CN-122004748-A - Intelligent intraocular pressure dynamic glasses monitoring method and system based on multi-mode sensing

CN122004748ACN 122004748 ACN122004748 ACN 122004748ACN-122004748-A

Abstract

The invention provides an intelligent intraocular pressure dynamic eyeglass monitoring method and system based on multi-mode sensing, which relate to the technical field of eyeglass monitoring, wherein an eye sensor and a pose sensor arranged in eyeglasses are used for determining monitoring points and acquiring point multi-mode data to obtain point multi-mode acquisition data; performing multi-mode and intraocular pressure data influence analysis on the point multi-mode acquisition data to obtain multi-mode influence analysis data and intraocular pressure influence analysis data; according to the multi-mode intraocular pressure dynamic analysis by combining the multi-mode influence analysis data with the intraocular pressure influence analysis data, the multi-mode intraocular pressure dynamic monitoring data are obtained, and scene flexibility and accuracy are greatly improved through multi-dimensional data acquisition and fusion analysis.

Inventors

LIU CHAO
ZHANG XIANGYANG
WANG JIAYIN
JIA ENZE

Assignees

哈尔滨医科大学

Dates

Publication Date: 20260512
Application Date: 20260329

Claims (10)

1. The intelligent intraocular pressure dynamic eyeglass monitoring method based on multi-mode sensing is characterized by comprising the following steps of: s1, determining monitoring points and acquiring point multi-mode data through an eye sensor and a pose sensor which are arranged in glasses, so as to obtain point multi-mode acquisition data; s2, carrying out multi-mode and intraocular pressure data influence analysis on the point multi-mode acquisition data to obtain multi-mode influence analysis data and intraocular pressure influence analysis data; s3, carrying out multi-mode intraocular pressure dynamic analysis according to the multi-mode influence analysis data and the intraocular pressure influence analysis data to obtain multi-mode intraocular pressure dynamic monitoring data.
2. The method for monitoring the intraocular pressure dynamic spectacles based on multi-modal sensing according to claim 1, wherein the S1 comprises: The glasses are provided with an eye sensor and a pose sensor, and the eye sensor and the pose sensor are combined to obtain a sensor group; Acquiring multi-mode sensing data according to the sensor group to obtain multi-mode sensing acquisition data; And generating a plurality of monitoring points according to the monitoring area of the glasses, and dividing the multi-mode sensing acquisition data according to the information of the monitoring points to obtain point multi-mode acquisition data.
3. The method for monitoring the intelligent intraocular pressure dynamic glasses based on the multi-modal sensing according to claim 2, wherein the generating a plurality of monitoring points according to the glasses monitoring area, dividing the multi-modal sensing acquisition data according to the monitoring point information, obtaining the point multi-modal acquisition data comprises the following steps: Acquiring information of a glasses monitoring area according to a sensor group, and carrying out area characteristic identification according to the information of the glasses monitoring area to acquire area characteristic identification information; Generating a plurality of monitoring points according to the regional characteristic identification information; Carrying out data feature extraction of each monitoring point location on the multi-mode sensing acquisition data to obtain point location feature extraction data; Performing similarity calculation on the point feature extraction data to obtain point feature similarity data; And carrying out representative point location analysis and determination on the point location feature extraction data according to the point location feature similarity data, and further obtaining point location multi-mode acquisition data of the representative point location.
4. The method for monitoring intelligent intraocular pressure dynamic glasses based on multi-modal sensing according to claim 3, wherein the performing representative point location analysis determination on the point location feature extraction data according to the point location feature similarity data to obtain point location multi-modal collection data of the representative point location comprises: Comparing the point location feature similarity data with a point location feature similarity threshold value to obtain point location feature similarity comparison data; Determining a plurality of similar monitoring points according to the point feature similarity comparison data, and combining the similar monitoring points to obtain a similar point combination; obtaining an average value of point location feature similarity data of similar point location combinations, and obtaining a similar point location combination average value; obtaining the difference absolute value of the combined mean value of each point feature similarity data and the similar point, and obtaining a point feature difference coefficient; Sequencing a plurality of point location characteristic difference coefficients of similar point location combinations from small to large to obtain a similar difference sequence; Selecting a monitoring point position corresponding to a first-ranking point position characteristic difference coefficient in the similar difference sequence to determine as a representative monitoring point position; The multi-mode collected data representing the monitoring point positions is the point position multi-mode collected data.
5. The method for monitoring intelligent intraocular pressure dynamic spectacles of claim 1, wherein S2 comprises: dividing the point-bit multi-mode acquisition data according to preset time sequence information to obtain time sequence point position information; acquiring time sequence multi-mode acquisition data of each time sequence node representing the monitoring point according to the time sequence point information; performing multi-mode influence analysis according to the time sequence multi-mode acquisition data to obtain multi-mode influence analysis data; And carrying out intraocular pressure influence analysis according to the time sequence multi-mode acquisition data to obtain intraocular pressure influence analysis data.
6. The method for monitoring the intelligent intraocular pressure dynamic glasses based on multi-modal sensing according to claim 5, wherein the performing multi-modal impact analysis according to the time-series multi-modal collected data to obtain multi-modal impact analysis data comprises: carrying out multiple preset pose data extraction on the time-series multi-mode acquisition data to obtain multiple pose extraction data; comparing the plurality of pose extraction data with preset pose data to obtain a plurality of pose comparison data; Determining normal pose extraction data and abnormal pose extraction data according to the multiple pose comparison data; determining normal pose influence time sequence information and abnormal pose influence time sequence information according to the normal pose extraction data and the abnormal pose extraction data; the normal pose influence time sequence information and the abnormal pose influence time sequence information are multi-mode influence analysis data.
7. The method for monitoring the intraocular pressure dynamic glasses based on the multi-modal sensing according to claim 6, wherein determining the normal pose extraction data and the abnormal pose extraction data according to the plurality of pose comparison data comprises: obtaining the difference value between each pose extraction data and preset pose data according to the multiple pose comparison data, and obtaining each pose difference value data; Comparing the pose difference value data with a preset pose difference value threshold value to obtain a pose difference value comparison result; when the pose difference value data is larger than a preset pose difference value threshold value, judging that pose extraction data corresponding to the pose difference value data is abnormal pose extraction data; And when the pose difference value data is smaller than or equal to a preset pose difference value threshold value, judging that the pose extraction data corresponding to the pose difference value data is normal pose extraction data.
8. The method for monitoring the intelligent intraocular pressure dynamic glasses based on the multi-mode sensing according to claim 5, wherein the analyzing the intraocular pressure influence according to the time sequence multi-mode collected data to obtain the intraocular pressure influence analysis data comprises the following steps: Extracting intraocular pressure data from the time-series multi-mode acquisition data to obtain intraocular pressure extraction data; comparing the intraocular pressure extraction data with a preset intraocular pressure range to obtain an intraocular pressure comparison result; Performing intraocular pressure abnormality judgment on the eye pressure extraction data according to the intraocular pressure comparison result to obtain abnormal intraocular pressure judgment data; the abnormal intraocular pressure judging data is the intraocular pressure influence analysis data.
9. The method for monitoring intelligent intraocular pressure dynamic spectacles of claim 1, wherein S3 comprises: performing time sequence node association on the multimode influence analysis data and the intraocular pressure influence analysis data according to preset time sequence information to obtain time sequence multimode intraocular pressure influence data; acquiring multi-modal influence analysis data corresponding to the intraocular pressure influence analysis data according to the time sequence multi-modal intraocular pressure influence data to acquire associated multi-modal influence data; Clustering and combining the ocular pressure influence analysis data to obtain ocular pressure clustering and combining information; Clustering and combining the associated multi-mode influence data according to the intraocular pressure clustering and combining information to obtain multi-mode clustering and combining information; And combining the intraocular pressure cluster combination information with the multi-mode cluster combination information to obtain multi-mode intraocular pressure dynamic monitoring data.
10. Intelligent intraocular pressure dynamic eyeglass monitoring system based on multi-modal sensing, characterized in that the system comprises: The acquisition module is used for determining monitoring points and acquiring point multi-mode data through an eye sensor and a pose sensor which are arranged in the glasses, so as to obtain point multi-mode acquisition data; the influence analysis module is used for carrying out multi-mode and intraocular pressure data influence analysis on the point multi-mode acquisition data to obtain multi-mode influence analysis data and intraocular pressure influence analysis data; And the monitoring module is used for carrying out multi-mode intraocular pressure dynamic analysis according to the multi-mode influence analysis data and the intraocular pressure influence analysis data to obtain multi-mode intraocular pressure dynamic monitoring data.

Description

Intelligent intraocular pressure dynamic glasses monitoring method and system based on multi-mode sensing Technical Field The invention provides an intelligent intraocular pressure dynamic eyeglass monitoring method and system based on multi-mode sensing, relates to the technical field of eyeglass monitoring, and particularly relates to the technical field of intelligent intraocular pressure dynamic eyeglass monitoring based on multi-mode sensing. Background The traditional intraocular pressure monitoring method is represented by Goldmann applanation intraocular pressure, has the defects that the operation depends on professional medical staff, needs to be carried out in a hospital scene and cannot realize dynamic continuous monitoring, is difficult to capture fluctuation change of intraocular pressure in daily activities, and is easy to cause early lesion missed diagnosis. The existing portable intraocular pressure monitoring equipment mostly adopts a single sensing mode, is interfered by factors such as head gestures and eyeball movements, has insufficient measurement precision, lacks comprehensive analysis on multidimensional data, and cannot effectively eliminate the influence of interference factors on intraocular pressure results. At present, no mature scheme is available for organically combining eye physiological sensing with pose sensing to realize dynamic and anti-interference monitoring of intraocular pressure. Disclosure of Invention The invention provides an intelligent intraocular pressure dynamic eyeglass monitoring method and system based on multi-mode sensing, which are used for solving the problems: the invention provides an intelligent intraocular pressure dynamic eyeglass monitoring method and system based on multi-mode sensing, wherein the method comprises the following steps: s1, determining monitoring points and acquiring point multi-mode data through an eye sensor and a pose sensor which are arranged in glasses, so as to obtain point multi-mode acquisition data; s2, carrying out multi-mode and intraocular pressure data influence analysis on the point multi-mode acquisition data to obtain multi-mode influence analysis data and intraocular pressure influence analysis data; s3, carrying out multi-mode intraocular pressure dynamic analysis according to the multi-mode influence analysis data and the intraocular pressure influence analysis data to obtain multi-mode intraocular pressure dynamic monitoring data. Further, the step S1 includes: The glasses are provided with an eye sensor and a pose sensor, and the eye sensor and the pose sensor are combined to obtain a sensor group; Acquiring multi-mode sensing data according to the sensor group to obtain multi-mode sensing acquisition data; And generating a plurality of monitoring points according to the monitoring area of the glasses, and dividing the multi-mode sensing acquisition data according to the information of the monitoring points to obtain point multi-mode acquisition data. Further, the generating a plurality of monitoring points according to the monitoring area of the glasses, dividing the multi-modal sensing acquisition data according to the monitoring point information to obtain point multi-modal acquisition data, including: Acquiring information of a glasses monitoring area according to a sensor group, and carrying out area characteristic identification according to the information of the glasses monitoring area to acquire area characteristic identification information; Generating a plurality of monitoring points according to the regional characteristic identification information; Carrying out data feature extraction of each monitoring point location on the multi-mode sensing acquisition data to obtain point location feature extraction data; Performing similarity calculation on the point feature extraction data to obtain point feature similarity data; And carrying out representative point location analysis and determination on the point location feature extraction data according to the point location feature similarity data, and further obtaining point location multi-mode acquisition data of the representative point location. Further, the performing representative point location analysis to the point location feature extraction data according to the point location feature similarity data to obtain point location multi-mode acquisition data of the representative point location includes: Comparing the point location feature similarity data with a point location feature similarity threshold value to obtain point location feature similarity comparison data; Determining a plurality of similar monitoring points according to the point feature similarity comparison data, and combining the similar monitoring points to obtain a similar point combination; obtaining an average value of point location feature similarity data of similar point location combinations, and obtaining a similar point location combination average value; obtaining t