding to the eyeball area, the position of the pupil is positioned by utilizing the characteristics of the gray scale, the shape and the like of the pupil, or the pupil position in each frame of video picture is determined according to the change of the characteristic vector corresponding to the eye characteristics in each frame of video picture, the pupil change speed, the pupil change direction and the pupil change acceleration are determined based on the pupil position in each frame of video picture and the change speed of each frame of picture, and the eyeball movement condition is determined according to the pupil change speed, the pupil change direction and the pupil change acceleration.

Wherein the pupil change speed can be obtained by calculating the movement distance of the pupil in the continuous frame picture divided by the time interval corresponding to the change speed of each frame picture, assuming that the pupil is at the first positionThe position of the frame isIn the first placeThe position of the frame isAt a time interval ofPupil change rateThe calculation can be referred to as the following formula (1):

(1)

in addition, the pupil change direction can be determined by the change trend of the pupil position. Which may include moving toward the center and out-diffusion, quantifying the trend of the pupil position, assuming a threshold When the threshold value is smaller, the pupil is moved towards the center, and the pupil can be considered to be diffusing outwards in other cases, and the calculation can be specifically performed by referring to the following formula (2):

(2)

Pupil change acceleration the pupil position change acceleration may be calculated by a second derivative or differential method. Specifically, the following formula (3) can be referred to:

(3)

According to the scheme provided by the embodiment, the relative position between the eyelid outline and the eyeball area is calculated, the opening and closing ratio of the eyelid is accurately determined, the opening and closing degree of the eyelid is accurately judged, the pupil position in each frame of video picture is tracked in real time, and the possibility is provided for calculation of the pupil change speed, direction and acceleration. Comprehensively considering the change speed, direction and acceleration of the pupil, realizing comprehensive evaluation of the scaling condition of the eyeball, obtaining more accurate eyeball movement condition, and being more convenient for subsequent detection and diagnosis of the eye disease of the patient.

In some embodiments, the environmental change condition is obtained in real time based on the eye movement condition, a pupil inspection mode is determined according to the environmental change condition, a pupil real-time position is determined according to the eye movement condition, a pupil real-time diameter is determined based on the pupil real-time position and the eye feature, and the pupil change condition is determined according to the pupil inspection mode and the pupil real-time diameter.

Based on the obtained eye movement situation, each time the eye changes, the current pupil inspection mode is determined by capturing the current environment change situation, such as the illumination situation, based on the environment change situation and the detection modes set in the above embodiment. And then determining the real-time position of the pupil at each moment according to the eyeball movement condition, namely, which position of the pupil is at the eyes at the current moment.

Thus, the diameter of the pupil at the current moment, namely the real-time diameter of the pupil, is determined by combining the real-time position of the pupil and the eye characteristics at the current moment. And determining pupil change conditions according to the pupil inspection mode and the real-time pupil diameter.

In a specific implementation, the condition of the patient can be utilized to establish a relationship between the eyeball scaling characteristics and the environmental factors, and then a model is established to judge the environmental change condition under different eye scaling conditions. In other implementations, the current image may also be analyzed to determine the distribution of light in the image, and then, in combination with a sensor disposed inside the eye detector, determine the illumination intensity and the light distribution of the environment where the patient is located, so as to infer the environment change condition.

According to the scheme provided by the embodiment, through real-time monitoring of the eyeball movement condition, the light of the surrounding environment under the condition of eye scaling is rapidly captured, and the pupil inspection mode is determined based on the determined surrounding environment change condition, so that analysis of the pupil change condition can be closely related to different inspection modes, and the analysis of the follow-up patient condition is facilitated. In addition, the eye examination requirements of the patient are met. The change of the pupil diameter can directly reflect the response degree of the pupil to different environmental stimuli, and is helpful for evaluating the vision adjustment capability of a patient, so that the condition of the patient can be analyzed more conveniently.

In some embodiments, the eye movement index is determined based on the pupil change speed, the pupil change direction, and the pupil change acceleration, and the eye movement condition is determined based on the eye movement index.

According to the pupil change speed, the pupil change direction and the pupil change acceleration, the eyeball movement condition is determined, and the following formula (4) is referred to:

(4)

Wherein, the Representing an eyeball movement index; Indicating pupil change rate; A weight coefficient representing the pupil change speed; representing the average value of the pupil change speed in a preset detection period; Standard deviation representing pupil variation rate; A weight coefficient representing the pupil variation direction; representing the quantized pupil change direction; a weight coefficient representing pupil change acceleration; indicating pupil change acceleration; Indicating the average value of the pupil variation acceleration in a preset detection period; standard deviation representing pupil variation acceleration;

According to the scheme provided by the embodiment, the pupil change speed, the pupil change direction and the pupil change acceleration are considered, the eyeball zoom index is calculated by the design formula, so that the eyeball motion condition is reflected by the eyeball zoom index, and on one hand, the evaluation of the eyeball motion condition can be more comprehensive and accurate. On the other hand, the design formula can digitize complex content, so that the eyeball movement condition is reflected more simply.

In some embodiments, last inspection data is determined according to a current detection mode, last eyelid opening and closing degree, last eyeball scaling degree and last pupil change condition are determined according to the last inspection data, eye change trend is determined according to eyelid opening and closing degree, eyeball movement condition, pupil change condition, last eyelid opening and closing degree, last eyeball scaling degree and last pupil change condition, and current eye disease state is determined according to the eye change trend.

The last examination data may be understood as examination data obtained by the last eye examination before the current examination, and may include the last eyelid opening and closing degree, eyeball movement condition, pupil change condition, i.e. last eyelid opening and closing degree, last eyeball scaling degree and last pupil change condition.

