CN111128382A - Artificial intelligence multimode imaging analytical equipment - Google Patents
Artificial intelligence multimode imaging analytical equipment Download PDFInfo
- Publication number
- CN111128382A CN111128382A CN201911392016.0A CN201911392016A CN111128382A CN 111128382 A CN111128382 A CN 111128382A CN 201911392016 A CN201911392016 A CN 201911392016A CN 111128382 A CN111128382 A CN 111128382A
- Authority
- CN
- China
- Prior art keywords
- module
- human body
- image
- lens
- light source
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/30—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for calculating health indices; for individual health risk assessment
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B3/00—Apparatus for testing the eyes; Instruments for examining the eyes
- A61B3/10—Objective types, i.e. instruments for examining the eyes independent of the patients' perceptions or reactions
- A61B3/12—Objective types, i.e. instruments for examining the eyes independent of the patients' perceptions or reactions for looking at the eye fundus, e.g. ophthalmoscopes
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
- A61B5/7271—Specific aspects of physiological measurement analysis
- A61B5/7275—Determining trends in physiological measurement data; Predicting development of a medical condition based on physiological measurements, e.g. determining a risk factor
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H30/00—ICT specially adapted for the handling or processing of medical images
- G16H30/20—ICT specially adapted for the handling or processing of medical images for handling medical images, e.g. DICOM, HL7 or PACS
Landscapes
- Health & Medical Sciences (AREA)
- Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Public Health (AREA)
- Biomedical Technology (AREA)
- Medical Informatics (AREA)
- Biophysics (AREA)
- Theoretical Computer Science (AREA)
- Molecular Biology (AREA)
- Data Mining & Analysis (AREA)
- Artificial Intelligence (AREA)
- Heart & Thoracic Surgery (AREA)
- Animal Behavior & Ethology (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- General Engineering & Computer Science (AREA)
- Veterinary Medicine (AREA)
- Computing Systems (AREA)
- Pathology (AREA)
- Evolutionary Computation (AREA)
- Epidemiology (AREA)
- Primary Health Care (AREA)
- General Physics & Mathematics (AREA)
- Surgery (AREA)
- Computational Linguistics (AREA)
- Ophthalmology & Optometry (AREA)
- Radiology & Medical Imaging (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Databases & Information Systems (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physiology (AREA)
- Psychiatry (AREA)
- Signal Processing (AREA)
- Eye Examination Apparatus (AREA)
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
Abstract
The invention relates to an artificial intelligence multimode imaging analysis device, which comprises: the human body characteristic image and video acquisition system is used for realizing the acquisition of motion video signals of all parts of a human body and the acquisition of human body static images; the human body characteristic image database is used for storing historical image information, image characteristic information and historical clinical diagnosis information; and the artificial intelligent hardware control and data analysis processing system is used for automatically controlling the human body characteristic image and video acquisition processing system to finish video and image acquisition, and analyzing and comparing the acquired video and image with the information stored in the human body characteristic image database to finish disease early warning and/or health state assessment.
Description
Technical Field
The invention relates to the technical field of multimode imaging based on Artificial Intelligence (AI), in particular to an Artificial Intelligence multimode imaging analysis device.
Background
The inspection method belongs to the four basic diagnosis methods of inspection, sniffing, inquiry and cutting in traditional Chinese medicine, and diagnosis of visceral diseases and analysis and prediction of health conditions of human bodies and the like are carried out by observing gait, physical signs, facial phase, tongue phase, hand phase, eye image (sclera or white eye, iris, vitreous body, fundus oculi and the like) and image characteristics of other parts of human bodies, such as color, plaque, blood vessels, points, protrusions and other characteristic information. The ancient medical books of Huangdi's classic, jin Kui Yao L.and Mai Jue Hui Bian have been recorded in detail for inspection, and the theory of "five figures and eight outlines" has been gradually established. The theory of "inspection" is further improved since the recent times, Chengdeli divides the sclera of the eye into 14 areas, Wang jin Dynasty works on "inspection eye syndrome differentiation diagnostics" divides the white eye into 17 areas, which correspond to different organs of the zang-fu organs, and diagnoses the disease by observing different characteristics of different areas of the white eye, thus having stronger clinical application value.
In the current stage, the eye diagnosis of traditional Chinese medicine mainly depends on the traditional Chinese medicine (including Zhuang medicine, Tibetan medicine and the like) with rich experience to carry out visual observation, and the white eye characteristic change phenomenon diagnosis is carried out through the accumulation of long-term experience and is associated with related diseases. Currently, the sclera visual inspection is very limited in application, on one hand, because the device for assisting in observing the sclera is a slit lamp, about 30 minutes is needed for inspecting one patient, and the efficiency is low; on the other hand, misdiagnosis is likely to occur when the requirements on doctors and medical environments are high, experiences are few or examination conditions are poor, and the method is difficult to popularize in primary hospitals and families.
The prior art discloses equipment for imaging eyes, a human health condition on-body analysis system and method based on white-eye shadowless imaging and the like, which solve the problems of partial image acquisition and image processing, but the problem of low intelligent degree still generally exists, and each stage of image acquisition, processing, diagnosis and the like depends on the high cooperation of a subject and the participation of professional technicians, so that the popularization is difficult. In the prior art, the software and hardware design and the image preliminary processing of image acquisition are mainly researched, the black eye is not intelligently and automatically captured, the black eye is positioned and acquired, the white eye image is subjected to spherical surface splicing and distortion correction, and the problems that the eye image acquired by shooting a plurality of images by the same eye is repeated, the shooting distortion causes inaccurate eye image characteristic identification and the like exist; the complete extraction method of the white eye region of the true color eye image in the prior art realizes the whole extraction of the white eye, but does not realize the extraction, classification and marking of the eye image characteristics on the white eye, and has very limited effect on improving the diagnosis efficiency.
In conclusion, a need exists for a system which has small dependence on the cooperation degree of a subject and the assistance degree of professionals, is efficient and accurate, is convenient to popularize in primary hospitals and families, and can realize important part images, video acquisition processing, feature identification marks and the like of a human body in a one-stop manner.
Disclosure of Invention
In view of the above problems, the present invention provides an artificial intelligence multimode imaging analyzer, so as to realize big data analysis of human body characteristic images and health conditions, and meet the information development requirements of telemedicine and intelligent medical treatment.
In order to achieve the purpose, the invention adopts the technical scheme that: an artificial intelligence multi-mode imaging analysis apparatus, the apparatus comprising:
the human body characteristic image and video acquisition system is used for realizing the acquisition of motion video signals of all parts of a human body and the acquisition of human body static images;
the human body characteristic image database is used for storing historical image information, image characteristic information and historical clinical diagnosis information;
and the artificial intelligent hardware control and data analysis processing system is used for automatically controlling the human body characteristic image and video acquisition processing system to finish video and image acquisition, and analyzing and comparing the acquired video and image with the information stored in the human body characteristic image database to finish disease early warning and/or health state assessment.
Preferably, the device comprises a microprocessor, the microprocessor is connected with MedNet cloud medical big data in a wired or wireless mode, and the artificial intelligence hardware control and data analysis and processing system and the human body characteristic image database are directly installed or installed in the microprocessor in a mirror mode for use.
Preferably, the human body characteristic image and video acquisition system comprises an optical imaging module, a normal incidence illumination light source module, an oblique incidence illumination light source module and an indication guide module;
the optical imaging module is used for carrying out image acquisition or video recording on the human body interested part;
the normal incidence illumination light source module is used for focusing emitted light into a part of human body which is interested by the human body;
the oblique incidence illumination light source module is used for dark field illumination on the human body feeling happy part;
the indication guiding module is used for generating a cursor indication to guide the subject to adjust the position and the angle of the feeling part.
Preferably, the optical imaging module includes more than one lens or lens, imaging detector and a supporting universal adjusting bracket or cloud platform that can move back and forth, lens or lens are used for the human interesting part of dynamic focus, imaging detector is used for carrying out image acquisition or video recording to human interesting part, supporting universal adjusting bracket or cloud platform is used for changing lens or lens, imaging detector's direction and angle, realize the tracking to human removal.
Preferably, the normal incidence illumination light source module includes more than one lens or lens capable of moving back and forth, a beam splitter, a polarizer, a coupling collimator and a light source, light emitted by the light source is transmitted to the beam splitter through the coupling collimator and the polarizer, and light reflected by the beam splitter is focused into a region of interest of a human body through the lens or lens.
Preferably, the oblique incidence illumination light source module includes first to fourth light sources capable of moving and illuminating in four different directions, namely up, down, left and right, and all or part of the first to fourth light sources is used for dark field illumination of the human body feeling-like part.
Preferably, the indication guiding module includes one or more than one lens or lens capable of moving back and forth, a beam splitter, two polarizing plates, a lens and a liquid crystal display, the liquid crystal display is used for generating a dynamic indication cursor, the cursor is sequentially transmitted to the beam splitter through the lens and the first polarizing plate, part of light reflected by the beam splitter is focused on a human body interest part through the lens or the lens, the light reflected by the cursor through the lens or the lens is transmitted to the second polarizing plate through the beam splitter and then is cut off, and the first polarizing plate and the second polarizing plate form an orthogonal polarization state.
Preferably, the artificial intelligence hardware control and data analysis processing system comprises a mode selection and illumination control module, an image and video acquisition control module, a feedback guide module, a target tracking module, a storage module and a data analysis module;
the mode selection and illumination control module is used for controlling the combination modes of the optical imaging module, the normal incidence illumination light source module, the oblique incidence illumination light source module and the indication guide module, wherein the optical imaging module is controlled to be used alone or the optical imaging module is controlled to be used in combination with any one or more of the normal incidence illumination light source module, the oblique incidence illumination light source module and the indication guide module;
the image and video acquisition control module is used for controlling the optical imaging module and dynamically acquiring human motion video signals and human static images;
the target tracking module is used for dynamically tracking the interested part of the human body, controlling the optical imaging module to automatically change an imaging angle along with the motion of the human body by dynamically acquiring the image of the interested part of the human body, and automatically focusing the optical imaging module, so that the interested part of the human body is always in the center of a view field and basically unchanged in size, and the interested part of the human body is tracked;
the feedback guiding module is used for guiding the subject to adjust the position and the angle through language prompt and/or cursor indication of the indication guiding module;
the storage module is used for storing the acquired information of the interested part;
and the data analysis module is used for comparing the acquired information with big data information stored in the human body characteristic image database by a machine learning method, such as similarity, image frequency spectrum, entropy, physical signs and the like, and performing disease early warning and/or health state assessment.
Preferably, the artificial intelligence hardware control and data analysis processing system further comprises:
the characteristic extraction module is used for identifying the acquired information and extracting the characteristics of the image;
the image splicing and distortion correction is used for splicing and distortion correction of the extracted image features by adopting a pre-established neural network model;
the characteristic marking module is used for identifying the characteristics and parameters of the characteristic images by splicing the corrected characteristic images and adopting a pre-trained deep convolutional neural network model, and marking the identified characteristics and parameters;
the data analysis module predicts diseases according to the characteristics and parameters of the interested part by adopting a pre-trained network model, and compares the predicted diseases with big data information stored in a human body characteristic image database, such as one or more of gait, blood vessels, color, texture, spots, points, strips, blood vessels, dunes, pterygium, metal rings, moons, nets, temperature, physiological and pathological parameters and the like, so as to early warn the diseases of a subject and/or evaluate the health of the subject.
Preferably, the specific process of the picture stitching and distortion correcting step is as follows:
firstly, detecting characteristic points by adopting an angular point matching method;
then, associating the feature points, and deleting unneeded corner points by adopting the RANSAC principle;
finally, a reference image is appointed, pixels of the input image are mapped to a plane defined by the reference image, and smooth transition among the spliced images is realized through an image fusion method;
the corner matching method adopts a Harris corner detection algorithm;
the characteristic point association adopts a mutual information method;
the image fusion adopts a feathering method, a pyramid method or a gradient method to realize smooth transition among spliced images.
Due to the adoption of the technical scheme, the invention has the following advantages:
1. the invention can quickly and accurately acquire video signals of gait, physical signs and the like of human motion and a plurality of characteristic image information of static eye image, facial image, tongue image, hand image, other parts of the body and the like of the human body, and is beneficial to the comprehensive analysis and evaluation of the human health information;
2. the invention can realize the intelligent tracking, automatic focusing and automatic amplification factor adjustment of human body movement, and carry out video signal acquisition and image signal acquisition, thereby greatly reducing the dependence on the cooperation degree of a subject and the professional skill level and proficiency of an operator;
3. the invention adopts oblique incidence illumination and AI feedback guidance to reduce the interference influence of reflected light of local imaging detection of human body; the subject is guided to change the observation direction and angle through language prompt, cursor indication and the like, the curvature radius of each part of the eyeball and the position of the pupil of the eye are automatically adjusted, so that the reflection images of each part of the eyeball to the oblique incidence illumination light source are gathered to one point and are overlapped with the pupil of the eye, and the imaging of the sclera, the iris and the like of the eye without reflection light interference is realized;
4. the invention adopts normal incidence illumination and AI feedback guidance, focuses light emitted by a normal incidence illumination light source module to enter the pupil of an eyeball, illuminates the vitreous body and retina of the eyeball, guides a subject to automatically adjust the curvature radius of each part of the eyeball and the position of the eye through language prompt, cursor indication and the like, and clearly guides an indication cursor through forward observation so that an optical imaging module can automatically realize clear imaging on the eyeground, the retina, the vitreous body and the like;
5. the system can manually input information and automatically store extracted characteristic information, expert medical advice information, health analysis data and the like, establish a large database for analyzing human characteristic images and health conditions, adopt an artificial intelligence algorithm combining a multilayer neural network and machine learning, realize multi-mode human characteristic information storage and human health condition analysis, and provide disease risk early warning measures and health management suggestions for a subject, wherein the multi-mode human characteristic information storage and the human health condition analysis comprise healthy person disease risk prediction, sub-health population screening and early warning, and quick non-invasive detection and long-term tracking health condition evaluation of disease population diseases (such as diabetes, thoracic cerebrovascular diseases, lung cancer, polycystic ovary syndrome, AIDS and the like);
6. the invention shares cloud resources, has a remote multimedia interaction function, can realize AI big data analysis of human body characteristic images and health conditions, and meets the information development requirements of remote medical treatment and intelligent medical treatment.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the provided drawings without creative efforts.
FIG. 1 is a schematic structural diagram of an artificial intelligence multimode imaging analysis device according to an embodiment of the invention;
FIG. 2 is a schematic diagram of a process for acquiring information such as gait, physical signs and the like in an embodiment of the invention;
FIG. 3 is a schematic diagram of a process for image capture of eye images, facial images, tongue images, hand images and other parts of the body in an embodiment of the present invention;
FIG. 4 is a schematic diagram of a process for implementing reflected light interference-free imaging under guidance of AI feedback language prompt and markup in an embodiment of the present invention;
FIG. 5 is a schematic diagram of a process for image acquisition of the fundus, retina and vitreous humor, according to an embodiment of the present invention;
FIG. 6 is a workflow diagram for AI multimodal imaging health analysis according to an embodiment of the invention;
FIG. 7 is a flowchart of the operation of the present invention for eye image acquisition location and distortion correction;
FIG. 8 is a flowchart of the operation of an embodiment of the present invention for white eye image feature recognition tagging;
FIG. 9 is a workflow diagram for machine learning health analysis according to an embodiment of the present invention.
Detailed Description
Exemplary embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. While exemplary embodiments of the invention are shown in the drawings, it should be understood that the invention can be embodied in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the invention to those skilled in the art.
It is to be understood that the terminology used herein is for the purpose of describing particular example embodiments only, and is not intended to be limiting. As used herein, the singular forms "a", "an" and "the" may be intended to include the plural forms as well, unless the context clearly indicates otherwise. The terms "comprises," "comprising," "including," and "having" are inclusive and therefore specify the presence of stated features, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, steps, operations, elements, components, and/or groups thereof. The method steps, processes, and operations described herein are not to be construed as necessarily requiring their performance in the particular order described or illustrated, unless specifically identified as an order of performance. It should also be understood that additional or alternative steps may be used.
As shown in fig. 1, the artificial intelligence multimode imaging analysis device provided in this embodiment has functions of identifying and positioning an interested target, tracking the target, feeding back a language prompt and mark guide, automatically focusing, collecting an image and a video signal, splicing and distortion correction of an image, identifying features, extracting and marking the features, analyzing data, and the like, and can realize the collection of a human motion video signal and the collection of images of human static eyes, facial phases, tongue phases, hand phases and other parts of a body, as well as image processing, feature identification and marking, and the like, and the device includes:
a human body characteristic image and video acquisition system CJXT, a human body characteristic image database BIGDATA and an artificial intelligence hardware control and data analysis processing system AI. Wherein:
the human body characteristic image and video acquisition system CJXT is used for realizing the acquisition of motion video signals of all parts of a human body and the acquisition of human body static images, wherein the acquisition of the human body static images comprises the acquisition of eye images, facial images, tongue images, hand images and other parts of the human body.
The human body characteristic image database BIGDATA stores a large amount of historical image information, image characteristic information and historical clinical diagnosis information;
an artificial intelligence hardware control and data analysis processing system AI for automatically controlling the human characteristic image and video acquisition processing system by adopting an artificial intelligence algorithm combining a multilayer neural network and machine learning, realizing the functions of target tracking, automatic focusing, dynamic amplification, light source control, video acquisition, image acquisition and the like, analyzing and comparing the information stored in a human characteristic image database BIGDATA, carrying out early warning on the diseases of a subject and/or evaluating the health state of the subject, wherein the acquired image and video signal can be subjected to characteristic recognition, characteristic extraction, characteristic marking and the like, and the extracted characteristic information is analyzed and compared with the historical information stored in the human characteristic image database BIGDATA; the collected images and video signals can be subjected to disease early warning and/or health state assessment directly by comparing similarity, frequency spectrum difference, entropy difference and the like between the test images and the database storage images without extracting features.
In some embodiments of the invention, the device comprises a microprocessor Processor, the microprocessor Processor has the functions of digital signal storage, analysis and processing, display language prompt MK, remote interaction and the like, can be connected with MedNet cloud medical big data in a wired or wireless mode, and the artificial intelligent hardware control and data analysis processing system AI and the human body characteristic image database BIGDATA can be directly installed or installed in the microprocessor Processor in a mirror image mode for use.
In some embodiments of the present invention, the human body feature image and video capturing system CJXT includes an optical imaging module, a normal incidence illumination light source module, an oblique incidence illumination light source module, and an indication guide module.
Specifically, the optical imaging module is used for image acquisition or video recording of a human body interested part, and comprises more than one lens or lens L3, an imaging detector CCD and a matched universal adjusting bracket or holder M, wherein the lens or lens L3 is used for focusing the human body interested part, the imaging detector CCD is used for image acquisition or video recording of the interested part, and the optical imaging module adjusts the angle or position of the mounted lens or lens L3 and the imaging detector CCD through the matched universal adjusting bracket or holder. As shown in fig. 2, the optical imaging module can track, automatically focus, and automatically adjust the magnification factor of the Object motion of the human body by artificial intelligence hardware control and data analysis processing system AI control, and is used for video signal acquisition and image signal acquisition to obtain information such as human gait, body shape, physical signs, and the like.
The normal incidence illumination light source module is used for focusing emitted light into a certain local part of a human body, such as the pupil of the eyeball of the human body, and illuminating the vitreous body and the retina of the eyeball. The normal incidence illumination light source module comprises more than one lens or lens L1 capable of moving back and forth, a beam splitter SP1, a polaroid P1, a coupling collimator lens L5 and a light source S0, wherein light emitted by the light source S0 is emitted to the beam splitter SP1 through the coupling collimator lens L5 and the polaroid P1, and light reflected by the beam splitter SP1 is focused into the pupil of the eyeball through the lens or lens L1 to illuminate the vitreous body and the retina of the eyeball. The polarizer P1 is used to make the light emitted from the light source S0 have linear polarization.
The oblique incidence illumination light source module is used for dark field illumination of a human body feeling part and reducing imaging influence of illumination light reflection on the optical imaging module, and comprises four movable light sources S1, S2, S3 and S4 illuminating from four different directions, namely, from top to bottom, from left to right, as shown in fig. 3, and under the control of an artificial intelligence hardware control and data analysis processing system, dark field illumination of all or part of the light sources S1-S4 on the human body feeling part (such as sclera or iris) can be realized.
The indication guiding module is used for generating cursor indication through a display screen and guiding a subject to adjust the position, the angle and the like of a measured sensitive part, and comprises one or more lenses or lens L2 capable of moving back and forth, a beam splitter SP2, two polaroids P2 and P3, a lens L4 and a liquid crystal display YJP. The liquid crystal display YJP is used for generating a dynamic indication cursor and guiding a subject to adjust the position, the angle and the like of a measured sensitive part, the cursor is sequentially emitted to the beam splitter SP2 through the lens L4 and the polaroid P2, part of light reflected by the beam splitter SP2 is imaged into a virtual image at a position of a 250mm photopic distance of human eyes through the lens or the lens L2, and then the virtual image is clearly imaged on the retina through eyeballs.
As shown in fig. 3, under the control of the artificial intelligence hardware control and data analysis processing system AI, the light emitted by the oblique incidence illumination light source module illuminates a region of interest of a human body (e.g. sclera or iris of a human eye) in a dark field illumination manner, the light reflected or scattered from the region of interest of the human body (e.g. sclera or iris of a human eye) is received by the mirror or lens L2, and is transmitted by the beam splitter SP2 and then is received by the optical imaging module through the polarizing plate P3, wherein the polarizing plate P3 and the polarizing plate P2 form orthogonal polarization states, so as to ensure that the reflected light from the light source S0 through the mirror or lens L1 is completely blocked and is not received by the imaging detector CCD, and it should be noted that the optical imaging module jointly uses the oblique incidence illumination light source module and/or the normal incidence illumination light source module, and the mirror or lens L1, the mirror or lens L2 and the mirror.
Preferably, the artificial intelligence hardware control and data analysis processing system AI can control the oblique incidence illumination light source module and the indication guiding module according to the image feedback detected by the imaging detector CCD to perform oblique incidence dark field illumination from any one direction or several directions, the light source S (which can be any one of S1, S2, S3 and S4) has a variable angle Ψ with the optical axis, and the subject is guided to change the observation direction and the angle of the pupil with the optical axis by the language prompt MK of the microprocessor Processor and/or the indication of the cursor (which can be an arrow or a cross or other mark) generated on the liquid crystal display YJP of the indication guiding module
As shown in fig. 4(a), the curvature radius of each part of the eyeball and the position of the pupil of the eye are automatically adjusted, so that the reflected images of each part of the eyeball to the obliquely incident illumination light source are gathered to one point and are overlapped with the pupil of the eye, and imaging without interference of reflected light to the sclera or the iris of the eye is realized, as shown in fig. 4 (b).
Preferably, the optical imaging module can be used alone, as shown in fig. 2; and can be freely combined with any one or more of the normal incidence illumination light source module, the oblique incidence illumination light source module and the indication guide module, as shown in fig. 3 and 5.
As shown in fig. 3, is an example of the use of an optical imaging module in conjunction with an oblique incidence illumination source module and an index guide module. The artificial intelligent hardware control and data analysis processing system can realize dark field illumination, automatic positioning, automatic focusing and automatic amplification factor adjustment on the local position of a human body by jointly controlling the optical imaging module, the oblique incidence illumination light source module and the indication guide module, acquire video signals and image signals and obtain image information of human eye images (images of sclera, iris, vitreous body and the like), facial images, tongue images, hand images, other parts of the body and the like.
As shown in fig. 5, is an example of the use of an optical imaging module in conjunction with a normal incidence illumination source module and an indication guide module. In fig. 5, the optical imaging module is used in combination with the normal incidence illumination light source module and the indication guide module, and the subject is guided to automatically adjust the curvature radius of each part of the eyeball and the position of the eye through the language prompt of the microprocessor Processor and/or the cursor indication generated on the liquid crystal display YJP of the indication guide module, and the cursor of the indication guide module is clearly observed in the forward direction; then, the artificial intelligence hardware control and data analysis processing system AI controls the normal incidence illumination light source module to emit light, and the light enters the pupil of the eyeball through the lens or lens L1 to illuminate the vitreous body and the retina of the eyeball, so that the optical imaging module can realize clear imaging of the eyeground, the retina, the vitreous body and the like.
In some embodiments of the present invention, the artificial intelligence hardware control and data analysis processing system AI is used to implement the functions of target identification and positioning of interest, target tracking, feedback language prompt and mark guidance, auto focusing, image and video signal acquisition, image stitching and distortion correction, feature identification, feature extraction, feature marking, data analysis, etc., as shown in fig. 6, the system includes a mode selection and illumination control module, an image and video acquisition control module, a target tracking module, a feedback guidance and auto focusing module, a feature extraction marking module, an image stitching and distortion correction and storage module, and a data analysis module;
the mode selection and illumination control module is used for controlling the combined mode of the incident illumination light source module, the oblique incident illumination light source module and the indication guide module;
the image and video acquisition control module is used for controlling the optical imaging module and dynamically acquiring human motion video signals, human static eye images, human face images, human tongue images, human hand images and other body part images;
the target tracking module is used for dynamically tracking the interested part of the human body, automatically and dynamically acquiring the image of the interested part of the human body, controlling a universal adjusting bracket or a holder matched with the optical imaging module, automatically changing an imaging angle along with the motion of the human body, automatically focusing and automatically adjusting the magnification factor of a lens or a lens L3 of the optical imaging module, ensuring that the interested part of the human body is always in the center of a view field and the size of the interested part is basically unchanged, and realizing the tracking measurement of the information of gait, body form, physical signs and the like of the motion of the human body.
The feedback guide module is used for guiding the testee to automatically adjust the curvature radius of each part of the eyeball and the position of the eyeball in a normal incidence illumination mode through language prompt of the microprocessor and cursor indication generated on a liquid crystal display YJP of the indication guide module, and the cursor of the indication guide module is clearly observed in the forward direction, so that the optical imaging module can realize clear imaging on the eyeground, the retina, the vitreous body and the like; in the oblique incidence illumination mode, a subject is guided to change the observation direction and the included angle between the pupil and the optical axis, the curvature radius of each part of the eyeball and the position of the pupil of the eye are automatically adjusted, so that the reflected images of each part of the eyeball to the oblique incidence illumination light source are gathered to one point and are overlapped with the pupil of the eye, and the imaging of the sclera or the iris of the eye and the like without reflected light interference is realized.
The characteristic extraction module is used for identifying and extracting the characteristics of the collected images, wherein the characteristic identification information comprises one or more of gait, blood vessel, color, texture, spot, point, strip, blood vessel, dune, pterygium, metal ring, mooncake and net, temperature, similarity, image spectrum, entropy, physiological and pathological parameters and the like, and the comprehensive evaluation on the health condition of the human body is facilitated;
picture stitching and distortion correction, which is used for stitching and distortion correction of extracted features by adopting a pre-established neural network model, as shown in fig. 7, the neural network model is the prior art, and the specific steps can include feature extraction, image registration, calculation of a homography matrix H, deformation and fusion, after an eye feature image is input: detecting characteristic points by adopting an angular point matching method, then correlating the characteristic points, calculating a single mapping matrix by using an RANSAC principle to estimate and delete unnecessary angular points, finally calculating a specified reference image, mapping pixels of an input image onto a plane defined by the reference image, and realizing smooth transition between spliced images by using an image fusion method.
The feature labeling module is configured to identify specific features and parameters of the feature images by using a pre-trained deep convolutional neural network model for the feature images after the splicing correction, and label the identified features and parameters, as shown in fig. 8, the neural network model may use the prior art, and the specific steps include framing a certain part of the whole-eye images obtained by the previous splicing by using sliding windows of different sizes as an eye image feature candidate region, extracting visual information related to the eye image feature candidate region, and identifying different eye image features by using a classifier, which are not specifically described in detail.
The storage module is used for marking and storing the characteristics and the parameters of the interested part;
the data analysis module can be used for carrying out disease early warning and/or health state assessment by comparing the acquired image or video information with big data information stored in the human body characteristic image database storage image, namely directly carrying out disease early warning or health state assessment by comparing similarity, frequency spectrum difference, entropy difference and the like between the test image and the database storage image; the characteristics and parameters of the interested part can also be called to predict diseases by adopting a pre-trained network model, and the characteristics and parameters are compared with the contents stored in a human body characteristic image database to early warn the diseases of the subject and/or evaluate the health of the subject, wherein the network model can be optimized and trained based on the association between the eye image and the diseases by adopting a BP neural network, and details are not repeated.
In some embodiments of the present invention, the corner matching method may use the existing Harris corner detection algorithm, which has many detection features, rotation invariance and scale variability. Calculating an autocorrelation matrix of each point (x, y) in the collected image, and then performing Gaussian filtering on each pixel point to obtain a new matrix M, wherein the discrete two-dimensional zero-mean Gaussian function is as follows:
where the displacement (u, v) is the displacement of the window each time a corner detection is performed.
Calculate the corner measure for each point (x, y):
R＝Det(M)-k*trace(M) (2)
and selecting a local maximum value to be compared with a set threshold value, and if the local maximum value is higher than the threshold value, determining that the local maximum value is a corner point.
In some embodiments of the present invention, the characteristic point association method uses a Mutual Information (MI) method.
In some embodiments of the present invention, the image fusion uses feathering (feathering), pyramid (pyramid), gradient (gradient), and the like to achieve smooth transition between the stitched images.
In some embodiments of the present invention, the BP neural network adopted in this embodiment specifically includes:
the neural network health analysis process is characterized by comprising 20 input nodes, two output nodes and two hidden layers, wherein each hidden layer comprises 10 nodes, and the activation value of the i node of the j layer is as follows:
in equation (2), n is a node on the j-1 level, and is a weight; a is the activation value and j is the bias.
The activation values for the nodes of level j are:
in the above equation, k is a node on the j level; n is a node on level j-1; ω is a weight; a is an activation value; j is the offset.
In some embodiments of the present invention, the human body feature image database BIGDATA is used for storing and managing the collected human body image, video signal, feature extraction information, expert medical advice information, and health analysis data; the feature recognition queue, the intermediate data, the feature recognition marking result and the like trained through neural network and machine learning are stored, and as shown in fig. 9, the function of manually inputting information and automatically classifying, storing and managing the extracted feature information, expert medical advice information, health analysis data and the like is realized.
In some embodiments of the invention, the artificial intelligence hardware control and data analysis processing system AI and the human body characteristic image database BIGDATA can share cloud resources, have a remote multimedia interaction function, share cloud data information resources, have a remote medical treatment interaction function, can realize AI big data analysis of human body characteristic images and health conditions, and meet the information development requirements of remote medical treatment and intelligent medical treatment. When the system is used specifically, a part of samples can be taken out from the obtained human body characteristic images and used for neural network parameter adjustment and machine learning of an artificial intelligence hardware control and data analysis processing system AI, a training characteristic identification queue, intermediate data, a characteristic identification marking result and the like are formed, all information is stored in a warehouse, a human body characteristic image database BIGDATA is constructed by combining the collected human body images, video signals, extracted characteristic information, expert medical advice information, health analysis data and the like so as to carry out health analysis on the rest samples of the testee, and finally, the health condition analysis result of the testee is given out and comprises health early warning and conditioning advice and the like provided for the testee.
The following describes in detail the process of performing imaging detection by using the artificial intelligence multimode imaging device provided by this embodiment, so as to acquire human motion video signals and human static eye images, facial images, tongue images, hand images and other body part images of a subject, and perform image processing, feature recognition marking, disease diagnosis and health analysis, including the following contents:
the human body movement video signal acquisition and the human body whole body or body feeling position image acquisition are carried out on the subject, the artificial intelligent hardware control and data analysis processing system AI controls the universal adjusting support or the holder matched with the optical imaging module, the lens L3 of the optical imaging module is automatically focused and the magnification factor is automatically adjusted, the human body movement tracking is realized, the video signal acquisition and the image signal acquisition are carried out, the information of human body gait, body shape, physical signs and the like is obtained, all the information is stored in the storage module after being processed, and the data analysis module analyzes and compares the information with the information in the human body characteristic image database BIGDATA to obtain the health condition of the subject.
When the human eye image (the image of sclera, iris, vitreous body, etc.), the facial phase, tongue phase, hand phase and other parts of the body, etc. of the tested person are required to be acquired, the artificial intelligence hardware control and data analysis processing system AI jointly controls the optical imaging module, the oblique incidence illumination light source module and the indication guiding module, firstly, all or part of the light sources S1-S4 of the oblique incidence illumination light source module are started to carry out dark field illumination on the human body feeling happy part, then, the lens L3 of the optical imaging module is automatically focused and adjusted in magnification, finally, the tested person is guided to adjust the position, angle, etc. of the measured feeling part through the language prompt MK of the microprocessor Processor and/or the cursor indication generated on the liquid crystal display YJP of the indication guiding module, so that the reflection image of the oblique incidence illumination light source completely disappears or gathers to a point and coincides with the pupil of the eye, clear images or recorded videos of the measured sensitive part can be obtained. And subsequently, storing all information into a storage module through digital image processing, distortion correction, AI characteristic identification, characteristic extraction and characteristic marking, and analyzing and comparing the information with the human characteristic image and the information in the big health condition analysis database BIGDATA by a data analysis system to obtain a health condition analysis result of the testee.
When the image information of the eyeground, the retina and the like of a human body needs to be acquired, an artificial intelligence hardware control and data analysis processing system AI is used for jointly controlling the optical imaging module, the normal incidence illumination light source module and the indication guide module, firstly, a language prompt MK of a microprocessor Processor and/or a cursor indication generated on a liquid crystal display YJP of the indication guide module are used for guiding the human body to automatically adjust the curvature radius of each part of an eyeball and the position of the eye, the cursor generated on the liquid crystal display YJP of the indication guide module is clearly indicated in the positive direction observation, then, a light source S0 of the normal incidence illumination light source module is started to focus light to enter the pupil of the eyeball, the vitreous body and the retina of the eyeball are illuminated, automatic focusing and magnification adjustment are synchronously carried out by a lens L3 of the optical imaging module, and finally the clear imaging or video recording and the like of the eyeground, the retina and the vitreous body are realized by the optical imaging module. And subsequently, storing all information into a storage module through digital image processing, distortion correction, AI characteristic identification, characteristic extraction and characteristic marking, and analyzing and comparing the information with the human characteristic image and the information in the big health condition analysis database BIGDATA by a data analysis system to obtain a health condition analysis result of the testee.
The above embodiments are only used for illustrating the present invention, and the structure, connection mode, manufacturing process, etc. of the components may be changed, and all equivalent changes and modifications performed on the basis of the technical solution of the present invention should not be excluded from the protection scope of the present invention.
Claims (10)
1. An artificial intelligence multimode imaging analysis device, characterized in that the device comprises:
the human body characteristic image and video acquisition system is used for realizing the acquisition of motion video signals of all parts of a human body and the acquisition of human body static images;
the human body characteristic image database is used for storing historical image information, image characteristic information and historical clinical diagnosis information;
and the artificial intelligent hardware control and data analysis processing system is used for automatically controlling the human body characteristic image and video acquisition processing system to finish video and image acquisition, and analyzing and comparing the acquired video and image with the information stored in the human body characteristic image database to finish disease early warning and/or health state assessment.
2. The artificial intelligence multimode imaging analysis device of claim 1, characterized in that, the device includes a microprocessor, the microprocessor is connected with MedNet cloud medical big data by wire or wirelessly, the artificial intelligence hardware control and data analysis processing system and the human body characteristic image database are directly installed or mirror-installed in the microprocessor for use.
3. The artificial intelligence multimode imaging analysis device according to claim 1 or 2, wherein the human body characteristic image and video acquisition system comprises an optical imaging module, a normal incidence illumination light source module, an oblique incidence illumination light source module and an indication guiding module;
the optical imaging module is used for carrying out image acquisition or video recording on the human body interested part;
the normal incidence illumination light source module is used for focusing emitted light into a part of human body which is interested by the human body;
the oblique incidence illumination light source module is used for dark field illumination on the human body feeling happy part;
the indication guiding module is used for generating a cursor indication to guide the subject to adjust the position and the angle of the feeling part.
4. The artificial intelligence multimode imaging analysis device of claim 3, wherein the optical imaging module comprises more than one lens or lens capable of moving back and forth, an imaging detector and a matching universal adjusting bracket or holder, the lens or lens is used for dynamically focusing the interested part of the human body, the imaging detector is used for carrying out image acquisition or video recording on the interested part of the human body, and the matching universal adjusting bracket or holder is used for changing the direction and angle of the lens or lens and the imaging detector to realize the tracking of the movement of the human body.
5. The artificial intelligence multimode imaging analysis device of claim 3, wherein the normal incidence illumination light source module comprises one or more than one lens or lens capable of moving back and forth, a beam splitter, a polarizer, a coupling collimator and a light source, wherein light emitted by the light source is transmitted to the beam splitter through the coupling collimator and the polarizer, and the light reflected by the beam splitter is focused into the region of interest of the human body through the lens or lens.
6. The artificial intelligence multimode imaging analysis device of claim 3, wherein the oblique incidence illumination light source module comprises a first light source to a fourth light source which can move and illuminate in four different directions, namely, up, down, left and right, and all or part of the first light source to the fourth light source is used for dark field illumination of the human body feeling part.
7. The artificial intelligence multimode imaging analysis device according to claim 3, wherein the indication guiding module comprises one or more than one lens or lens capable of moving back and forth, a beam splitter, two polarizers, a lens and a liquid crystal display, the liquid crystal display is configured to generate a dynamic indication cursor, the cursor is transmitted to the beam splitter through the lens and the first polarizer in sequence, a part of light reflected by the beam splitter is focused on the region of interest of the person through the lens or lens, the light reflected by the cursor through the lens or lens is transmitted to the second polarizer through the beam splitter and then is cut off, and the first polarizer and the second polarizer form an orthogonal polarization state.
8. The artificial intelligence multimode imaging analysis device of claim 3, wherein the artificial intelligence hardware control and data analysis processing system comprises a mode selection and illumination control module, an image and video acquisition control module, a feedback guidance module, a target tracking module, a storage module and a data analysis module;
the mode selection and illumination control module is used for controlling the combination modes of the optical imaging module, the normal incidence illumination light source module, the oblique incidence illumination light source module and the indication guide module, wherein the optical imaging module is controlled to be used alone or the optical imaging module is controlled to be used in combination with any one or more of the normal incidence illumination light source module, the oblique incidence illumination light source module and the indication guide module;
the image and video acquisition control module is used for controlling the optical imaging module and dynamically acquiring human motion video signals and human static images;
the target tracking module is used for dynamically tracking the interested part of the human body, controlling the optical imaging module to automatically change an imaging angle along with the motion of the human body by dynamically acquiring the image of the interested part of the human body, and automatically focusing the optical imaging module, so that the interested part of the human body is always in the center of a view field and basically unchanged in size, and the interested part of the human body is tracked;
the feedback guiding module is used for guiding the subject to adjust the position and the angle through language prompt and/or cursor indication of the indication guiding module;
the storage module is used for storing the acquired information of the interested part;
and the data analysis module is used for comparing the acquired information with big data information stored in the human body characteristic image database by a machine learning method to carry out disease early warning and/or health state assessment.
9. The artificial intelligence multimodal imaging analysis apparatus according to claim 8 wherein the artificial intelligence hardware control and data analysis processing system further comprises:
the characteristic extraction module is used for identifying the acquired information and extracting the characteristics of the image;
the image splicing and distortion correction is used for splicing and distortion correction of the extracted image features by adopting a pre-established neural network model;
the characteristic marking module is used for identifying the characteristics and parameters of the characteristic images by splicing the corrected characteristic images and adopting a pre-trained deep convolutional neural network model, and marking the identified characteristics and parameters;
the data analysis module predicts diseases by adopting a pre-trained network model according to the characteristics and parameters of the interested part, compares the predicted diseases with big data information stored in a human body characteristic image database, and performs early warning on the diseases of the subject and/or evaluates health.
10. The artificial intelligence multimode imaging analysis device of claim 9, wherein the specific process of the picture stitching and distortion correction step is as follows:
firstly, detecting characteristic points by adopting an angular point matching method;
then, associating the feature points, and deleting unneeded corner points by adopting the RANSAC principle;
finally, a reference image is appointed, pixels of the input image are mapped to a plane defined by the reference image, and smooth transition among the spliced images is realized through an image fusion method;
the corner matching method adopts a Harris corner detection algorithm;
the characteristic point association adopts a mutual information method;
the image fusion adopts a feathering method, a pyramid method or a gradient method to realize smooth transition among spliced images.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911392016.0A CN111128382B (en) | 2019-12-30 | 2019-12-30 | Artificial intelligence multimode imaging analysis device |
| PCT/CN2020/123568 WO2021135557A1 (en) | 2019-12-30 | 2020-10-26 | Artificial intelligence multi-mode imaging analysis apparatus |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911392016.0A CN111128382B (en) | 2019-12-30 | 2019-12-30 | Artificial intelligence multimode imaging analysis device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111128382A true CN111128382A (en) | 2020-05-08 |
| CN111128382B CN111128382B (en) | 2022-06-28 |
Family
ID=70504781
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911392016.0A Active CN111128382B (en) | 2019-12-30 | 2019-12-30 | Artificial intelligence multimode imaging analysis device |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN111128382B (en) |
| WO (1) | WO2021135557A1 (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2021135557A1 (en) * | 2019-12-30 | 2021-07-08 | 清华大学 | Artificial intelligence multi-mode imaging analysis apparatus |
| CN113180971A (en) * | 2021-04-29 | 2021-07-30 | 新乡市中心医院 | Traditional Chinese medicine nursing device and control method |
| CN113390885A (en) * | 2021-08-17 | 2021-09-14 | 济南邦德激光股份有限公司 | Laser head cutting protective glass state detection device and detection method |
| CN114664410A (en) * | 2022-03-11 | 2022-06-24 | 北京医准智能科技有限公司 | Video-based focus classification method and device, electronic equipment and medium |
| CN116636808A (en) * | 2023-06-28 | 2023-08-25 | 交通运输部公路科学研究所 | Intelligent cockpit driver visual health analysis method and device |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113379397B (en) * | 2021-07-16 | 2023-09-22 | 北京华博创科科技股份有限公司 | Cloud workflow frame intelligent management and scheduling system based on machine learning |
| CN113724874A (en) * | 2021-09-08 | 2021-11-30 | 北京新清泰克体育科技有限公司 | Evaluation-based test system |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20050119642A1 (en) * | 2001-12-21 | 2005-06-02 | Horia Grecu | Method and apparatus for eye registration |
| CN105310646A (en) * | 2015-12-09 | 2016-02-10 | 博奥颐和健康科学技术（北京）有限公司 | Human health condition in-vivo analysis system and method based on white eye shadowless imaging |
| CN105426695A (en) * | 2015-12-18 | 2016-03-23 | 北京铭光正讯科技有限公司 | Health status detecting system and method based on irises |
| CN206497027U (en) * | 2017-02-28 | 2017-09-15 | 福建师范大学 | Monochromatic light spectrometer polarization domain optical coherence tomography system based on photoswitch |
| CN107307848A (en) * | 2017-05-27 | 2017-11-03 | 天津海仁医疗技术有限公司 | A kind of recognition of face and skin detection system based on the micro- contrast imaging of high speed large area scanning optics |
| CN107680683A (en) * | 2017-10-09 | 2018-02-09 | 上海睦清视觉科技有限公司 | A kind of AI eye healths appraisal procedure |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110021445A (en) * | 2017-08-15 | 2019-07-16 | 北京蜂鸟互动科技有限公司 | A kind of medical system based on VR model |
| US11756667B2 (en) * | 2018-05-30 | 2023-09-12 | Siemens Healthcare Gmbh | Decision support system for medical therapy planning |
| US11630995B2 (en) * | 2018-06-19 | 2023-04-18 | Siemens Healthcare Gmbh | Characterization of amount of training for an input to a machine-learned network |
| CN111128382B (en) * | 2019-12-30 | 2022-06-28 | 清华大学 | Artificial intelligence multimode imaging analysis device |
-
2019
- 2019-12-30 CN CN201911392016.0A patent/CN111128382B/en active Active
-
2020
- 2020-10-26 WO PCT/CN2020/123568 patent/WO2021135557A1/en active Application Filing
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20050119642A1 (en) * | 2001-12-21 | 2005-06-02 | Horia Grecu | Method and apparatus for eye registration |
| CN105310646A (en) * | 2015-12-09 | 2016-02-10 | 博奥颐和健康科学技术（北京）有限公司 | Human health condition in-vivo analysis system and method based on white eye shadowless imaging |
| CN105426695A (en) * | 2015-12-18 | 2016-03-23 | 北京铭光正讯科技有限公司 | Health status detecting system and method based on irises |
| CN206497027U (en) * | 2017-02-28 | 2017-09-15 | 福建师范大学 | Monochromatic light spectrometer polarization domain optical coherence tomography system based on photoswitch |
| CN107307848A (en) * | 2017-05-27 | 2017-11-03 | 天津海仁医疗技术有限公司 | A kind of recognition of face and skin detection system based on the micro- contrast imaging of high speed large area scanning optics |
| CN107680683A (en) * | 2017-10-09 | 2018-02-09 | 上海睦清视觉科技有限公司 | A kind of AI eye healths appraisal procedure |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2021135557A1 (en) * | 2019-12-30 | 2021-07-08 | 清华大学 | Artificial intelligence multi-mode imaging analysis apparatus |
| CN113180971A (en) * | 2021-04-29 | 2021-07-30 | 新乡市中心医院 | Traditional Chinese medicine nursing device and control method |
| CN113180971B (en) * | 2021-04-29 | 2023-09-22 | 新乡市中心医院 | Traditional Chinese medicine nursing device and control method |
| CN113390885A (en) * | 2021-08-17 | 2021-09-14 | 济南邦德激光股份有限公司 | Laser head cutting protective glass state detection device and detection method |
| CN113390885B (en) * | 2021-08-17 | 2021-11-09 | 济南邦德激光股份有限公司 | Laser head cutting protective glass state detection device and detection method |
| CN114664410A (en) * | 2022-03-11 | 2022-06-24 | 北京医准智能科技有限公司 | Video-based focus classification method and device, electronic equipment and medium |
| CN116636808A (en) * | 2023-06-28 | 2023-08-25 | 交通运输部公路科学研究所 | Intelligent cockpit driver visual health analysis method and device |
| CN116636808B (en) * | 2023-06-28 | 2023-10-31 | 交通运输部公路科学研究所 | Intelligent cockpit driver visual health analysis method and device |
Also Published As
| Publication number | Publication date |
|---|---|
| WO2021135557A1 (en) | 2021-07-08 |
| CN111128382B (en) | 2022-06-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111128382B (en) | Artificial intelligence multimode imaging analysis device | |
| US11963750B2 (en) | Systems, devices and methods for non-invasive hematological measurements | |
| US9445713B2 (en) | Apparatuses and methods for mobile imaging and analysis | |
| CN110448267B (en) | Multimode fundus dynamic imaging analysis system and method | |
| JP6200168B2 (en) | Image processing apparatus and image processing method | |
| US10219693B2 (en) | Systems and methods for combined structure and function evaluation of retina | |
| US20190045170A1 (en) | Medical image processing device, system, method, and program | |
| AU2003290334A1 (en) | A pupilometer | |
| EP4045138A1 (en) | Systems and methods for monitoring the functionality of a blood vessel | |
| WO2021106967A1 (en) | Ocular fundus image processing device and ocular fundus image processing program | |
| CN116172507A (en) | Eye motion capturing and tear film detecting system and equipment | |
| CN106073801A (en) | A kind of external cavum nasopharyngeum vena systemica blood oxygen saturation formation method and device | |
| Jansen et al. | A torsional eye movement calculation algorithm for low contrast images in video-oculography | |
| JP2003532461A (en) | Non-invasive measurement method using retinal image | |
| US11896345B2 (en) | Device for producing an image of depth-dependent morphological structures of skin lesions | |
| JP6604743B2 (en) | Information processing apparatus, operating method thereof, and computer program | |
| EP3695775B1 (en) | Smartphone-based handheld optical device and method for capturing non-mydriatic retinal images | |
| CN115375611A (en) | Model training-based refraction detection method and detection system | |
| US20230301511A1 (en) | Slit lamp microscope | |
| CN214048773U (en) | Eyeball motion inspection instrument | |
| US20220151482A1 (en) | Biometric ocular measurements using deep learning | |
| RU40704U1 (en) | AUTOMATED IRIDOLOGICAL DEVICE | |
| M’hiri et al. | Quality Assessment of Retinal Hyperspectral Images Using SURF and Intensity Features | |
| CN117530654A (en) | Real-time binocular pupil inspection system and detection method | |
| Talib et al. | A General Overview of Retinal Lesion Detecting Devices |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |