An ophthalmic imaging method and system for automatic retinal feature detection

Abstract

The present invention discloses an ophthalmic imaging method and system for automatic retinal feature detection, and relates to the technical field of ophthalmic detection. The present invention acquires multiple fundus images collected in a non-mydriatic state; acquires a fundus image collected in a dilated state corresponding to the fundus image collected in a non-mydriatic state; acquires a mapping relationship between the graphic features of the fundus image collected in the non-mydriatic state and the graphic features of the fundus image collected in the dilated state according to the fundus image collected in the non-mydriatic state and the fundus image collected in the corresponding dilated state; acquires the fundus image to be detected collected in the non-mydriatic state in real time; acquires the image features of the fundus image to be detected collected in the non-mydriatic state according to the fundus image to be detected collected in the non-mydriatic state acquired in real time. The present invention reduces the discomfort and harm caused to the eyes by the detection beam by controlling the detection beam.

Description

Technical Field

[0001] The present invention belongs to the technical field of ophthalmic detection, and in particular relates to an ophthalmic imaging method and system for automatic retinal feature detection. Background Art

[0002] After the fundus has lesions, a fundus examination is required. The conventional method is to dilate the pupil, illuminate the fundus, and then collect images of the fundus to obtain retinal features.

[0003] After the eye is instilled with medicine to dilate the pupil, the ciliary muscle of the eye is completely paralyzed. Prolonged strong light stimulation may cause discomfort. Therefore, it is necessary to reduce the discomfort and harm caused by the detection light beam to the eye according to the actual situation of the patient's fundus. Summary of the invention

[0004] The object of the present invention is to provide an ophthalmic imaging method and system for automatic retinal feature detection, which can reduce the discomfort and harm caused by the detection beam to the eyes by controlling the detection beam.

[0005] In order to solve the above technical problems, the present invention is achieved through the following technical solutions:

[0006] The present invention provides an ophthalmic imaging method for automatic retinal feature detection, which comprises:

[0007] Acquire multiple fundus images collected in a non-mydriatic state;

[0008] Acquire a fundus image collected in a dilated mydriatic state corresponding to a fundus image collected in a non-dilated mydriatic state;

[0009] Acquire a mapping relationship between graphic features of the fundus image collected in the non-mydriatic state and graphic features of the fundus image collected in the mydriatic state according to the fundus image collected in the non-mydriatic state and the corresponding fundus image collected in the mydriatic state;

[0010] Acquire the fundus image to be detected in real time under the non-mydriasis state;

[0011] Acquire image features of the fundus image to be detected collected in the non-mydriatic state according to the fundus image to be detected collected in the non-mydriatic state acquired in real time;

[0012] Acquire predicted graphic features of the fundus image collected in the mydriatic state according to the image features of the fundus image to be detected collected in the non-mydriatic state and the mapping relationship between the graphic features of the fundus image collected in the non-mydriatic state and the graphic features of the fundus image collected in the mydriatic state;

[0013] Obtaining a key fundus area to be inspected according to predicted graphic features of the fundus image collected in a dilated pupil state;

[0014] In the process of collecting fundus images in the mydriatic state, the fundus key areas to be inspected are collected first, and the fundus images collected in the mydriatic state are obtained;

[0015] The retinal features are obtained based on the fundus images collected in the mydriatic state.

[0016] In one embodiment of the present invention, the step of acquiring a mapping relationship between graphic features of the fundus image collected in the non-mydriatic state and graphic features of the fundus image collected in the mydriatic state based on the fundus image collected in the non-mydriatic state and the corresponding fundus image collected in the mydriatic state includes:

[0017] Acquiring graphic features of the fundus image collected in the non-mydriatic state according to the fundus image collected in the non-mydriatic state;

[0018] Acquiring graphic features of the fundus image collected in the mydriatic state according to the fundus image collected in the mydriatic state;

[0019] According to the graphic features of the fundus image collected in the non-mydriatic state and the corresponding graphic features of the fundus image collected in the dilated state, a mapping relationship between the graphic features of the fundus image collected in the non-mydriatic state and the graphic features of the fundus image collected in the dilated state is obtained.

[0020] In one embodiment of the present invention, in the process of collecting fundus images in a dilated pupil state, the fundus key areas to be inspected are collected first, and the step of obtaining the fundus images collected in a dilated pupil state includes:

[0021] Use a low-brightness positioning beam to illuminate the eye in a non-dilated pupil state to obtain the shape and position information of the fundus;

[0022] When the detection light beam starts to be emitted, the predicted graphic features of the fundus image collected in the mydriasis state are acquired;

[0023] A high-brightness detection beam is emitted into the eye in a dilated pupil state according to the shape and position information of the fundus, so that the detection beam irradiates the fundus to obtain a fundus image collected in the dilated pupil state.

[0024] In one embodiment of the present invention, the step of emitting a high-brightness detection beam into the eye in a dilated pupil state according to the shape and position information of the fundus so that the detection beam irradiates the fundus to obtain a fundus image collected in the dilated pupil state includes:

[0025] An electrode patch is arranged on the head of the human body to be sampled, wherein the electrode patch is arranged on the scalp surface corresponding to the visual imaging area of ​​the brain;

[0026] According to the detection current value of the electrode patch, the brightness of the detection beam emitted to the eye in the dilated pupil state is gradually increased until the brightness limit is reached.

[0027] In one embodiment of the present invention, the step of gradually increasing the brightness of the detection beam emitted to the eye in the dilated pupil state according to the detection current value of the electrode patch until the brightness limit is reached includes:

[0028] Send a safe low-brightness detection beam to the dilated eye;

[0029] Acquire a detection current value of the electrode patch corresponding to the brightness of a safe low brightness detection beam projected into an eye in a dilated pupil state;

[0030] According to the obtained value of the detection current value of the electrode patch and the safe low-brightness detection beam emitted to the eye in the dilated pupil state, a functional relationship between the brightness of the detection beam emitted to the eye in the dilated pupil state and the obtained value of the detection current value of the electrode patch is obtained;

[0031] Based on the functional relationship between the brightness of the detection beam incident on the eye in the dilated pupil state and the acquired value of the detection current value of the electrode patch, as well as the preset maximum value of the detection current value of the electrode patch, the maximum brightness of the detection beam incident on the eye in the dilated pupil state is obtained.

[0032] In one embodiment of the present invention, the step of obtaining the detection current value of the electrode patch corresponding to the brightness of the safe low-brightness detection beam emitted to the eye in the dilated pupil state includes:

[0033] Use multiple electrode patches to detect bioelectric current values ​​on the scalp surface corresponding to the visual imaging area of ​​the brain;

[0034] The corresponding weight is adjusted according to the distance between each electrode patch and the visual imaging area of ​​the brain.

[0035] In one embodiment of the present invention, the step of adjusting the corresponding weight according to the distance between each electrode patch and the visual imaging area of ​​the brain includes:

[0036] Divide each electrode patch into a number of planar units;

[0037] Divide the visual imaging area of ​​the brain into several volume units;

[0038] The weight of the detection current value of the electrode patch is adjusted according to the distance between each planar unit in each electrode patch and the volume unit in the visual imaging area of ​​the brain.

[0039] In one embodiment of the present invention, the step of adjusting the weight of the detection current value of the electrode patch according to the distance between each planar unit in each electrode patch and the volume unit in the visual imaging area of ​​the brain includes:

[0040] Get the center point position of each plane unit;

[0041] The weight of the detection current value of the electrode patch is adjusted according to the distance between the center point of each plane unit in each electrode patch and the volume unit in the visual imaging area of ​​the brain.

[0042] In one embodiment of the present invention, the step of adjusting the weight of the detection current value of the electrode patch according to the distance between the center point of each planar unit in each electrode patch and the center point of the volume unit in the visual imaging area of ​​the brain includes:

[0043] Get the center point position of each volume unit;

[0044] The weight of the detection current value of the electrode patch is adjusted according to the distance between the center point of each plane unit in each electrode patch and the center point of the volume unit in the visual imaging area of ​​the brain.

[0045] In one embodiment of the present invention, the multiple fundus images collected in a non-mydriatic state are obtained by a fundus camera, and the fundus camera includes:

[0046] S1: the lens of the mobile fundus camera is aimed at the pupil in a non-mydriasis state;

[0047] S2: Control the lens of the fundus camera to approach the eyeball to acquire an image, and adjust the image presented in the fundus camera;

[0048] S3: selecting the image pictures presented in the fundus camera adjusted at different focal lengths according to the distance between the lens of the fundus camera and the eyeball, and obtaining a plurality of fundus images collected in a non-mydriasis state; wherein S3 also includes:

[0049] Analyze whether the image reaches a preset clarity. When the image does not reach the preset clarity, adjust the clarity of the image. When the image reaches the preset clarity, determine whether a light spot phenomenon occurs in the image.

[0050] If a light spot phenomenon appears in the image, the position of the light spot phenomenon is adjusted until the position of the light spot phenomenon coincides with the center of the image, and the image adjusted under the focal length state is obtained. Then the focal length is adjusted, and the image is re-analyzed to see whether it reaches the preset clarity. This cycle is repeated until the target number of image adjusted under different focal lengths are obtained.

[0051] In one embodiment of the present invention, the detection beam is emitted by a detection instrument, and the monitoring instrument also analyzes the detection beam when emitting the detection beam, and starts to emit the detection beam when the detection beam meets the requirements, wherein the analysis of the detection beam includes: estimating the intensity of the detection beam, and obtaining detection beam intensity estimation data:

[0052]

[0053] Where R represents the estimated data of the detection beam intensity; C represents the central intensity of the maximum bright spot in the central area of ​​the detection beam; J α (X) represents the α-order Bessel function, where X is represented by α is 1; π is the ratio of circumference to circumference; r is the aperture of the emission outlet of the monitoring instrument; L is the wavelength; t is the angle between the emission light and the main optical axis; d is the image distance;

[0054] Determine the analysis result based on the detection beam intensity estimation data:

[0055]

[0056] Among them, W represents the sensitive range of human eyes to light beams; G represents the analysis result. When G=T, it means that the detection beam meets the requirements and the detection beam is emitted. When G=F, it means that the detection beam does not meet the requirements and the detection beam is not emitted.

[0057] The present invention also discloses an ophthalmic imaging system for automatic retinal feature detection, which is characterized by comprising:

[0058] A history record collection unit, used for acquiring a plurality of fundus images collected in a non-mydriasis state;

[0059] Acquire a fundus image collected in a dilated mydriatic state corresponding to a fundus image collected in a non-dilated mydriatic state;

[0060] Acquire a mapping relationship between graphic features of the fundus image collected in the non-mydriatic state and graphic features of the fundus image collected in the mydriatic state according to the fundus image collected in the non-mydriatic state and the corresponding fundus image collected in the mydriatic state;

[0061] A detection unit, used for acquiring in real time the fundus image to be detected collected in a non-mydriasis state;

[0062] Acquire image features of the fundus image to be detected collected in the non-mydriatic state according to the fundus image to be detected collected in the non-mydriatic state acquired in real time;

[0063] Acquire predicted graphic features of the fundus image collected in the mydriatic state according to the image features of the fundus image to be detected collected in the non-mydriatic state and the mapping relationship between the graphic features of the fundus image collected in the non-mydriatic state and the graphic features of the fundus image collected in the mydriatic state;

[0064] Obtaining a key fundus area to be inspected according to predicted graphic features of the fundus image collected in a dilated pupil state;

[0065] In the process of collecting fundus images in the mydriatic state, the fundus key areas to be inspected are collected first, and the fundus images collected in the mydriatic state are obtained;

[0066] The retinal features are obtained based on the fundus images collected in the mydriatic state.

[0067] The present invention reduces the discomfort and harm caused by the detection light beam to eyes by controlling the detection light beam.

[0068] Of course, any product implementing the present invention does not necessarily need to achieve all of the advantages described above at the same time. BRIEF DESCRIPTION OF THE DRAWINGS

[0069] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for describing the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0070] Figure 1 A schematic diagram of an implementation flow of an ophthalmic imaging method for automatic retinal feature detection according to an embodiment of the present invention;

[0071] Figure 2 A schematic diagram of an implementation flow of the step S3 of the present invention of acquiring a mapping relationship between graphic features of a fundus image collected in a non-mydriatic state and graphic features of a fundus image collected in a dilated state according to a fundus image collected in a non-mydriatic state and a corresponding fundus image collected in a dilated state in one embodiment;

[0072] Figure 3 It is a schematic diagram of an implementation flow of an embodiment of the step of obtaining the fundus image collected in the mydriatic state in the process of collecting the fundus image in the mydriatic state in step S8 of the present invention, giving priority to collecting the fundus key area to be inspected, and obtaining the fundus image collected in the mydriatic state;

[0073] Figure 4A schematic diagram of an implementation flow of the step S83 of the present invention of emitting a high-brightness detection beam into an eye in a dilated pupil state according to the shape and position information of the fundus, so that the detection beam irradiates the fundus, so as to obtain a fundus image collected in a dilated pupil state, in one embodiment;

[0074] Figure 5 This is a schematic diagram of an implementation flow of the step S832 of the present invention, in which the brightness of the detection light beam incident on the eye in the dilated pupil state is gradually increased according to the detection current value of the electrode patch until the brightness limit is reached, in one embodiment;

[0075] Figure 6 A schematic diagram of module connections of an ophthalmic imaging system for automatic retinal feature detection according to an embodiment of the present invention.

[0076] In the accompanying drawings, the components represented by the reference numerals are listed as follows:

[0077] 1-history collection unit, 2-detection unit. DETAILED DESCRIPTION

[0078] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0079] See also Figure 1As shown, the present invention provides an ophthalmic imaging method for automatic retinal feature detection. During the execution process, step S1 can be first executed to obtain multiple fundus images collected in the non-mydriatic state, and then step S2 can be executed to obtain the fundus image collected in the dilated state corresponding to the fundus image collected in the non-mydriatic state. Next, step S3 can be executed to obtain the mapping relationship between the graphic features of the fundus image collected in the non-mydriatic state and the graphic features of the fundus image collected in the dilated state based on the fundus image collected in the non-mydriatic state and the corresponding fundus image collected in the dilated state. Next, step S4 can be executed to obtain the fundus image to be detected collected in the non-mydriatic state in real time. Next, step S5 can be executed to obtain the image features of the fundus image to be detected collected in the non-mydriatic state based on the fundus image to be detected collected in the non-mydriatic state acquired in real time. Next, step S6 can be executed to obtain the predicted graphic features of the fundus image collected in the mydriatic state based on the image features of the fundus image to be detected collected in the non-mydriatic state and the mapping relationship between the graphic features of the fundus image collected in the non-mydriatic state and the graphic features of the fundus image collected in the mydriatic state. Next, step S7 can be executed to obtain the key fundus areas to be inspected based on the predicted graphic features of the fundus image collected in the mydriatic state. Next, step S8 can be executed to preferentially collect the key fundus areas to be inspected during the process of collecting fundus images in the mydriatic state, and obtain the fundus images collected in the mydriatic state. Finally, step S9 can be executed to obtain retinal features based on the fundus images collected in the mydriatic state. The above-mentioned method can reduce the discomfort and harm caused to the eyes by the detection beam.

[0080] See also Figure 2 As shown, in order to obtain the mapping relationship between the graphic features of the fundus image collected in the non-mydriatic state and the graphic features of the fundus image collected in the dilated state, in the above step S3, step S31 can be first executed to obtain the graphic features of the fundus image collected in the non-mydriatic state according to the fundus image collected in the non-mydriatic state. Next, step S32 can be executed to obtain the graphic features of the fundus image collected in the dilated state according to the fundus image collected in the dilated state. Finally, step S33 can be executed to obtain the mapping relationship between the graphic features of the fundus image collected in the non-mydriatic state and the graphic features of the fundus image collected in the dilated state according to the graphic features of the fundus image collected in the non-mydriatic state and the corresponding graphic features of the fundus image collected in the dilated state.

[0081] See also Figure 3As shown, in order to obtain the fundus image collected in the mydriatic state, in the above step S8, step SS81 can be first executed to use a low-brightness positioning beam to illuminate the eye in the non-mydriatic state to obtain the shape and position information of the fundus. Next, step S82 can be executed to start emitting a detection beam and start acquiring the predicted graphic features of the fundus image collected in the mydriatic state. Finally, step S83 can be executed to emit a high-brightness detection beam into the eye in the mydriatic state according to the shape and position information of the fundus, so that the detection beam is irradiated to the fundus to obtain the fundus image collected in the mydriatic state. The fundus image collected in the mydriatic state is obtained in the above manner.

[0082] See also Figure 4 As shown, in order to obtain the fundus image collected in the mydriatic state, in the above step S83, step S831 can be first executed to set an electrode patch on the head of the person to be collected, and the electrode patch is set on the scalp surface corresponding to the brain visual imaging area. Next, step S832 can be executed to gradually increase the brightness of the detection beam injected into the eye in the mydriatic state according to the detection current value of the electrode patch until the brightness limit is reached. The fundus image collected in the mydriatic state is obtained in the above manner.

[0083] Please refer to Figure 5. In order to protect the eyes, in the above step S832, step S8321 can be first executed to inject a safe low-brightness detection beam into the eyes in the dilated pupil state. Next, step S8322 can be executed to obtain the detection current value of the electrode patch corresponding to the brightness of the safe low-brightness detection beam injected into the eyes in the dilated pupil state. Next, step S8323 can be executed to obtain the functional relationship between the brightness of the detection beam injected into the eyes in the dilated pupil state and the acquired value of the detection current value of the electrode patch according to the acquired value of the safe low-brightness detection beam injected into the eyes in the dilated pupil state and the detection current value of the electrode patch. Finally, step S8324 can be executed to obtain the maximum brightness of the detection beam injected into the eyes in the dilated pupil state according to the functional relationship between the brightness of the detection beam injected into the eyes in the dilated pupil state and the acquired value of the detection current value of the electrode patch, as well as the preset maximum value of the detection current value of the electrode patch. Limiting the brightness by limiting means plays a role in protecting the eyes.

[0084] In one embodiment of the present solution, the step of obtaining the detection current value of the electrode patch corresponding to the brightness of the safe low brightness detection beam emitted to the eye in the dilated pupil state can firstly perform the step of using multiple electrode patches to detect the bioelectric current value of the scalp surface corresponding to the brain visual imaging area. Finally, the step of adjusting the corresponding weight according to the distance between each electrode patch and the brain visual imaging area can be performed.

[0085] In one embodiment of the present scheme, the step of adjusting the corresponding weight according to the distance between each electrode patch and the visual imaging area of ​​the brain can first be performed by dividing each electrode patch into a number of planar units, then dividing the visual imaging area of ​​the brain into a number of volume units, and then adjusting the weight of the detection current value of the electrode patch according to the distance between each planar unit in each electrode patch and the volume unit in the visual imaging area of ​​the brain.

[0086] In one embodiment of the present scheme, the step of adjusting the weight of the detection current value of the electrode patch according to the distance between each planar unit in each electrode patch and the volume unit in the visual imaging area of ​​the brain can firstly obtain the center point position of each planar unit, and then adjust the weight of the detection current value of the electrode patch according to the distance between the center point of each planar unit in each electrode patch and the volume unit in the visual imaging area of ​​the brain.

[0087] In one embodiment of the present scheme, the step of adjusting the weight of the detection current value of the electrode patch according to the distance between the center point of each planar unit in each electrode patch and the center point of the volume unit in the brain visual imaging area can be performed by first obtaining the center point position of each volume unit, and then adjusting the weight of the detection current value of the electrode patch according to the distance between the center point of each planar unit in each electrode patch and the center point of the volume unit in the brain visual imaging area.

[0088] In one embodiment of the present solution, a plurality of fundus images collected in a non-mydriasis state are obtained by a fundus camera, and the fundus camera includes:

[0089] S1: the lens of the mobile fundus camera is aimed at the pupil in a non-mydriasis state;

[0090] S2: Control the lens of the fundus camera to approach the eyeball to acquire an image, and adjust the image presented in the fundus camera;

[0091] S3: selecting the image pictures presented in the fundus camera adjusted at different focal lengths according to the distance between the lens of the fundus camera and the eyeball, and obtaining a plurality of fundus images collected in a non-mydriasis state; wherein S3 also includes:

[0092] Analyze whether the image reaches a preset clarity. When the image does not reach the preset clarity, adjust the clarity of the image. When the image reaches the preset clarity, determine whether a light spot phenomenon occurs in the image.

[0093] If a light spot phenomenon appears in the image, the position of the light spot phenomenon is adjusted until the position of the light spot phenomenon coincides with the center of the image, and the image adjusted under the focal length state is obtained. Then the focal length is adjusted, and the image is re-analyzed to see whether it reaches the preset clarity. This cycle is repeated until the target number of image adjusted under different focal lengths are obtained.

[0094] If a light spot phenomenon appears in the image, the focal length is directly adjusted so that the position of the light spot phenomenon is adjusted after the light spot phenomenon appears in the image presented by the fundus camera. The above-mentioned fundus camera can not only obtain the fundus image with high precision, but also facilitate the adjustment of the image during the process of obtaining the fundus image, and the obtained image can more accurately present the fundus condition through the clarity adjustment and the light spot phenomenon adjustment, so that more accurate fundus information can be obtained from the image. In addition, the image presented in the fundus camera adjusted at different focal lengths to obtain multiple fundus images collected in the non-mydriatic state can effectively eliminate accidental conditions, reduce information errors, and improve the accuracy of automatic retinal feature detection.

[0095] In one embodiment of the present solution, the detection beam is emitted by the detection instrument. When the detection beam is emitted, the monitoring instrument also analyzes the detection beam. When the detection beam meets the requirements, the detection beam is emitted. The analysis of the detection beam includes: estimating the intensity of the detection beam and obtaining the detection beam intensity estimation data:

[0096]

[0097] Where R represents the estimated data of the detection beam intensity; C represents the central intensity of the maximum bright spot in the central area of ​​the detection beam; J α (X) represents the α-order Bessel function, where X is represented by α is 1; π is the ratio of circumference to circumference; r is the aperture of the emission outlet of the monitoring instrument; L is the wavelength; t is the angle between the emission light and the main optical axis; d is the image distance;

[0098] Determine the analysis results based on the test beam intensity estimation data:

[0099]

[0100] Wherein, W represents the sensitive range of the human eye to the light beam; G represents the analysis result. When G=T, it means that the detection beam meets the requirements and the detection beam is emitted. When G=F, it means that the detection beam does not meet the requirements and the detection beam is not emitted. When the above-mentioned monitoring instrument emits the detection beam, it analyzes the detection beam so that the emitted detection beam can better detect the fundus. At the same time, it can also avoid the damage of the detection beam to the eyes and reduce the instantaneous effect of the detection beam on the eyes, so as to reduce the error between the predicted graphic features of the fundus image collected in the dilated pupil state and the actual situation, thereby improving the accuracy of automatic retinal feature detection.

[0101] The present solution also provides an ophthalmic imaging system for automatic retinal feature detection, which may include a history record collection unit 1 and a detection unit 2. The history record collection unit 1 is used to obtain multiple fundus images collected in a non-mydriatic state, and obtain a fundus image collected in a dilated state corresponding to the fundus image collected in the non-mydriatic state. A mapping relationship between the graphic features of the fundus image collected in the non-mydriatic state and the graphic features of the fundus image collected in the dilated state is obtained based on the fundus image collected in the non-mydriatic state and the corresponding fundus image collected in the dilated state. The detection unit 2 is used for acquiring the fundus image to be detected collected in the non-mydriatic state in real time, acquiring the image features of the fundus image to be detected collected in the non-mydriatic state according to the fundus image to be detected collected in the non-mydriatic state acquired in real time, acquiring the predicted graphic features of the fundus image collected in the dilated state according to the image features of the fundus image to be detected collected in the non-mydriatic state and the mapping relationship between the graphic features of the fundus image collected in the non-mydriatic state and the graphic features of the fundus image collected in the dilated state, acquiring the key fundus area to be inspected according to the predicted graphic features of the fundus image collected in the dilated state, and in the process of collecting the fundus image in the dilated state, giving priority to collecting the key fundus area to be inspected, acquiring the fundus image collected in the dilated state, and acquiring the retinal features according to the fundus image collected in the dilated state.

[0102] In summary, by controlling the detection light beam, the discomfort and harm caused by the detection light beam to the eyes can be reduced.

[0103] The above description of the illustrated embodiments of the present invention (including the contents in the abstract of the specification) is not intended to be an exhaustive list or to limit the present invention to the precise form disclosed herein. Although specific embodiments of the present invention and examples of the present invention are described herein for illustrative purposes only, as will be recognized and appreciated by those skilled in the art, various equivalent modifications are possible within the spirit and scope of the present invention. As noted, these modifications may be made to the present invention in accordance with the above description of the embodiments of the present invention, and these modifications will be within the spirit and scope of the present invention.

[0104] Systems and methods have been generally described herein as details that aid in understanding the present invention. In addition, various specific details have been given to provide an overall understanding of embodiments of the present invention. However, those skilled in the relevant art will recognize that embodiments of the present invention may be practiced without one or more of the specific details, or may be practiced using other devices, systems, accessories, methods, components, materials, parts, etc. In other cases, well-known structures, materials, and / or operations are not specifically shown or described in detail to avoid confusion with various aspects of embodiments of the present invention.

[0105] Thus, although the invention has been described herein with reference to specific embodiments thereof, freedom of modification, various changes and substitutions are intended within the foregoing disclosure, and it should be understood that in some cases, some features of the invention will be employed without the corresponding use of other features without departing from the scope and spirit of the proposed invention. Thus, many modifications may be made to adapt a particular environment or material to the essential scope and spirit of the invention. The invention is not intended to be limited to the specific terms used in the claims below and / or the specific embodiments disclosed as the best mode contemplated for carrying out the invention, but the invention will include any and all embodiments and equivalents falling within the scope of the appended claims. Thus, the scope of the invention will be determined solely by the appended claims.

Claims

1. An ophthalmic imaging method for automatic retinal feature detection, characterized in that: include, Acquire multiple fundus images collected in a non-mydriatic state; Acquire a fundus image collected in a dilated mydriatic state corresponding to a fundus image collected in a non-dilated mydriatic state; Acquire a mapping relationship between graphic features of the fundus image collected in the non-mydriatic state and graphic features of the fundus image collected in the mydriatic state according to the fundus image collected in the non-mydriatic state and the corresponding fundus image collected in the mydriatic state; Acquire the fundus image to be detected in real time under the non-mydriasis state; Acquire image features of the fundus image to be detected collected in the non-mydriatic state according to the fundus image to be detected collected in the non-mydriatic state acquired in real time; Acquire predicted graphic features of the fundus image collected in the mydriatic state according to the image features of the fundus image to be detected collected in the non-mydriatic state and the mapping relationship between the graphic features of the fundus image collected in the non-mydriatic state and the graphic features of the fundus image collected in the mydriatic state; Obtaining a key fundus area to be inspected according to predicted graphic features of the fundus image collected in a dilated pupil state; In the process of collecting fundus images in the mydriatic state, the fundus key areas to be inspected are collected first, and the fundus images collected in the mydriatic state are obtained; The retinal features are obtained based on the fundus images collected in the mydriatic state.

2. The method according to claim 1, characterized in that The step of acquiring a mapping relationship between the graphic features of the fundus image collected in the non-mydriatic state and the graphic features of the fundus image collected in the mydriatic state according to the fundus image collected in the non-mydriatic state and the corresponding fundus image collected in the mydriatic state includes: Acquiring graphic features of the fundus image collected in the non-mydriatic state according to the fundus image collected in the non-mydriatic state; Acquiring graphic features of the fundus image collected in the mydriatic state according to the fundus image collected in the mydriatic state; According to the graphic features of the fundus image collected in the non-mydriatic state and the corresponding graphic features of the fundus image collected in the dilated state, a mapping relationship between the graphic features of the fundus image collected in the non-mydriatic state and the graphic features of the fundus image collected in the dilated state is obtained.

3. The method according to claim 2, characterized in that In the process of collecting fundus images in the mydriatic state, the fundus key areas to be inspected are collected first, and the fundus images collected in the mydriatic state are obtained. include, Use a low-brightness positioning beam to illuminate the eye in a non-dilated pupil state to obtain the shape and position information of the fundus; When the detection beam starts to be emitted, the predicted graphic features of the fundus image collected in the mydriasis state are obtained; A high-brightness detection beam is emitted into the eye in a dilated pupil state according to the shape and position information of the fundus, so that the detection beam irradiates the fundus to obtain a fundus image collected in the dilated pupil state.

4. The method according to claim 3, characterized in that The step of emitting a high-brightness detection beam into the eye in a dilated pupil state according to the shape and position information of the fundus, so that the detection beam irradiates the fundus, and obtaining a fundus image collected in a dilated pupil state, includes: An electrode patch is arranged on the head of the human body to be sampled, wherein the electrode patch is arranged on the scalp surface corresponding to the visual imaging area of ​​the brain; According to the detection current value of the electrode patch, the brightness of the detection beam emitted to the eye in the dilated pupil state is gradually increased until the brightness limit is reached.

5. The method according to claim 4, characterized in that The step of gradually increasing the brightness of the detection beam emitted to the eye in the dilated pupil state according to the detection current value of the electrode patch until the brightness limit is reached includes: Send a safe low-brightness detection beam to the dilated eye; Acquire a detection current value of the electrode patch corresponding to the brightness of a safe low brightness detection beam projected into an eye in a dilated pupil state; According to the obtained value of the detection current value of the electrode patch and the safe low-brightness detection beam emitted to the eye in the dilated pupil state, a functional relationship between the brightness of the detection beam emitted to the eye in the dilated pupil state and the obtained value of the detection current value of the electrode patch is obtained; Based on the functional relationship between the brightness of the detection beam incident on the eye in the dilated pupil state and the acquired value of the detection current value of the electrode patch, as well as the preset maximum value of the detection current value of the electrode patch, the maximum brightness of the detection beam incident on the eye in the dilated pupil state is obtained.

6. The method according to claim 5, characterized in that The step of obtaining the detection current value of the electrode patch corresponding to the brightness of the safe low-brightness detection light beam emitted to the eye in the dilated pupil state, include, Use multiple electrode patches to detect bioelectric current values ​​on the scalp surface corresponding to the visual imaging area of ​​the brain; Adjust the corresponding weight according to the distance between each electrode patch and the visual imaging area of ​​the brain; The step of adjusting the corresponding weight according to the distance between each electrode patch and the visual imaging area of ​​the brain includes: Divide each electrode patch into a number of planar units; Divide the visual imaging area of ​​the brain into several volume units; The weight of the detection current value of the electrode patch is adjusted according to the distance between each planar unit in each electrode patch and the volume unit in the visual imaging area of ​​the brain.

7. The method according to claim 6, characterized in that The step of adjusting the weight of the detection current value of the electrode patch according to the distance between each plane unit in each electrode patch and the volume unit in the visual imaging area of ​​the brain, include, Get the center point position of each plane unit; Adjusting the weight of the detection current value of the electrode patch according to the distance between the center point of each plane unit in each electrode patch and the volume unit in the visual imaging area of ​​the brain; The step of adjusting the weight of the detection current value of the electrode patch according to the distance between the center point of each planar unit in each electrode patch and the center point of the volume unit in the visual imaging area of ​​the brain includes: Get the center point position of each volume unit; The weight of the detection current value of the electrode patch is adjusted according to the distance between the center point of each plane unit in each electrode patch and the center point of the volume unit in the visual imaging area of ​​the brain.

8. The method according to claim 3, characterized in that The detection beam is emitted by a monitoring instrument. When emitting the detection beam, the monitoring instrument also analyzes the detection beam. When the detection beam meets the requirements, the detection beam is emitted. The analysis of the detection beam includes: estimating the intensity of the detection beam and obtaining detection beam intensity estimation data: Where R represents the estimated data of the detection beam intensity; C represents the central intensity of the maximum bright spot in the central area of ​​the detection beam; J α (X) represents the α-order Bessel function, where X is represented by α is 1; π is the ratio of circumference to circumference; r is the aperture of the emission outlet of the monitoring instrument; L is the wavelength; t is the angle between the emission light and the main optical axis; d is the image distance; Determine the analysis result based on the detection beam intensity estimation data: Among them, W represents the sensitive range of human eyes to light beams; G represents the analysis result. When G=T, it means that the detection beam meets the requirements and the detection beam is emitted. When G=F, it means that the detection beam does not meet the requirements and the detection beam is not emitted.

9. The method according to claim 1, characterized in that: The fundus images collected in the non-mydriasis state are obtained by a fundus camera, and the fundus camera includes the following steps when obtaining the fundus images: In a non-mydriatic state, move the lens of the fundus camera to align with the pupil; Controlling the lens of the fundus camera to approach the eyeball to acquire an image, and adjusting the image presented in the fundus camera; Selecting the image pictures presented in the fundus camera adjusted at different focal lengths according to the distance between the lens of the fundus camera and the eyeball, and obtaining a plurality of fundus images collected in a non-mydriasis state, wherein: Analyze whether the image reaches a preset clarity. When the image does not reach the preset clarity, adjust the clarity of the image. When the image reaches the preset clarity, determine whether a light spot phenomenon occurs in the image. If a light spot phenomenon appears in the image, the position of the light spot phenomenon is adjusted until the position of the light spot phenomenon coincides with the center of the image, and the image adjusted under the focal length state is obtained. Then the focal length is adjusted, and the image is re-analyzed to see whether it reaches the preset clarity. This cycle is repeated until the target number of image adjusted under different focal lengths are obtained.

10. An ophthalmic imaging system for automatic retinal feature detection, characterized in that: include, A history record collection unit, used for acquiring a plurality of fundus images collected in a non-mydriasis state; Acquire a fundus image collected in a dilated mydriatic state corresponding to a fundus image collected in a non-dilated mydriatic state; Acquire a mapping relationship between graphic features of the fundus image collected in the non-mydriatic state and graphic features of the fundus image collected in the mydriatic state according to the fundus image collected in the non-mydriatic state and the corresponding fundus image collected in the mydriatic state; A detection unit, used for acquiring in real time the fundus image to be detected collected in a non-mydriasis state; Acquire image features of the fundus image to be detected collected in the non-mydriatic state according to the fundus image to be detected collected in the non-mydriatic state acquired in real time; Acquire predicted graphic features of the fundus image collected in the mydriatic state according to the image features of the fundus image to be detected collected in the non-mydriatic state and the mapping relationship between the graphic features of the fundus image collected in the non-mydriatic state and the graphic features of the fundus image collected in the mydriatic state; Obtaining a key fundus area to be inspected according to predicted graphic features of the fundus image collected in a dilated pupil state; In the process of collecting fundus images in the mydriatic state, the fundus key areas to be inspected are collected first, and the fundus images collected in the mydriatic state are obtained; The retinal features are obtained based on the fundus images collected in the mydriatic state.

Images