Specifically, last inspection data matched with the current detection mode is extracted from a database according to the current detection mode, last inspection data are classified, the last eyelid opening and closing degree, last eyeball scaling degree and last pupil change condition are determined, the eyelid opening and closing degree, eyeball movement condition and pupil change condition at the current moment are respectively compared with the corresponding last eyelid opening and closing degree, last eyeball scaling degree and last pupil change condition, the trend of eye change compared with the last time, namely the eye change trend, in the current detection mode is determined, for example, the eyelid opening and closing degree is larger than the last eyelid opening and closing degree in the consent detection mode, the condition is improved if the eye opening and closing degree is gradually enlarged based on the condition of a patient, the eye change trend can be considered to be changed in a good direction at the moment, so that whether the current eye condition is improved compared with the last time or not is determined based on the eye change trend, and the current eye condition is determined by combining the condition diagnosed by a doctor again.

By the scheme provided by the embodiment, the data of multiple dimensions such as eyelid opening and closing degree, eyeball movement condition, pupil change condition and the like are integrated, and a solid foundation is provided for comprehensive evaluation of eye states. And comparing the current detection data with the last detection data, and accurately capturing the change trend of the eye state. The current eye disease state can be determined by analyzing the eye change trend, and a basis is provided for the follow-up judgment of whether the eye disease is improved.

Fig. 3 is a schematic structural diagram of a system for monitoring the progress of thyroid-related eye disease according to an embodiment of the present application, and as shown in fig. 3, the system 300 for monitoring the progress of thyroid-related eye disease according to the present embodiment includes a feature determining module 301, a feature analyzing module 302, a pupil analyzing module 303, a status analyzing module 304, and a disorder analyzing module 305.

The feature determining module 301 is configured to obtain an eye monitoring video captured by the imaging device, analyze the eye monitoring video, and determine an eye feature;

The feature analysis module 302 is configured to analyze the eye feature and determine an eyelid opening and closing degree and an eyeball movement condition;

the pupil analysis module 303 is configured to determine a pupil change condition according to the eyeball motion condition;

the state analysis module 304 determines the current eye disease state according to the eyelid opening and closing degree, the eyeball movement condition and the pupil change condition;

The condition analysis module 305 is used for acquiring and analyzing a history eye disease diagnosis and treatment record to obtain a history eye disease state, comparing the current eye disease state with the history eye disease state, and determining whether the thyroid-related eye disease is improved.

Optionally, the feature determining module 301 is specifically configured to:

Optionally, the eye detector comprises an imaging device, and the feature determining module 301 is specifically configured to:

Optionally, the feature determining module 301 is specifically configured to:

determining eyelid shadow areas according to the eyelid shadow areas;

Optionally, the feature determining module 301 is specifically configured to:

analyzing the eye monitoring video to determine the brightness change;

determining a current detection mode according to the brightness change;

Optionally, the feature analysis module is specifically configured to:

analyzing the eye features to determine eyelid contours and eyeball areas;

Optionally, the pupil analysis module 303 is specifically configured to:

Optionally, the feature analysis module 302 is specifically configured to:

;

And determining the eyeball motion according to the eyeball motion index.

Optionally, the state analysis module 304 is specifically configured to, when determining the current eye disease state according to the eyelid opening and closing degree, the eyeball motion condition, and the pupil variation condition:

determining last check data according to the current detection mode;

The system of the present embodiment may be used to perform the method of any of the foregoing embodiments, and its implementation principle and technical effects are similar, and will not be described herein.

Claims

1. A method for monitoring the progress of thyroid-related eye diseases is characterized by being applied to an eye detector, wherein the eye detector comprises an imaging device and a data processing chip, the method is applied to the data processing chip, and the method comprises the following steps:

comparing the current eye disease state with the historical eye disease state to determine whether the thyroid-related eye disease is improved or not;

the analyzing the eye monitoring video to determine eye characteristics includes:

analyzing the eye monitoring video to determine the brightness change;

determining a current detection mode according to the brightness change;

according to the current detection mode, analyzing the eye monitoring video frame by frame to determine eye characteristics;

The analyzing the eye characteristics to determine the opening and closing degree of eyelids and the movement condition of eyeballs comprises the following steps:

analyzing the eye features to determine eyelid contours and eyeball areas;

Determining the eyeball movement condition according to the pupil change speed, the pupil change direction and the pupil change acceleration;

the determining the current eye disease state according to the eyelid opening and closing degree, the eyeball movement condition and the pupil change condition comprises the following steps:

determining last check data according to the current detection mode;

2. The method of claim 1, wherein the acquiring the eye monitoring video captured by the imaging device comprises:

3. The method of claim 2, wherein the acquiring and analyzing the initial eye image and determining whether an eye shift exists based on the image analysis result comprises:

4. A method according to claim 3, wherein said analyzing the ocular characteristics within each structure based on the structure boundaries to determine if an ocular shift exists comprises:

determining eyelid shadow areas according to the eyelid shadow areas;

5. The method of claim 1, wherein said determining a pupil change condition from said eye movement condition comprises:

6. The method of claim 1, wherein the eye movement is determined from the pupil change speed, the pupil change direction, and the pupil change acceleration, with reference to the following formula:

;

Wherein E represents an eyeball motion index, V represents the pupil change speed, a represents a weight coefficient of the pupil change speed; representing the average value of the pupil change speed in a preset detection period; a standard deviation representing the pupil change rate; a weight coefficient representing the pupil variation direction; D represents the pupil variation direction after quantization; A weight coefficient representing the pupil change acceleration; a represents the pupil variation acceleration; Indicating the average value of the pupil variation acceleration in a preset detection period; A standard deviation representing the pupil variation acceleration;

And determining the eyeball motion according to the eyeball motion index.

7. A system for monitoring the progression of thyroid-related eye disease, applied to the method of any one of claims 1-6, comprising: