Vision screening apparatus including color imaging

Abstract

Described herein is a visual screening device for performing a visual screening test on a patient to determine the presence or absence of a disease and / or abnormality in an eye of the patient.SOLUTION: The visual screening device may include associated methods and systems configured to perform the operations of the visual screening test. The apparatus may include a radiation source configured to generate near infrared (NIR) radiation, a sensor configured to capture a grayscale image representative of the radiation reflected by the patient's eye, a white light source, and a camera configured to capture a color image of the patient's eye. The apparatus may also be configured to generate a composite image based on the grayscale image and / or the color image, determine a difference between an eye-related value and an expected value, and generate an output indicative of an eye-related condition based on the difference.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] This application relates to medical devices, and more particularly to vision screening devices and associated systems and methods for detecting and assessing ocular diseases and disorders. [Background technology]

[0002] Vision screening typically involves screening for ocular diseases. Such screening may also include transillumination tests, such as the Bruckner red reflex test. In a red reflex test, a clinician uses an ophthalmoscope to shine visible light into a patient's eye and examine the color and other properties of the light reflected by the eye's choroidal and retinal surfaces. This test can be used to detect a variety of ocular diseases and abnormalities, including corneal and medial opacities, cataracts, and retinal abnormalities, including tumors and retinoblastoma. Vision screening for diseases is recommended for all age groups. For example, newborns may be screened for congenital eye diseases, and older adults may be screened for the development of age-related degenerative diseases, such as cataracts and retinal diseases. Vision screening under visible light can also be used to detect the presence of foreign bodies in the eye. Summary of the Invention [Problem to be solved by the invention]

[0003] Vision screening also typically includes one or more tests to determine various deficiencies related to a patient's eyes. Examples of such vision tests include refractive error testing, accommodation testing, visual acuity testing, and color vision screening. Some vision screening tests require the use of infrared or near-infrared imaging, while others require visible light imaging and / or a display screen to guide the patient. However, ophthalmic testing devices, such as phoropters, autorefractors, and photorefractors, may only be capable of performing a limited range of tests. It would be advantageous to be able to screen for the majority of vision defects and disorders using a single, integrated device.

[0004] Various examples of the present disclosure are directed to overcoming one or more of the problems set forth above. [Means for solving the problem]

[0005] In one example of the present disclosure, a vision screening device includes a radiation source configured to emit radiation at a first wavelength (e.g., in the near-infrared band), a sensor configured to capture radiation reflected by a patient's eye, a white light source, and a camera configured to capture a color image of the patient's eye. The vision screening device also includes a processor operatively connected to the radiation source, the sensor, the white light source, and the camera, and a memory storing instructions executable by the processor that, when executed, cause the radiation source to emit radiation at the first wavelength for a first time period, cause the sensor to capture a portion of the radiation reflected by the patient's eye for the first time period, cause the white light source to illuminate the patient's eye for a second time period after the first time period, and cause the camera to capture a color image of the patient's eye for the second time period. The instructions, when executed, also cause the processor to generate a composite image of the eye including a first plurality of pixels representing a grayscale image indicative of the captured portion of the radiation and a second plurality of pixels representing a color image, determine a difference between a value associated with the eye and an expected value based on the composite image, and generate an output indicative of a condition associated with the eye based at least in part on the difference.

[0006] In another example of the present disclosure, a method includes causing a radiation source to illuminate a patient's eye for a first time period, causing a sensor to capture a grayscale image of the eye for the first time period, causing a white light source to illuminate the eye for a second time period separate from the first time period, and causing a camera to capture a color image of the eye for the second time period, and also including generating a composite image of the eye deriving a first plurality of pixel values ​​from the grayscale image and a second plurality of pixel values ​​from the color image, and determining, based at least in part and independently on at least one of the color image or the composite image, a characteristic of the eye revealed by the NIR and visible light, an output associated with the patient.

[0007] In yet another example of the present disclosure, a system includes a memory; a processor; and computer-executable instructions stored in the memory and executable by the processor, which, when executed, cause the processor to: cause a radiation source to emit near-infrared (NIR) radiation for a first time period; cause a sensor to capture a portion of the NIR radiation reflected by a patient's eye for the first time period; cause a white light source to illuminate the eye for a second time period separate from the first time period; and cause a camera to capture a color image of the eye for the second time period. The instructions, when executed, also cause the processor to determine a difference between a value associated with the eye and an expected value based on the color image and the portion of the NIR radiation; determine that the difference is greater than or equal to a threshold; and generate an output indicative of a condition of the eye based at least in part on determining that the difference is greater than or equal to the threshold.

[0008] The features of the present disclosure, its nature and various advantages may become more apparent from the following detailed description taken in conjunction with the accompanying drawings. [Brief explanation of the drawings]

[0009] [Figure 1] 1 illustrates an exemplary vision screening device and system of the present disclosure. In some embodiments, the components of the exemplary system illustrated in FIG. 1 can be used to perform one or more screening tests related to vision screening and / or detection of ocular diseases or abnormalities. [Figure 2] FIG. 2 illustrates an exemplary visual screening device of the present disclosure. [Figure 3A] FIG. 3A illustrates another exemplary visual screening device of the present disclosure. [Figure 3B] FIG. 3B illustrates the placement of radiation sources in an exemplary vision screening device of the present disclosure. [Figure 3C] FIG. 3C is a schematic diagram of an exemplary visual screening system of the present disclosure. [Figure 3D] FIG. 3D is another schematic diagram of an exemplary visual screening device of the present disclosure. [Figure 4A]FIG. 4A illustrates exemplary features that can be used by a vision screening device to determine eye diseases and abnormalities from eye images, according to examples of the present disclosure. [Figure 4B] FIG. 4B illustrates exemplary features that can be used by a vision screening device to determine eye diseases and abnormalities from eye images, according to examples of the present disclosure. [Figure 4C] FIG. 4C illustrates exemplary features that may be used by a vision screening device to determine eye diseases and abnormalities from eye images, according to examples of the present disclosure. [Figure 4D] FIG. 4D illustrates exemplary features that can be used by a vision screening device to determine eye diseases and abnormalities from eye images, according to examples of the present disclosure. [Figure 5A] FIG. 5A shows an exemplary visualization produced by the visual screening device of the present disclosure. [Figure 5B] FIG. 5B shows an exemplary visualization produced by the visual screening device of the present disclosure. [Figure 6] FIG. 6 is a first flow chart illustrating an exemplary method of the present disclosure. [Figure 7] FIG. 7 is a second flow chart illustrating an exemplary method of the present disclosure. [Figure 8] FIG. 8 is a third flow chart illustrating an exemplary method of the present disclosure.

[0010] In the drawings, the most significant digit(s) of a reference number indicates the drawing number in which the reference number first appears. The use of the same reference number in different drawings indicates similar or identical elements or features. The drawings are not to scale. DETAILED DESCRIPTION OF THE INVENTION

[0011] The present disclosure relates, in part, to a vision screening device and corresponding method. Such an exemplary vision screening device may be configured to administer one or more vision screening tests to a patient and output the results of the vision screening tests to an operator of the device, such as a clinician or physician's assistant. Specifically, the disclosure relates to devices and methods for screening for ocular diseases and abnormalities. For example, the vision screening device may capture one or more images of an eye illuminated with radiation in different wavelength ranges of the electromagnetic spectrum (e.g., infrared, near-infrared, and visible light). Based on analysis of the captured images, the device may determine one or more ocular diseases and / or abnormalities, such as cataracts, tumors, refractive errors, intraocular foreign bodies, corneal abrasions, retinal detachments or lesions, congenital disorders, etc., associated with one or both of the patient's eyes.

[0012] Based at least in part on the analysis of the captured images, the device can generate an output including at least one of a recommendation or a diagnosis for the patient. Such output (e.g., a recommendation and / or diagnosis) may indicate a detected disease or abnormality, that the patient needs additional screening, or that the screening was normal (e.g., no disease or abnormality was detected). For example, the device can determine the difference between an image of the patient's left eye and an image of the patient's right eye and compare the difference to standard test data corresponding to a normal eye to provide a recommendation and / or diagnosis. In particular, the standard test data may provide one or more thresholds or ranges of values, and the output generated by the device may be based on the difference being below the threshold or within the range of values. The device can also generate a visualization of the captured image for display to a clinician or operator of the vision screening device to assist the clinician or operator in determining a diagnosis. In this manner, the methods described herein can provide an automated diagnosis based on the analysis of images captured by the vision screening device. The methods described herein can also provide an automated recommendation based on and / or indicative of such a diagnosis.

[0013] As described with at least reference to FIG. 1 , an exemplary vision screening device associated with screening for ocular diseases and abnormalities may include components for capturing images of a patient's eye under near-infrared and visible light. The device may also include components for controlling near-infrared and visible light emission and corresponding capture of radiation reflected from the eye during screening. In some examples, near-infrared images may be captured before capturing visible light images so that the pupil of the patient's eye does not constrict, i.e., adjust, in response to visible light during screening, and screening may be completed without the need for eye dilation. Additionally, the device may include components for analyzing the captured images to determine a disease state and / or abnormality in the patient's eye and components for determining and reporting an output indicative of a disease state and / or abnormality detected during screening.

[0014] Additional details regarding the above-described devices and techniques are provided below with reference to Figures 1-7. While these figures illustrate devices and systems that may utilize the claimed methods, it will be understood that the methods, processes, functions, acts and / or techniques described herein may be applied to other devices, systems, etc. as well.

[0015] FIG. 1 illustrates an exemplary environment 100 for administering a vision screening test, particularly a screening test for detecting eye diseases and / or abnormalities, according to some embodiments. As shown in FIG. 1 , in some examples, an operator 102 may administer a vision screening test to a patient 106 via a vision screening device 104 to determine the eye health of the patient 106. As described herein, the vision screening device 104 may perform one or more vision screening tests, including screening for eye diseases and / or abnormalities when visible light is illuminated on the eye. The vision screening device 104 may also be configured to perform other vision screening tests, such as a visual acuity test, a refractive error test, an accommodation test, a dynamic eye tracking test, a color vision screening test, and / or any other vision screening test, configured to evaluate and / or diagnose the visual health of the patient 106. In some examples, the vision screening device 104 may include a portable device configured to administer one or more vision screening tests. The portability of the vision screening device 104 allows vision screening tests to be administered anywhere, from traditional screening settings such as schools and clinics, to doctor's offices, hospitals, ophthalmology facilities, and / or other remote and / or mobile locations. It is also envisioned that the vision screening device 104 can be used to administer vision screening tests to all age groups, including newborns, young children, and geriatric patients.

[0016] As described herein, the vision screening device 104 may be configured to administer one or more vision screening tests to the patient 106. In various examples, the one or more vision screening tests may include illuminating the eye of the patient 106 with infrared or near-infrared (NIR) radiation and capturing radiation reflected from the eye of the patient 106. For example, U.S. Pat. No. 9,237,846, the disclosure of which is incorporated herein by reference in its entirety, describes a system and method for determining refractive error based on photorefraction using pupil images captured under different illumination patterns produced by a near-infrared (NIR) radiation source. In other examples, a vision screening test, such as a red reflex test, may include illuminating the eye of the patient 106 with visible light and capturing color images of the eye under visible light illumination. The vision screening device 104 may acquire data, including color image and / or video data of the eye under visible light illumination, to detect the pupil, retina, and / or lens of the eye of the patient 106. This data may be used to determine differences between the left and right eyes, compare captured images to standard images, or generate visualizations to assist the operator 102 or clinician in diagnosing ocular diseases or abnormalities in the patient. The vision screening device 104 may transmit the data via the network 108 to the vision screening system 110 for analysis to determine an output 112 related to the patient 106. Alternatively or additionally, the vision screening device 104 may perform some or all of the analysis locally to determine the output 112. Indeed, in any of the examples described herein, some or all of the disclosed methods may be performed, in whole or in part, by the vision screening device 104 independently (e.g., without the vision screening system 110 or its components), or by the vision screening system 110 independently (e.g., without the vision screening device 104 or its components).For example, in some examples, the vision screening device 104 may be configured to perform any of the vision screening tests and / or other methods described herein without being connected to or otherwise communicating with the vision screening system 110 via the network 108. In other examples, the vision screening system 110 may include one or more components that are similar and / or identical to those included in the vision screening device 104, and thus the vision screening system 110 may be configured to perform any of the vision screening tests and / or other methods described herein without being connected to or otherwise communicating with the vision screening device 104.

[0017] As shown generally in FIG. 1 , the vision screening device 104 may include one or more radiation sources 114 configured to perform functions related to administering one or more vision screening tests. The radiation sources 114 may include individual radiation emitters, such as light-emitting diodes (LEDs), arranged in a pattern to form an LED array. In some examples, the radiation sources 114 may include near-infrared (NIR) radiation emitters, such as NIR LEDs, to measure the refractive error of the patient's 106 eye using photorefraction. The NIR radiation emitters of the radiation sources 114 may be used to measure the gaze angle or gaze direction of the patient's 106 eye. Additionally, the radiation sources 114 may include color LEDs to generate color stimuli for display to the patient 106 during color vision screening tests.

[0018] The vision screening device 104 may include one or more radiation sensors 116, such as an infrared camera, configured to capture radiation reflected from the patient's eye during a vision screening test. For example, the vision screening device 104 may emit one or more radiation beams via a radiation source 114 and may be configured to direct such beams toward the eye of the patient 106. The vision screening device 104 may then capture corresponding radiation reflected back (e.g., from the eye's pupil) via the radiation sensor 116. In some examples, the radiation sensor 116 may include an NIR radiation sensor that captures reflected NIR radiation while the NIR radiation source 114 is shining light onto the patient's 106 eye. The data captured by the NIR radiation sensor 116 may be used to measure the refractive error and / or gaze angle of the patient's 106 eye. This data may include images and / or videos of the pupil, retina, and / or lens of the patient's 106 eye. In some examples, the images and / or video may be grayscale (e.g., values ​​between 0 and 128 or between 0 and 256). The data may be captured intermittently during a specific period of the vision screening test or throughout the entire test. The vision screening device 104 may also process the image and / or video data to determine changes in the refractive error and / or gaze angle of the patient's 106 eye. Grayscale images of the eye captured under NIR illumination may also be used to screen for eye diseases and abnormalities, such as refractive error, strabismus, and occlusion.

[0019] In some examples, the vision screening device 104 may further include a visible white light source 118 and a camera 120 configured to capture color images and / or videos of the eye under illumination by the white light source 118. The white light source 118 may include light-emitting diodes (LEDs), such as an array of LEDs configured to generate white light, e.g., blue LEDs with a phosphor coating that converts blue light to white light, or a combination of red, blue, and green LEDs configured to generate white light by varying the intensity at which the individual red, blue, and green LEDs are activated. The individual LEDs in the array of LEDs may be arranged in a pattern configured to operate individually to project light from different angles during a vision screening test. The white light source 118 may also be configured to generate white light at different intensity levels. The camera 120 may be configured to capture white light reflected from the patient's eye and generate digital color images and / or videos. The camera 120 may include a high-resolution, autofocus digital camera with custom optics for imaging the eye in clinical applications, as described in further detail with reference to FIG. 2. The color images and / or videos captured by the camera 120 may be stored in a variety of formats, such as JPEG, BITMAP, TIFF, etc. (for images) and MP4, MOV, WMV, AVI, etc. (for videos). In some examples, the pixel values ​​of the color images and / or videos may be in the RGB (red, green, blue) color space. Color images and / or videos of the eye captured under white light illumination can be used to screen for eye diseases and abnormalities, such as cataracts, intraocular fluid clouding in the aqueous and vitreous humor, tumors, retinal cancer, and retinal detachment. Color images and / or videos can also be used in combination with grayscale images captured under NIR illumination to generate visualizations to aid in the detection of a wide range of eye disease conditions.

[0020] The vision screening device 104 may include one or more display screens, such as display screen 122 and display screen 124, which may be color LCD (liquid crystal display) or OLED (organic light-emitting diode) display screens. Display screen 122 may be an operator display screen facing the operator 102 configured to provide information related to the vision screening test to the operator 102. In any of the examples described herein, the display screen 122 facing the operator 102 may be configured to display and / or provide output 112 generated by the vision screening device 104 and / or generated by the vision screening system 110. The output 112 may include test parameters, the current status and progress of the screening test, measurements determined during the test, images captured or generated during the screening test, a diagnosis and / or diagnosis-related recommendations based on one or more tests, and the like. The display screen 122 facing the operator 102 may also display patient-related or patient-specific information and the patient's medical history.

[0021] In some examples, the vision screening device 104 may include a display screen 124 facing the patient 106 and configured to display content to the patient 106. The content may include attention-grabbing images and / or videos to attract the patient's attention and keep the patient's gaze directed toward the vision screening device 104. Content corresponding to various vision screening tests may also be presented to the patient 106 on the display screen 124. For example, the display screen 124 may display color stimuli to the patient 106 during a color vision screening test or a Snellen chart during a vision screening test. The display screens 122, 124 may be integrated with the vision screening device 104 or may be external to the device and under the control of the device's 104 computer program.

[0022] The vision screening device 104 can transmit data captured by the radiation sensor 116 and the camera 120 over the network 108 using the network interface 126 of the vision screening device 104. Similarly, the vision screening device 104 can also transmit other test data related to the vision screening test being administered (e.g., test type, test time, patient identification information, etc.). The network interface 126 of the vision screening device 104 can be operatively connected to one or more processors 128 of the vision screening device 104 and can enable wired and / or wireless communication between the vision screening device 104 and one or more components of the vision screening system 110, as well as with one or more other remote systems and / or other networked devices. For example, the network interface 126 can include a personal area network component that enables communication over one or more short-range wireless communication channels and / or a wide area network component that enables communication over a wide area network. In any of the examples described herein, the network interface 126 can enable communication between, for example, the processor 128 of the vision screening device 104 and the vision screening system 110 over the network 108. The network 108 shown in Figure 1 can be any type of wireless network or other communication network known in the art. Examples of the network 108 include the Internet, an intranet, a wide area network (WAN), a local area network (LAN), a virtual private network (VPN), a cellular network connection, and connections made using protocols such as 802.11a, b, g, n, and / or ac.

[0023] The vision screening system 110 may be configured to receive data collected during the administration of a vision screening test from the vision screening device 104 via the network 108. In some examples, based at least in part on processing the data, the vision screening system 110 may determine an output 112 related to the patient 106. For example, the output 112 may include a recommendation and / or diagnosis related to the ocular health of the patient 106 based on an analysis of the color image data and / or NIR image data that is indicative of a disease and / or abnormality related to the patient's 106's eye. The vision screening system 110 may communicate the output 112 via the network 108 to the processor 128 of the vision screening device 104. As noted above, in any of the examples described herein, one or more such recommendations, diagnoses, or other outputs may alternatively or additionally be generated by the vision screening device 104.

[0024] As described herein, a processor, such as processor 128, can be a single processing unit or multiple processing units, and a processor can include single or multiple arithmetic units or multiple processing cores. Processor 128 can be implemented as one or more microprocessors, microcomputers, microcontrollers, digital signal processors, central processing units, state machines, logic circuits, and / or any device that manipulates signals based on operational instructions. For example, processor 128 can be one or more hardware processors and / or logic circuits of any suitable type specifically programmed or configured to execute the algorithms and processes described herein. As shown generally in FIG. 1, the vision screening device 104 can also include a computer-readable medium 130 operably connected to processor 128. Processor 128 can be configured to fetch and execute computer-readable instructions stored on computer-readable medium 130, thereby programming processor 128 to perform the functions described herein.

[0025] Computer-readable medium 130 may include volatile and nonvolatile memory and / or removable and non-removable media implemented in any type of technology for storing information, such as computer-readable instructions, data structures, program modules, or other data. Such computer-readable medium 130 includes, but is not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, optical storage, solid-state storage, magnetic tape, magnetic disk storage, RAID storage systems, storage arrays, network-attached storage, storage area networks, cloud storage, or any other medium that can be used to store desired information and that can be accessed by a computing device. Computer-readable medium 130 may be a type of computer-readable storage medium and / or may be a tangible, non-transitory medium, to the extent that, when referred to, non-transitory computer-readable medium excludes media such as energy, carrier signals, electromagnetic waves, and signals themselves.

[0026] The computer-readable medium 130 may be used to store any number of functional components executable by the processor(s) 128. In various examples, these functional components include instructions or programs executable by the processor(s) 128 that, when executed, specifically configure the one or more processors 128 to perform operations associated with one or more vision screening tests used in the detection and diagnosis of ocular diseases and abnormalities. For example, the computer-readable medium 130 may store one or more functional components for conducting a vision screening test, such as a patient screening component 132, an image capture control component 134, a data analysis and visualization component 136, and / or an output generation component 138, as shown in FIG. 1 . At least some of the functional components of the vision screening device 104 are described in detail below.

[0027] In various examples, the patient screening component 132 may be configured to store and / or access patient data 140 associated with the patient 106. For example, the patient data 140 may include demographic information such as name, age, and ethnicity. When the vision screening device 104 and / or the vision screening system 110 initiates a vision screening test, the patient 106 may provide the patient data 140 regarding the patient's demographics, medical information, preferences, and the like, or the operator 102 may request this information from the patient 106 or the patient's guardian. In such examples, the operator 102 may request the data during or prior to the start of the screening. In some examples, the operator 102 may be provided with a predetermined category associated with the patient 106, such as a predetermined age range (e.g., newborn to 6 months, 6 months to 12 months, 1 year to 5 years), or the operator 102 may request the patient data 140 to select the appropriate category associated with the patient 106. In other examples, the operator 102 may be provided with free-form input associated with the patient data 140. In yet another example, the input element may be provided directly to the patient 106 .

[0028] Alternatively or additionally, the vision screening device 104 and / or vision screening system 110 can determine and / or detect patient data 140 during a vision screening test. For example, the vision screening device 104 can include one or more digital cameras, motion sensors, proximity sensors, or other image capture devices configured to collect image and / or video data of the patient 106, and one or more processors of the vision screening device 104 can analyze this data to determine patient data 140, such as the age category of the patient 106 or the distance of the patient 106 from the screening device. For example, the vision screening device 104 can be equipped with a rangefinder, such as an ultrasonic rangefinder, an infrared rangefinder, and / or any other proximity sensor capable of determining the distance of the patient 106 from the screening device.

[0029] Alternatively or additionally, the vision screening device 104 may be configured to transmit image / video data to the vision screening system 110 via the network 108 for analysis to determine patient data 140. Furthermore, the patient screening component 132 may be configured to receive, access, and / or store patient data 140 associated with the patient 106 and / or additional patients. For example, the patient screening component 132 may store previous patient information associated with the patient 106 and / or other patients. For example, the patient screening component 132 may store the patient's 106 prior screening history, including data obtained during previous screenings, such as color images, NIR images, and / or videos of the patient's 106 eyes. The patient screening component 132 may receive the patient data 140 and / or access such information via the network 108. For example, the patient screening component 132 may access an external database, such as a screening database 144, that stores data associated with the patient 106 and / or other patients. The screening database 144 may be configured to store the patient data 140 in association with a patient ID. Once the operator 102 and / or patient 106 enters the patient ID, the patient screening component 132 may access or receive patient data 140 stored in association with the patient ID for the patient 106 .

[0030] In various examples, the patient screening component 132 may be configured to determine a vision screening test to administer to the patient 106 based at least in part on the patient data 140. For example, the patient screening component 132 may utilize the patient data 140 to determine the test category to which the patient 106 belongs (e.g., based on age, medical history, etc.). The patient screening component 132 may determine the vision screening test to administer based on the test category. For example, if the patient data 140 indicates that the patient is a newborn, the selected vision screening test may include screening for congenital eye conditions such as congenital cataracts, retinoblastoma, corneal opacities, and strabismus. Additionally, eye abnormalities may be associated with systemic genetic disorders such as Marfan syndrome and Tay-Sachs disease. For example, screening for a distinctive red spot in the eye may indicate Tay-Sachs disease. As another example, if the patient data 140 indicates that the patient is over 50 years old, the patient screening component 132 may determine that the vision screening exam includes screening for the development of cataracts, macular degeneration, and other age-related eye diseases.

[0031] The patient screening component 132 can also determine vision screening tests based on the patient's medical history. For example, the screening database 144 can store in the patient data 140 medical history related to the patient's 106 previous vision screening tests, including test results, eye images, measurements, recommendations, etc. The patient screening component 132 can access the patient data 140, including the medical history, from the screening database 144 to determine vision screening tests to administer to monitor the status and changes of previously detected visual health issues. For example, if a progressive eye disease, such as the development of cataracts or macular degeneration, is detected in a previous screening, further screening may be administered to track the development of the disease. As another example, if the patient 106 has undergone surgery to remove a tumor in the eye, the vision screening test may include screening for other tumors or scars in the eye. The patient screening component 132 can determine a list of vision screening tests to be administered to the patient 106 during a vision screening session and track the vision screening tests already administered during the vision screening session as well as the remaining vision screening tests on the list of vision screening tests to be administered.

[0032] In some examples, the computer-readable medium 130 can further store an image capture control component 134. The image capture control component 134 can be configured to operate the radiation source 114, radiation sensor 116, white light source 118, and camera 120 of the vision screening device 104 so that eye images are captured under specific lighting conditions required for each specific vision screening test. As described above, the radiation source 114 can include a near-infrared (NIR) LED for illuminating the eye during capture of grayscale images for measuring the refractive error and / or gaze angle of the patient's 106 eye, and the white light source 118 can include a white light LED for illuminating the eye during capture of color images of the eye by the camera 120. In some examples, the image capture control component 134 can generate commands to operate and control individual radiation sources, such as the LEDs of the white light source 118 and the LEDs of the NIR LEDs. Control parameters for the LEDs can include intensity, duration, pattern, and cycle time. For example, the commands may selectively activate or deactivate individual LEDs of radiation source 114 and white light source 118 to generate illumination from different angles as required by the vision screening test indicated by patient screening component 132. Image capture control component 134 may activate the NIR LEDs of radiation source 114 used to measure the refractive error and / or gaze angle of the patient's 106 eye in synchronization with the capture of an image of the eye by radiation sensor 116 during the performance of the vision screening test. Similarly, image capture control component 134 may activate the LEDs of white light source 118 in synchronization with the capture of a color image of the eye by camera 120.

[0033] Individual radiation sources, such as the LEDs of radiation source 114 or white light source 118, may be controlled by image capture control component 134 according to control parameters stored in computer-readable medium 130. For example, the control parameters may include the intensity, duration, pattern, cycle time, etc., of the NIR LEDs of radiation source 114 and / or the white light-producing LEDs of white light source 118. For example, image capture control component 134 may use the control parameters to determine the duration (e.g., 50 ms, 100 ms, 200 ms, etc.) for each individual LED of radiation source 114, 118 to emit radiation. Image capture control 134 may also utilize the control parameters to vary the intensity and display pattern of the NIR LEDs of radiation source 114 for determining ocular photorefraction and / or ocular refractive error based on gaze angle. With respect to intensity, the image capture control component 134 may control parameters to cause the LEDs of the white light source 118 to emit light at an intensity bright enough to capture color images of the eye using the camera 120, while also limiting the brightness to avoid or reduce pupil constriction or accommodation. The image capture control component 134 may also control the intensity of the white light source 118 to gradually increase in intensity at a steady rate while the camera 120 is activated to capture images and / or video of the eye to record the response of the pupil of the patient's eye to increasing illumination intensity.

[0034] Additionally, the image capture control component 134 can command the emission of radiation from the radiation source 114, the white light source 118 such that the NIR LED is activated before the LED of the white light source 118 is activated to capture an image of the eye under the NIR radiation. In some examples, this command can prevent the eye's pupil from constricting in response to the white light impinging on the eye and / or can capture an image of the eye's internal structure without having to dilate the patient's 106's pupil. In some examples, the image capture control component 134 can also control the radiation sources 114, 118 to generate patterns such as circular patterns, alternating light patterns, blinking patterns, patterns shaped like circles or rectangles, etc. to attract the patient's 106's attention, and / or control the colored LEDs of the radiation sources 114, 118 to display color stimuli, such as colored dot patterns, to the patient 106 during vision screening.

[0035] The image capture control component 134 can also control the radiation sensor 116 and the camera 120 to capture images and / or video of the patient's 106 eye during the performance of a vision screening test. For example, the radiation sensor 116 can capture data indicative of radiation reflected from the patient's 106 eye while one or more of the radiation sources 114 are operating. This data can include grayscale image data and / or video data of the eye. The image capture control component 134 can synchronize the operation of the camera 120 for capturing color image and / or video data of the eye with the operation of the white light source 118 so that the eye is illuminated with white light radiation when the color image and / or video data is captured. In some examples, images of the left and right eyes can be captured under different lighting conditions (e.g., from different individual light sources) so that the relative angle of illumination with respect to the optical axis of a particular eye is the same for the left and right eyes. In other examples, images of both eyes can be captured simultaneously under the same lighting. As described herein, the image capture control component 134 of the vision screening device 104 may generate grayscale images of the eye illuminated under NIR radiation and color images of the eye illuminated under white light. Capturing both grayscale and color images can enable the detection of a wider range of eye diseases and abnormalities.

[0036] In some examples, the computer-readable medium 130 may store a data analysis and visualization component 136. The data analysis and visualization component 136 may be configured to analyze image and / or video data collected, detected, and / or captured by components of the vision screening device 104 (e.g., the radiation sensor 116 and the camera 120) during one or more vision screening tests. For example, the data analysis and visualization component 136 may analyze the data to determine the location of the eye's pupil within the image and identify a portion of the image (e.g., a pupil image) that corresponds to the pupil. The data analysis and visualization component 136 may analyze the pupil image to determine visual characteristics of the pupil in the pupil image. For example, in the case of a color image captured by the camera 120, the features may include values ​​corresponding to the mean color, the variance of the color, a measure of uniformity, the presence of inclusions, etc. In the case of an infrared image captured by the radiation sensor 116, the features may include the mean grayscale value and the variance of the grayscale values ​​instead of color, in addition to the measure of uniformity and the presence of inclusions. The data analysis and visualization component 136 may further compare the left and right pupil images to determine visual differences between the left and right pupils. For example, the differences may correspond to differences in average color value, average grayscale value, or uniformity between the left and right pupil images. In a normal eye, the expected value of a characteristic associated with one pupil image (e.g., average color value, grayscale value, uniformity measure, etc.) may be approximately identical to the value of the characteristic in the other pupil image. The data analysis and visualization component 132 may compare the pupil image with standard pupil images and / or pupil images of the patient 106 captured during a previous vision screening to determine visual differences, such as differences in average color or grayscale values, differences in uniformity measures, or differences in detected inclusions. In such an example, the expected value of the characteristic in the pupil image may correspond to the value of the characteristic in the standard pupil image or the previously captured pupil image of the patient 106.In any of the above examples, all captured images or a subset of the captured grayscale and / or color images can be used to determine differences. In some examples, grayscale images may not be used, and differences may be determined based on color images. Note that pixels in grayscale images may also be considered to have color values, where the color values ​​are determined using the same grayscale value for each of the three color channels (e.g., RGB). For example, a pixel with a grayscale value of 128 may be determined to have a color value of (128, 128, 128) in the RGB color space. The data analysis and visualization component 136 may apply additional image processing steps to the grayscale and / or color images that may improve detection of disease states. For example, the image may be sharpened, certain colors may be enhanced or attenuated, the color or brightness of the image may be balanced, etc. Further details regarding the above-described data analysis for determining differences are described below with reference to FIG. 4.

[0037] Additionally, the data analysis and visualization component 136 may be configured to receive, access, and / or analyze normative data related to vision screening. For example, the data analysis and visualization component 136 may be configured to access or receive data from one or more additional databases (e.g., screening database 144, third-party databases, etc.) storing test data, measurements, and / or values ​​indicating various ranges or thresholds within which measurements should fall. Such thresholds or ranges may be known or determined from standard tests for patients with normal visual health. The data analysis and visualization component 136 may utilize the normative data for comparison with average values ​​and differences determined during vision screening tests, such as those described above. For example, the normative data may indicate a threshold or range for the difference between the color values ​​of the left and right pupil images, with differences greater than the threshold or outside the range corresponding to an abnormality in the patient's eye. Alternatively or additionally, the data analysis and visualization component 136 may access previous vision screenings of the patient 106 and compare the values ​​and differences to corresponding data from the previous screenings. For example, the average color value of the pupil may be compared to the average color value from the previous screenings to determine the difference. As described above, the difference can be compared to standard thresholds and ranges to determine the presence or absence of an abnormality. Different thresholds and / or ranges may be indicated in the standard data for different types of diseases or abnormalities. Additionally, the thresholds and / or ranges associated with a vision screening test may be based on the test category of the patient 106 (e.g., the age group or medical history of the patient 106), and different test categories may have different thresholds and / or ranges.The data analysis and visualization component 136 may store images and / or video captured or generated during the vision screening test, measurements related to the vision screening test, test results, and other data in a database (e.g., screening database 144) as part of the patient data 140 in order to compare the data over time to monitor visual health and changes in visual health. In some examples, the stored images may include images of the patient's 106 face or partial face (e.g., both eyes and part of the nose).

[0038] Based on the comparison with the above-mentioned thresholds and / or ranges, the data analysis and visualization component 136 can generate a normal / abnormal or pass / reference determination for each eye of the patient 106. For example, if all measured values ​​and differences are below the corresponding thresholds of the normative data or fall within the corresponding ranges of the normative data, a "normal" or "pass" determination may be made by the data analysis and visualization component 136; otherwise, an "abnormal" or "referral" determination may be made to indicate a referral for further screening. Alternatively or additionally, the data analysis and visualization component 136 may generate a normal / abnormal determination for each disease and / or abnormality screened for during the vision screening session.

[0039] In several examples, the data analysis and visualization component 136 can utilize one or more machine learning techniques to generate a diagnosis of a particular disease and / or abnormality type. For example, a machine learning (ML) model may be trained using normal images of the eye and eye images classified as indicative of various disease states and abnormalities. The trained ML model can then generate an output indicative of a disease or abnormality diagnosis when given as input an eye image captured during a vision screening of the patient 106. In such examples, the data analysis and visualization component 136 can directly generate the output by providing the eye image as input to the trained ML model without calculating differences between pupil images or applying comparisons to thresholds and / or ranges. In some examples, multiple trained ML models may be used, each trained to detect a particular disease or abnormality. In such examples, each ML model outputs a binary presence / absence indication indicating whether the input image exhibits the disease or abnormality that the ML model is trained to detect. The data analysis and visualization component 136 may provide the eye image as input to each ML model of multiple trained ML models for detecting one or more of a plurality of diseases and abnormalities. In some examples, the ML model may be a neural network, such as a convolutional neural network (CNN). In other examples, the ML model may include a regression algorithm, a decision tree algorithm, a Bayesian classification algorithm, a clustering algorithm, a support vector machine (SVM), etc.

[0040] The data analysis and visualization component 136 can also generate visualizations of the eye using the image and / or video data captured by the radiation sensor 116 and / or the camera 120. For example, a first visualization may include a composite image of the eye incorporating both grayscale information from a grayscale image and color information from a color image captured by the radiation sensor 116 under NIR illumination. Generating the first visualization may include detecting and identifying eye structures, such as the pupil and / or lens, followed by aligning these types of images so that the pupil is in the same location in the grayscale and color images. The composite image may then be generated using grayscale pixel values ​​from the grayscale image for one portion of the composite image and color pixel values ​​from the color image for another portion of the composite image. The portions of the composite image using grayscale pixel values ​​and the portions using color pixel values ​​may correspond to regions of different eye structures (e.g., the fovea, the retina, the cornea, etc.). The composite image may be used to more clearly depict eye structures for improved detection and evaluation of eye diseases and / or abnormalities.

[0041] In another example, the second visualization may include a series of still images or an animated video including a series of still images. In some examples, a series of images may be captured by the radiation sensor 116 or the camera 120 while the radiation source 114 or the white light source 118 illuminates the eye at different angles along different axes relative to the optical axis, including an angle of approximately 0° relative to the optical axis. For example, in some examples, the radiation emitted by the radiation source 114 or the white light source 118 may be emitted substantially parallel to and / or substantially along the optical axis. In such examples, the emitted radiation may be coaxial or nearly coaxial with the optical axis. In some examples, the visualization may include a graphic and / or color identification indicating areas of the image of the eye that are flagged as abnormal. Further examples of visualizations, including further details regarding the first and second visualizations, are described below with reference to FIGS. 5A and 5B. As described herein, the data analysis and visualization component 136 of the vision screening device 104 can process grayscale and color images of the eye captured during the administration of a vision screening test to determine diseases and / or abnormalities associated with the patient's eye. Based on the grayscale and color images, the data analysis and visualization component 136 can also generate visualizations of the eye that can assist a clinician or operator of the vision screening device 104 in identifying diseases and / or abnormalities in the eye.

[0042] The computer-readable medium 130 may further store an output generation component 138. The output generation component 138 may be configured to receive, access, and / or analyze data obtained from the data analysis and visualization component 136 to generate the output 112. For example, the output generation component 138 may utilize the normal / abnormal determinations of the data analysis and visualization component 136 to generate recommendations in the output 112. The recommendations may indicate whether the patient's 106 screening results indicate normal ocular health or whether further screening is necessary based on one or more screening tests determined to be “abnormal.” The output generation component 138 may also incorporate all or a subset of visualizations generated by the data analysis and visualization component 136 into the output 112 to aid in diagnosing the ocular condition. Portions of images and / or video captured by the radiation sensor 116 or camera 120 may also be included in the output 112. Furthermore, if an abnormality is determined, the output generation component 138 can incorporate an appropriate diagnosis into the output 112 based on the analysis results by the data analysis and visualization component 136. The output 112 can be presented to the device operator via the device interface (e.g., on the display screen 122 of the vision screening device 104). In some examples, the operator display screen can be invisible to the patient, for example, the operator display screen can face away from the patient. The output generation component 138 can store the output 112, which can include recommendations, diagnoses, measurements, captured images / video, and / or generated visualizations, in a database, such as a screening database 144, for evaluation by a clinician or for access during subsequent vision screening of the patient 106. The screening database 144 can provide access to authorized medical professionals to print reports related to the screening of the patient 106 or to further evaluate the data.

[0043] 1 illustrates an exemplary processor 128 and computer-readable medium 130 storing the patient screening component 132, image capture control component 134, data analysis and visualization component 136, output generation component 138, and / or other components and / or items as components of the vision screening device 104, in any of the examples described herein, the vision screening system 110 may include similar and / or identical components. In such examples, the vision screening system 110 may include a processor 146 and computer-readable memory 148 configured to perform the functions of some or all of the components of the computer-readable memory 130 of the vision screening device 104. For example, one or more of the components of the computer-readable memory 130 may be included in the analysis component 150 of the computer-readable memory 148 and may be executable by the processor 146. In such an example, the vision screening system 110 may communicate with the vision screening device 104 using the network interface 152 and via the network 108 to receive data from the vision screening device 104 and send results (e.g., output 112) back to the vision screening device 104. The vision screening system 110 may be implemented on a computer in close proximity to the vision screening device 104 or may be located at a remote location. For example, the vision screening system 110 may be implemented as a cloud service on a remote cloud server.

[0044] The network interfaces 152 may enable wired and / or wireless communication between the components and / or devices shown in the system 100 and / or with one or more other remote systems and other network-connected devices. For example, at least some of the network interfaces 152 may include a personal area network component to enable communication over one or more short-range wireless communication channels. Additionally, at least some of the network interfaces 152 may include a wide area network component to enable communication over a wide area network. Such network interfaces 152 may enable communication between the vision screening system 110 and the vision screening device 104 and / or other components of the system 100, for example, via the network 108. For example, the network interface 152 may be configured to connect to an external database (e.g., the screening database 144) to receive, access, and / or transmit screening data using a wireless connection. Wireless connections include cellular network connections as well as connections made using protocols such as 802.11a, b, g, and / or ac. In other examples, wireless connectivity can be achieved directly between the vision screening device 104 and an external system using one or more wireless protocols, such as Bluetooth, Wi-Fi Direct, radio frequency identification (RFID), infrared signals, and / or Zigbee. Other configurations are possible. Communication of data to an external database allows for a report of the patient's vision test data to be printed or further evaluated. For example, collected data and corresponding test results may be transmitted wirelessly and stored in a remote database accessible to authorized medical professionals.

[0045] While FIG. 1 illustrates system 100 as including a single vision screening system 110, it should be understood that in additional examples, system 100 may include any number of local or remote vision screening systems substantially similar to vision screening system 110, configured to operate independently and / or in combination, and configured to communicate via network 108.

[0046] 1 illustrates an exemplary vision screening device 104 that includes components for administering a vision screening test to a patient. In some examples, one or more components may be implemented in a remote vision screening system 110 that communicates with the vision screening device 104 over a network 108. The vision screening device 104 and its components are described in more detail with reference to the remaining figures.

[0047] 2 illustrates one embodiment of a visual screening device 200 according to some implementations. The exemplary visual screening device 200 may include one or more of the same components included in the visual screening device 104 of the system 100. In some additional examples, the visual screening device 200 may include different components that provide similar functionality as the visual screening device 104.

[0048] The vision screening device 200 may be a tablet-like device and may include one or more processors, computer-readable media, and associated network interfaces (not shown) within a housing 202. The housing 202 may include a front surface 204 configured to face a patient (such as patient 106) when the vision screening device 200 is in use, and a back surface 206 opposite the front surface 204 and configured to face an operator (such as operator 102) of the vision screening device 200 when the vision screening device 200 is in use. The front surface 204 may include a display screen 208, which may be substantially similar to or identical to the display screen 124, a radiation source 210, which may be substantially similar to or identical to the radiation source 114, a radiation sensor 212, which may be substantially similar to or identical to the radiation sensor 116, a white light source 214, which may be substantially similar to or identical to the white light source 118, and / or a camera 216, which may be substantially similar to or identical to the camera 120.

[0049] The radiation source 210 may be configured to emit radiation in the infrared and / or near-infrared (NIR) bands. For example, the radiation source 210 may include an arrangement of NIR LEDs configured to determine refractive error associated with one or more eyes of a patient. The NIR LEDs of the radiation source 210 may be arranged radially about a central axis 211 of the vision screening device 200, with the radiation sensor 212 positioned substantially along the central axis 211. By aligning the central axis 211 with the optical axis of the eye, the NIR LEDs can be used to provide off-center illumination of the patient's eye during a vision screening test (e.g., to measure refractive error via photorefraction techniques). The arrangement of the NIR LEDs is described in further detail with reference to FIG. 3B.

[0050] The vision screening device 200 may also include a white light source 214 and a visible light camera 216 configured to capture color images and / or video of the patient's eye. In some examples, the white light source 214 and the camera 216 may be included in an image capture module 218. The camera 216 of the image capture module 218 may include a high-resolution lens with a narrow field of view suitable for imaging the eye in a vision screening setting. Such a lens may incorporate folded prism slim lens technology, allowing for telephoto zoom while reducing height. The optics used in folded prism lenses bend and focus light by bouncing it back and forth within an optical prism, reducing the thickness of the lens and enabling a substantially low-profile form factor. As discussed above with reference to FIG. 1 , some vision screening tests may require color images of the pupil and / or lens of the patient's eye to determine the presence or absence of disease and / or abnormality. In some examples, the camera 216 may be equipped with high-resolution zoom capabilities that allow for the capture of close-up images of the patient's eye, from which the pupil and / or lens of the eye can be localized. In another example, the camera 216 may use a short-throw lens with eye placement adjusted to provide a focused image. The white light source 214, more commonly referred to as a flash, may include one or more visible light LEDs with adjustable intensity. The intensity level of the white light source 214 may be controlled by one or more processors of the vision screening device 200. The one or more processors of the vision screening device 200 may also synchronize the activation of the white light source 214 with the capture of images by the camera 216.

[0051] The vision screening device 200 may include a display screen 220 disposed on the rear surface 206 of the housing 202 that substantially faces an operator (e.g., operator 102) during operation of the vision screening device 200. The display screen 220 may be touch-sensitive to receive input from the operator and may display a graphical user interface configured to display information to and / or receive input from the operator during the vision screening test. For example, the operator may use the display screen 220 to input information about the patient or the vision screening test being performed. Additionally, the display screen 220 may be configured to display information to the operator regarding the vision screening test being performed (e.g., parameter settings, screening progress, options for transmitting data from the vision screening device 200, one or more measurements and / or images or visualizations generated during the vision screening, etc.). The display screens 208, 220 may include, for example, a liquid crystal display (LCD) or an active matrix organic light-emitting display (AMOLED).

[0052] In some examples, the vision screening device 200 may include handgrips 222a and 222b for holding the vision screening device 200 steady during a vision screening test. As described herein, FIG. 2 illustrates an exemplary vision screening device 200 including components for administering one or more vision screening tests to a patient. The vision screening device 200 is intended to perform an entire vision screening, which may include multiple different vision screening tests, including screening for multiple diseases, abnormalities, and conditions of the patient's eye. As illustrated, the vision screening device 200 has the additional feature of being lightweight enough to be handheld, for example, by using handgrips 222a and 222b for the operator's right and left hands, respectively, allowing for ease of use and portability for patients as young as newborns. The vision screening device 200 provides the radiation source and image capture sensor required for NIR imaging, as well as color imaging under white light illumination required for one or more vision screening tests, in a compact, substantially planar arrangement, allowing for the lightweight and portable form factor of the vision screening device 200.

[0053] 3A illustrates another embodiment of a vision screening device 300 according to some embodiments. The exemplary vision screening device 300 can include one or more of the same components included in the vision screening devices 104, 200. In some additional examples, the vision screening device 300 can include different components that provide similar functionality to the vision screening devices 104, 200.

[0054] In the illustrated example, the vision screening device 300 includes a housing 302 with a transparent display screen 304, such as a transparent organic light-emitting display (OLED), facing a first end 306 of the vision screening device 300, which first end 306 faces a patient (e.g., patient 106). The display screen 304 may cover optical components of the vision screening device 300, such as an array of LEDs 308 acting as a radiation source, which may be substantially similar to or identical to the radiation source 114, a radiation sensor 310, which may be substantially similar to or identical to the radiation sensor 116, and an image capture module 312 including a white light source 312a and a digital camera 312b, which may be substantially similar to or identical to the white light source 118 and camera 120. Although the white light source 312a is shown proximate the digital camera 312b, in some examples, the white light source 312a and / or additional white light sources may be located elsewhere on the housing 302 (e.g., the white light source 312a and / or additional white light sources may be located at one or more corners 315 of the housing 302, along or proximate one or more sides or edges of the housing 302, and / or in any other location). The display screen 304 is transparent so that radiation from the radiation source 308 and / or white light from the white light source 312a of the image capture module 312 can reach the patient's eye, and reflected radiation from the patient's eye can be received by the radiation sensor 310 and / or camera 312b of the image capture module 312 by passing through the display screen 304 without attenuation or redirection. The array 308 may be composed of individual NIR LEDs (e.g., NIR LEDs 308a, 308b, 308c, 308d) distributed in a pattern around the radiation sensor 310, as shown. As also shown, the NIR LEDs 308a-308d may be arranged along different axes, such as axes A-A', B-B', and C-C', which will be described in more detail with reference to Figure 3B. The radiation sensor 310 is arranged substantially along a central axis 314 of the vision screening device 300, similar to the central axis 211 of the vision screening device 200.Although the individual NIR LEDs of the array 308 are shown here radiating outward from a centrally located radiation sensor 310, other arrangements of the NIR LEDs of the array 308, including more or fewer individual NIR LEDs, are contemplated. As discussed with reference to Figure 2, the arrangement of the NIR LEDs of the array 308 described herein can provide the off-center illumination necessary to measure refractive error using photorefraction techniques.

[0055] FIG. 3B shows an expanded view of an array 308 of NIR LEDs. In various examples, the array 308 may include more or fewer individual LEDs than those illustrated. The exemplary arrangement of the individual NIR LEDs in the array 308 shown in FIG. 3B includes a column 316 of NIR LEDs extending generally coaxially across the array 308 along a first axis M1 of the array 308, which may correspond to axis A-A' in FIG. 3A. The array 308 may also include a column 318 along a second axis M2, which may correspond to axis B-B' in FIG. 3A, at an angle θ relative to the column 316, as shown, and a column 320 along a third axis M3, which may correspond to axis C-C', at an angle α relative to the column 318. The angles θ and α may be any acute angle (e.g., 60°). In some examples, axis M1 may also be referred to as the first meridian of array 308, axis M2 may also be referred to as the second meridian of array 308, and axis M3 may also be referred to as the third meridian of array 308, and the three axes M1, M2, and M3 may intersect at a central axis 314 of vision screening device 300. In some examples, additional LEDs may be positioned along one or more of axes M1, M2, and M3 spaced apart from array 308.

[0056] As described in further detail with reference to FIG. 3C , the individual LEDs of array 308 may be activated sequentially (e.g., by image capture control component 134) to generate illumination light at different angles or eccentricities. For example, LED 322 may be activated first, followed by LED 324 adjacent to LED 322, and then LED 326, progressing along axis M2. The activation sequence may move the individual LEDs along a first axis M1, followed by a second axis M2, and then a third axis M3. LEDs 322 and 324 may also be activated substantially simultaneously to simulate a source position of the combined radiation using a diffuser (not shown). In such an example, the amount of current applied to LEDs 322 and 324 may be further controlled (e.g., by image capture control component 134) to achieve a desired simulated source position of the combined radiation and enable illumination light to be generated at additional angles or eccentricities without having to mechanically move array 308.

[0057] FIG. 3C is a schematic diagram of an exemplary vision screening system 301 using a vision screening device 300 to administer a vision screening test to a patient 106 by an operator 102, according to examples of the present disclosure. In examples, the central axis 314 of the vision screening device 300 may be substantially aligned or collinear with the optical axis 328 of the patient's eye, as shown. As described above with reference to FIG. 1, the image capture control component 134 of the vision screening device may control individual radiation sources, such as LEDs 308a-308d or 322, 324 of the array 308, to emit radiation. The radiation emitted by the individual LEDs may strike the eye of the patient 106 at different angles relative to the optical axis 328. For example, the radiation beam 330A emitted by LED 308b may be at an angle 332A relative to the optical axis 328, and the radiation beam 330B emitted by LED 308c may be at an angle 332B different from angle 332A. Thus, activation of the LEDs in the array 308 can be utilized individually or collectively to generate radiation that strikes the eye of the patient 106 at different angles. In some examples, the angles 332A, 332B may be approximately 0 degrees, for example, relative to the optical axis 328 (e.g., the illumination may be coaxial or nearly coaxial with the central axis 314). For each angle of illumination, radiation reflected from the eye of the patient 106 traveling along the optical axis 328 can be captured by a radiation sensor 310 located along the central axis 314 aligned with the optical axis 328 to generate an image of the eye under illumination from each angle relative to the optical axis 328. Although described herein with reference to the array 308 of NIR LEDs, illumination from different angles relative to the optical axis 328 can also be generated by the white light source 118, 214, 312a using an array or set of individual white light LEDs arranged in a two-dimensional array or linear pattern, similar to the NIR LEDs in the array 308 described above, as described further below with reference to FIG. 3D .

[0058] The vision screening device 300 may include an additional display screen 334 disposed on the housing 302 of the vision screening device 300 on a side 336 opposite the front side 306. Similar to the display screen 122 described with reference to FIG. 1, the display screen 334 may face the operator 102 and be configured to provide information regarding the vision screening test to the operator 102. In some examples, the display screen 334 may be separate from the vision screening device 300 (e.g., not attached to the housing 302), but operably connected to and under the control of the vision screening device 300.

[0059] 3D illustrates an exemplary system 303 including components of the vision screening system 301 according to several examples of the present disclosure. The exemplary system 303 illustrates a camera 338, a white light LED array 340 including the white light source 118, a diffuser 342, and a partial reflector 344. For clarity, the exemplary system 303 omits other components of the vision screening system 301, but it is understood that, for example, any of the components of the vision screening device 300 or the components of the vision screening system 301 described above may be included in the exemplary system 303 illustrated in FIG. 3D.

[0060] In some examples, radiation 346 (e.g., light) emitted by one or more LEDs of the LED array 340 passes through a diffuser 342 and impinges on a partial reflector 344. In some examples, the partial reflector 344 may be a beam splitter, an array of mirrors, a prism, or some other optical component configured to reflect a first portion of the impinging radiation while transmitting a second portion of the radiation. In some examples, the partial reflector 344 is positioned at an angle of approximately 45 degrees with respect to a central axis 314 of the vision screening system 301, 303, which may be substantially aligned or coaxial with the optical axis of the eye of the patient 106, as described with reference to FIG. 3C . As shown, the central axis 314 may also be substantially aligned or coaxial with the lens of the camera 338. The diffuser 342 may act as a blur-smoothing filter for the radiation 346 emitted by the LEDs of the LED array 340. In some examples, the lens 348 may be configured to focus the radiation 346 emitted by the LED array 340 onto the partial reflector 344. However, one or more additional optical components may be included to modify the radiation 346 reaching the partial reflector 344. In the example shown in FIG. 3D , at least a portion 350 of the radiation 346 may be reflected off the partial reflector 344 and directed toward one or both eyes of the patient 106. While the portion 350 of the radiation 346 is directed toward the eye of the patient 106, the camera 338 may capture one or more images and / or videos of the eye of the patient 106. In some examples, the images and / or videos may be indicative of the radiation reflecting off the pupil of the eye of the patient 106.

[0061] In various examples, as described herein with reference to Figures 2 and 3A-3D, the vision screening device 200, 300 may include an NIR radiation source and sensor for capturing NIR images of the patient's eye, and a white light source and color camera for capturing color images of the patient's eye. The vision screening device 200, 300 may also capture images of the eye while under illumination from the radiation source at different angles relative to the optical axis of the eye. While the radiation sources 114, 210, 308 have been described as including infrared or near-infrared (NIR) sources, in additional examples, the radiation sources 114, 210, 308 may include LEDs that emit radiation at different wavelengths, and the radiation sensors 116, 212, 310 may capture images of the eye while illuminated with radiation of different wavelengths and / or different wavelength bands (e.g., infrared, NIR, visible, ultraviolet, etc.). Different wavelengths of radiation in the visible spectrum may include wavelengths corresponding to specific colors. In such instances, the vision screening device 104, 200, 300 may be able to detect eye diseases and / or abnormalities that may be more apparent in images captured under illumination of particular wavelengths. Additionally, because colored light and white light are also forms of electromagnetic radiation, the term "radiation source," as used herein, may refer to both visible light emitters and radiation emitters in the infrared / NIR and ultraviolet regions of the electromagnetic spectrum.

[0062] 4A-4D illustrate images of an eye captured by a radiation sensor 116, 212, 310 or camera 120, 216, 312b of a vision screening device 104, 200, 300. Various ocular abnormalities and / or diseases that may be detected using analysis of image data captured by the vision screening device 104, 200, or 300 will now be described with reference to FIGS. 4A-4D. FIG. 4A illustrates an image 402 of a patient's eye with normal ocular health and no detectable disease states or abnormalities. The image 402 includes the patient's right eye 404a and left eye 404b. As shown, the iris 406a and pupil 408a of the right eye 404a appear substantially similar to the corresponding iris 406b and pupil 408b of the patient's left eye 404b, which are indicative of normal ocular health. As described with reference to FIG. 1 , the data analysis and visualization component 136 can process the captured images to determine the location of the eye's pupil and generate an image of the pupil (e.g., a pupil image). Because the pupil allows radiation to enter the eye and for reflected radiation to interact with different layers of the eye before returning out of the eye, the pupil image captures the appearance of eye layers, such as the cornea, lens, aqueous and vitreous humors, and retina, that are illuminated by the radiation striking the eye. U.S. Patent Application No. 17 / 347,079, filed June 14, 2021, the entire disclosure of which is incorporated herein by reference, describes exemplary systems and methods for detecting pupil images captured under different illumination patterns generated by a near-infrared (NIR) source to determine refractive error based on photorefraction.

[0063] FIG. 4B illustrates an exemplary image 410 of a disease state that may be detected by comparing a pupil image 412a of one eye 414a with a pupil image 412b of the other eye 414b. Image 410 may be a grayscale image captured by radiation sensors 116, 212, 310 under NIR illumination and shows an example of a grayscale difference between the left and right eye images resulting from clouding of the intraocular fluid or lens, typical of cataract development. As described with reference to FIG. 1, the data analysis and visualization component 136 can compare the pupil images of the left and right eyes to determine the difference in grayscale values ​​between the two eyes. For example, the grayscale values ​​of image 410 may range from 0 to 128, and the average grayscale value of the pupil portion of one eye's image may be 24, while that of the other eye may be 80. The calculated difference can be compared to thresholds and / or ranges in standard test data corresponding to normal eyes to determine whether an abnormality exists. However, grayscale images such as image 410 may not capture differences between the eyes that correspond to some diseases or abnormalities that are easily discernible in color images. For example, a tumor in the retina or cornea of ​​the eye may appear as a uniform gray area similar in appearance to a normal retina in a grayscale image captured under NIR illumination.

[0064] FIG. 4C shows a color image 416 captured by a camera 120, 216, 312b of a vision screening device 104, 200, 300 under white light illumination (e.g., from a white light source 118, 214, 312a). Because pupil images 420a, 420b in image 416 are generated from white light reflected from the eye's retina and through the cornea, retinal and corneal diseases and abnormalities can be visualized in such images. For example, because the retina is highly vascular, reflected light appears orange-red in a healthy eye, but may appear white or yellow in an eye with a retinal or corneal tumor. While differences in retinal pigmentation among patients of different ethnicities may result in variations in color appearance in pupil images 420a, 420b, comparing two pupil images 420a, 420b from the same patient reliably reveals differences in color values ​​when a disease or abnormality is present in only one of the two eyes. As described with reference to FIG. 1, the data analysis and visualization component 136 can compare pupil images of the left and right eyes to determine differences in color values. The data analysis and visualization component 136 can also compare the color values ​​of each pupil image with a standard image of a pupil of a healthy eye. The image 416, such as that captured by the camera 120, 216, or 312b, may be a typical digital color image in which each pixel represents an RGB (red, green, blue) value ranging from 0 to 256 for each of three color channels. As is known in the art, the RGB color space is often unsuitable for applications requiring the determination of differences between multiple colors because it is sensitive to lighting variations and has a low correlation between the distance between colors in the RGB color space and the perceived difference between colors. In some examples, the color image 416 can be converted to a color space (e.g., CIE L*a*b*, CIE L*u*v*, CIE 1931 model, HSI (hue, saturation, luminance), etc.) that is more suitable for determining differences between colors. The difference between the average color values ​​of pupil image 420a and the average color values ​​of pupil image 420b can be determined in the transformed color space. In some examples, a measurement of ocular refractive error may be used to further adjust for differences in color values ​​between the patient's eyes to eliminate the effects of ocular refractive error, which may also cause differences in pupil appearance.Difference values ​​that are greater than a threshold and / or outside the range of typical data corresponding to a normal, healthy eye can be flagged as detected anomalies. For example, the color difference shown in image 416 may have resulted from a tumor, such as retinoblastoma, in one eye.

[0065] FIG. 4D further illustrates an image 422 of eyes 424a, 424b, including pupil images 426a, 426b. Image 422 may be a grayscale image captured by radiation sensor 116, 212, 310 under NIR illumination or a color image captured by camera 120, 216, 312b under white light illumination. Even if the images of pupils 426a and 426b have similar average grayscale or color values, as shown, there may be other types of differences between them, such as inhomogeneities, inclusions, or other structures, which may indicate disease states and / or abnormalities. For example, small inhomogeneities or inclusions may indicate the early stages of cataract formation, the presence of foreign bodies in the intraocular fluid, scratches on the cornea or lens, etc. The data analysis and visualization component 136 can determine these types of differences between pupil images 426a, 426b by various methods. For example, after aligning pixel images 426a and 426b, subtracting the pixel values ​​of pixels at corresponding locations would result in a difference image that shows the areas where the differences primarily exist. The sum of the pixel values ​​in the difference image can be compared to a threshold to determine whether the differences are greater than the threshold (e.g., an abnormal case). The data analysis and visualization component 136 can also determine differences by calculating the variance of grayscale or color values ​​in pupil image 426b. If the variance exceeds a threshold or falls outside the range expected for a healthy eye, an abnormal condition may be determined.

[0066] In some examples, the data analysis and visualization component 136 can determine certain eye conditions by evaluating each individually captured pupil image for uniformity of characteristics within the pupil image. These characteristics may include color, brightness, texture, etc. For example, the data analysis and visualization component 136 may determine the standard deviation (or variance) of the characteristics within the pupil image, and an abnormal condition may be determined if the standard deviation exceeds a threshold or falls outside of a range expected for a healthy eye.

[0067] 4A-4D illustrate examples of some eye conditions that may be determined by the techniques described herein, it should be understood that additional conditions may also be determined. Additionally, the data analysis and visualization component 136 may analyze color images under white light illumination, grayscale images under NIR illumination, and / or composite images (as described with reference to FIG. 1) to determine an eye condition. For example, the presence or absence of a cataract in the eye may be determined based on composite images, whereas the presence or absence of blastoma may be determined primarily based on color images. In some examples, color and / or grayscale images may be extracted from color and / or grayscale video of the eye, e.g., one or more frames of video.

[0068] 4A-4D illustrate processing of images captured by radiation sensors 116, 212, 310 and / or cameras 120, 216, 312b that may be performed by the data analysis and visualization component 136 of the vision screening device 104 to determine differences between pupil images that are indicative of disease states and / or abnormalities in the patient's eye. Other examples of processing tailored to detect specific disease states and abnormalities are also contemplated. For example, images captured under different wavelengths of radiation may be used to detect characteristic differences in grayscale or color values ​​or structure of the images that are indicative of specific disease states.

[0069] 5A and 5B illustrate exemplary visualizations of pupil images that may be generated by the vision screening device 104, 200, or 300. As described with reference to FIG. 1, the data analysis and visualization component 136 may process the captured pupil images to generate one or more visualizations that may assist a clinician or vision screening device operator in diagnosing an ocular abnormality or disease in a patient (e.g., patient 106). As described above, a first visualization may generate a composite image 502 that incorporates information obtained from grayscale images captured by the radiation sensors 116, 212, 310 under NIR illumination and information obtained from color images captured by the cameras 120, 216, 312b under white light illumination. The data analysis and visualization component 136 may align the grayscale and color images, e.g., by performing an image registration process, so that the pupils 504 of the eyes in the images are superimposed on one another. In this way, the data analysis and visualization component 136 can perform a more precise image registration process based on specific features of the eye, such as the optic disc 506 and / or the fovea 508, to ensure that the grayscale and color images are accurately aligned, for example, when the features of the eye overlap one another. In some examples, the registration process may be performed even when the vision screening device is not moving during multiple image captures, since slight patient movement or patient eye movement can cause misalignment between images.

[0070] In some examples, a composite image 502 can be generated from the aligned grayscale and color images, e.g., by the data analysis and visualization component 136, such that the composite image 502 includes pixel values ​​from the grayscale image in a portion 510 of the pupil region 504 and pixel values ​​from the color image in the remaining portion. The selection of images to capture in different portions of the composite image 502 can be based on whether features in those portions are more discernible under NIR illumination or white light illumination. For example, the portion 510 of the eye may include features or conditions that are better discerned under NIR illumination, while vascular structures in the remaining portion of the pupil image 504 may be more clearly visible in a color image captured under white light illumination. In such an example, the data analysis and visualization component 136 can generate the composite image 502 using grayscale values ​​from the grayscale image in the region 510 of the composite image 502 and color values ​​from the color image in the remaining portion of the pupil image 504. In an example where multiple grayscale and color images are captured, the data analysis and visualization component 136 can generate a composite grayscale image and a composite color image using the average grayscale value at each pixel location in the grayscale image and the average color value at each pixel location in the color image. In other examples, a single image from the multiple images may be selected for use in the composite image 502 based on factors such as image quality (e.g., sharpness, angle of illumination, or visibility of particular ocular features). As described above, the composite image derives a first plurality of pixel values ​​from a grayscale image captured under NIR illumination and a second plurality of pixel values ​​from a color image captured under white light illumination. Thus, the composite image includes characteristics of the eye revealed separately by NIR and visible light illumination.

[0071] In some examples, the composite image may include portions of the patient's face (e.g., nose, forehead) or even the entire face in addition to the eyes. In some examples, the data analysis and visualization component 136 may also or alternatively add graphics 512 (e.g., pseudocolor) to the composite image 502 to highlight portions or areas of the composite image 502 where differences are detected between the left and right pupil images or between the captured image and the standard image. In some examples, the colors used in the highlighting (e.g., in the graphic 512) may be based on a heat map visualization scheme, where cooler colors or bluer hues may indicate smaller differences and warmer colors or redder hues may indicate larger differences. Portions of the composite image 502 where differences exceed a standard threshold or fall outside the standard range may be assigned pseudocolors from the top end of the heat map. Other features of the eye may also be highlighted using different graphics or different color legends. In such examples, the clinician or operator may have the option to turn the highlighting and / or graphics on, off, or switch between them to aid in diagnosing the eye condition. The composite image 502 may be stored as part of the patient data 140 in a database (eg, the screening database 144).

[0072] FIG. 5B illustrates a visualization 514 that includes a sequence of still images or an animated video including a sequence of still images. This visualization may correspond to the second visualization described with reference to the data analysis and visualization component 136 of FIG. 1. The radiation sources 516(1-7) may correspond to, for example, NIR LEDs along the axes or meridians or 320 of FIG. 3B. Alternatively or additionally, the radiation sources 516(1-7) may include an array of white light LEDs of the white light source 118, 214, 312a. The pupil images 518(1-7) represent images captured by the radiation sensors 116, 212, 310 or the cameras 120, 216, 312b under illumination from the corresponding radiation sources 516(1-7). For example, image 518(1) may be captured when radiation source 516(1) is activated, image 518(3) may be captured when radiation source 516(3) is activated, etc. Note that sequence 514 need not include an image corresponding to each radiation source (e.g., 516(2, 4, 6)). In various examples, any number of images corresponding to radiation sources 516(1-7) may be used to generate visualization 514. Also, more or fewer radiation sources 516 are contemplated.

[0073] As described above with reference to FIG. 5A , images captured under illumination from individual radiation sources (e.g., 516(1, 3, 5, 7)) are aligned so that the pupil is in the same position throughout each image. This alignment or registration prevents jitter when the images are presented sequentially (e.g., on display screen 122, 220). Visualization 514 may include presenting images 518(1-7) sequentially on display screen 122, 220 from left to right (e.g., as a series of images 518(1), 518(3), 518(5), 518(7)) and / or from right to left (e.g., as a series of images 518(7), 518(5), 518(3), 518(1)). Data analysis and visualization component 136 can also generate animated video in which each frame of video contains a single image in the sequence. The animated video may include repeatedly displaying a left-to-right sequence followed by a right-to-left sequence, creating the appearance of an illumination source moving back and forth from end to end. The vision screening device may present a graphical user interface (e.g., on the display screen 122, 220) that allows the clinician or operator of the vision screening device to pause the video at any frame and / or zoom in or out. Additionally, the visualization 514 may use, for each image 518 (1, 3, 5, 7), a grayscale image captured under NIR illumination, a color image captured under white light illumination, or a composite image as described above with reference to FIG. 5A. The graphical user interface may provide options for displaying a grayscale image, a color image, or a composite image.

[0074] 5A and 5B, the vision screening device 300 may provide visualizations to an operator of the vision screening device 300 to aid in diagnosing a patient's ocular disease and / or abnormality. Additionally, these visualizations may be stored as part of the patient data 140 (e.g., in the screening database 144) so ​​that the visualizations can be accessed by a clinician for review or comparison during future vision screening tests for the same patient.

[0075] FIGS. 6-8 illustrate flow diagrams illustrating exemplary methods for visual screening as described herein. The methods illustrated in FIGS. 6-8 are illustrated as a collection of blocks in a logical flow graph, which represent a sequence of operations that can be implemented in hardware, software, or a combination thereof. In the software context, the blocks represent computer-executable instructions stored on one or more computer-readable storage media that, when executed by a processor, perform the described operations. Generally, computer-executable instructions include routines, programs, objects, components, data structures, etc. that perform particular functions or implement particular abstract data types. The order in which the operations are described is not intended to be construed as a limitation, and any number of the described blocks can be combined in any order and / or in parallel to implement the methods illustrated in FIGS. 6-8. In some embodiments, one or more blocks of the methods illustrated in FIGS. 6-8 may be omitted entirely.

[0076] The operations described below with respect to the method illustrated in Figures 6-8 may be performed by any of the devices or systems 104, 200, and 300 described herein and / or various components thereof. Unless otherwise specified, for ease of explanation, the method illustrated in Figures 6-8 will be described below with reference to the system 100 illustrated in Figure 1 and the vision screening devices 104, 200, and 300 illustrated in Figures 1-3. In particular, any of the operations described with respect to the method illustrated in Figures 6-8 may be performed, alone or in combination, by the image capture control component 134, data analysis and visualization component 136, and / or output generation component 138 executed by the processor 128 of the vision screening device 104 and / or the analysis component 150 executed by the processor 146 of the vision screening system 110.

[0077] 6, in operation 602, the image capture control component 134 and / or one or more associated processors may cause a radiation source to emit radiation (e.g., near-infrared (NIR) radiation). For example, the radiation source may include an NIR LED of the radiation source 114 of the vision screening device 104 configured to emit NIR radiation during a time period corresponding at least in part to the administration of a vision screening test administered to the patient by the patient screening component 132. In some examples, the image capture control component 134 may emit radiation of different wavelengths, e.g., a first radiation source may emit radiation of a first wavelength and a second radiation source may emit radiation of a second wavelength. As described with reference to FIG. 3B, the image capture control component 134 may activate the LEDs of the radiation source 114 individually or in groups to generate radiation that strikes the eye at different angles relative to the eye's optical axis. For example, the image capture control component 134 can set the activation pattern of the NIR LEDs along axes 316, 318, and 320, as described above with reference to FIG. 3B . The image capture control component 134 can also activate different wavelengths of radiation in combination with different angles of incidence on the eye. In some examples, arranging the activation pattern of the NIR LEDs can present different illumination patterns to the eye of the patient 106, allowing for accurate measurement of the refractive error of the eye based on images captured under the selected illumination patterns. Additional details regarding illumination patterns used in testing protocols for determining refractive error can be found in the above-mentioned U.S. Patent Application No. 9,237,846, which is incorporated herein by reference.

[0078] In operation 604, the image capture control component 134 can cause a sensor of the vision screening device (e.g., the radiation sensor 116 of the vision screening device 104) to capture radiation reflected from the patient's eye under illumination by the radiation source 114. The image capture control component 134 can receive data indicative of the radiation captured by the radiation sensor 116. This data can include grayscale images and / or videos of the eye illuminated by radiation from different angles, as described above in operation 602. For example, the image capture control component 134 can cause the sensor to capture a first image under illumination from a first set of NIR LEDs and a second image under illumination from a second set of NIR LEDs. In some examples, near-infrared radiation can be a first wavelength band emitted by the first radiation source, and the image capture control component 134 can further activate a second radiation source emitting radiation in a second wavelength band and cause the sensor to capture a third image under illumination from the second radiation source. Additionally, the image capture control component 134 can cause the sensor to capture images of both eyes simultaneously or one eye at a time. For example, the image capture control component 134 can alter the activation of the illumination source (e.g., activate a different LED in the radiation sources 114, 118) after capturing an image of the left eye and before capturing an image of the right eye so that both eyes are illuminated from the same angle relative to the eye during image capture. In some examples, the image capture control component 134 can cause the sensor to capture multiple images of the eye while the patient 106 is instructed to look in various directions, such as to the left, right, up, and / or down relative to the optical axis of the vision screening device 104.

[0079] In operation 606, the image capture control component 134 and / or one or more associated processors may cause a white light source (e.g., the white light source 118 of the vision screening device 104) to emit white light to illuminate the patient for a period of time after operations 602 and 604 are completed and during at least a portion of the administration of the vision screening test. Similar to the radiation sources described in operation 602, the individual white light sources of the white light source 118 may also be operated to illuminate the eye from different angles relative to the optical axis. In some examples, the illumination from the white light source 118 may be coaxial or nearly coaxial with the optical axis, e.g., the angle may be substantially 0°. In some examples, multiple images of the eye may be captured corresponding to different gaze directions of the patient 106, as described above, e.g., a first color image of the eye may be captured corresponding to a first gaze direction of the patient, and a second color image of the eye may be captured corresponding to a second gaze direction of the patient. The image capture control component 134 can store the capture time and the illumination angle and / or patient gaze direction at the time of image capture as metadata associated with the image.

[0080] In operation 608, the image capture control component 134 can cause a camera (e.g., the camera 120 of the vision screening device 104) to capture a color image of the patient's eye while under white light illumination. In some examples, the image capture control component 134 can also cause the camera to capture video data. For example, the video data can be captured for a first period of time before the onset of white light illumination and continue for a second period of time after the onset of white light illumination. For example, the video data can be useful for determining the patient's pupillary response (e.g., pupil size) and / or pupillary adjustment to various levels of illumination and / or sudden changes in illumination (e.g., caused by the onset of white light illumination). The image capture control component 134 can also store the color image and / or video in a database for review by a clinician. The image capture control component 134 can also cause the camera to capture a color image of the patient's face and store the image in the patient data 140 as a photo identifier for the patient. The data analysis and visualization component 136 can utilize the color images to generate and differentiate a composite color image at operations 610 and 612, as described below. In some examples, color images captured at different gaze directions of the patient may be combined to generate a composite color image to show the retina of the eye.

[0081] At operation 610, the data analysis and visualization component 136 may generate a composite image of the eye by combining information from the grayscale image captured at operation 604 and information from the color image captured at operation 608. As described above with reference to FIG. 5A , the data analysis and visualization component 136 may detect a pupil image corresponding to the eye's pupil in the grayscale and color images and align the grayscale and color pupil images so that the eye structures overlap. The data analysis and visualization component 136 may further generate a composite image that incorporates grayscale values ​​from the grayscale image into portions of the composite image and color values ​​from the color image into the remainder of the composite image. The data analysis and visualization component 136 may also annotate portions of the composite image (e.g., using graphics and / or false color values), for example, to indicate areas of interest.

[0082] In operation 612, the data analysis and visualization component 136 or the output generation component 138 can determine one or more differences between pixel values ​​in the grayscale image, color image, and / or composite image and expected pixel values. For example, the data analysis and visualization component 136 or the output generation component 138 can calculate a first difference as the average difference between pixel values ​​at corresponding pixel locations in the patient's left and right pupil images. As another example, the data analysis and visualization component 136 or the output generation component 138 can calculate a second difference between average pixel values ​​in a first region of the pupil image and a second region of the same pupil image. In yet another example, the data analysis and visualization component 136 or the output generation component 138 can calculate a third difference between the average pixel value of the pupil image and a standard value obtained for a normal, healthy eye, which may be stored in a database (e.g., screening database 144). Additionally, the data analysis and visualization component 136 or the output generation component 138 may calculate a fourth difference between pixel values ​​in a pupil image captured during a vision screening test and pixel values ​​in a pupil image during a previous vision screening test performed on the same patient.

[0083] At operation 614, the output generation component 138 can compare the pixel value differences obtained at operation 612 to thresholds and / or ranges to determine an output indicative of an eye-related condition, which may include a diagnosis or recommendation. For example, if the difference is below the threshold (operation 614—Yes), the output generation component 138 can generate a first output related to the patient at operation 616, and if the difference is equal to or greater than the threshold (operation 614—No), the output generation component 138 can generate a second output at operation 618. The thresholds and / or ranges may be predetermined or may be utilized as part of standard data, which may be stored in the screening database 144 or the computer-readable medium 130, 148. The standard data may include different thresholds and ranges for each of the above-mentioned types of differences, including separate thresholds and ranges for grayscale values ​​and color values. The thresholds and ranges may also vary based on the patient's examination category (e.g., the patient's age group or medical history).

[0084] At operation 616, the output generation component 138 may generate a first output as described above (operation 614—Yes). The first output may correspond to an indication that a disease or abnormality has been detected, a recommendation for further screening, and / or a diagnosis of the detected disease or abnormality. The first output may include a link to stored images, including captured grayscale and color images and / or generated visualizations. At operation 618, the output generation component 138 may generate a second output (operation 614—No). The second output may correspond to an indication that the patient passed the vision screening or that the patient's eyes appear normal and healthy.

[0085] As described above, the exemplary method 600 may be performed by components of the vision screening device 104 executed by the processor 128 of the device 104. The exemplary method 600 illustrates operations performed during at least a portion of a vision screening test administered to a patient (e.g., the patient 106) to determine diseases and / or abnormalities associated with the patient's eye based on images of the eye captured under illumination of different wavelengths. In another example, some or all of the operations of the method 600 may be performed by the processor 146 of the vision screening system 110 connected to the vision screening device 104 via the network 108.

[0086] 7 illustrates an exemplary method 700 for vision screening according to some embodiments of the present disclosure. As noted above, the operations of process 700 are described as being performed by processor 128 of vision screening device 104, even though these operations may alternatively or additionally be performed by processor 146 of remote vision screening system 110.

[0087] In operation 702, the data analysis and visualization component 136 can determine a pupil image from the captured image. As described above, the captured image may include a grayscale image captured by the radiation sensor 116, 212, 310 under NIR illumination or a color image captured by the camera 120, 216, 312b under white light illumination. The pupil image may also be determined from the image captured under NIR radiation illumination using techniques described in U.S. Pat. No. 9,237,846, incorporated herein by reference. The data analysis and visualization component 136 can also determine the pupil image from the color image using various techniques, such as, for example, detecting edges using image processing techniques, followed by arc fitting and comparing the detected edge arcs to a model edge map of the eye image. The pupil color may also be used to determine the pupil image from the color image and segment the pupil region of the eye image. A combination of edge and color-based segmentation may also be used.

[0088] In operation 704, the data analysis and visualization component 136 can determine differences between the patient's left and right pupil images. These differences can be determined using grayscale values ​​in a grayscale image and / or color values ​​in a color image. As described above with reference to operation 612 in FIG. 6 , the differences can be calculated as the average difference between pixel values ​​at corresponding pixel locations or portions of the patient's left and right pupil images. In other examples, the data analysis and visualization component 136 can determine the differences by subtracting the left pupil image from the right pupil image, or vice versa, and then summing the resulting pixel values.

[0089] In operation 706, the data analysis and visualization component 136 may compare the difference obtained in operation 704 to a first threshold to determine whether the difference is less than the first threshold. For example, the first threshold may be predetermined, available as part of normative data that may be stored in the screening database 144, and may indicate the maximum difference expected between two pupil images of the same patient when the patient presents with a normal, healthy eye. If the difference is greater than or equal to the first threshold (operation 706—Yes), then the output generation component 138 may generate an output reporting the abnormality in operation 716, as described in more detail below.

[0090] In operation 708 (operation 706-No), the data analysis and visualization component 136 can determine the difference between the patient's left or right pupil image and a standard image of a normal, healthy eye. The left and right pupil images may be captured simultaneously or at different times during the vision screening test. The standard image may be available as part of the standard data, which may be stored in the screening database 144 or computer-readable media 130, 148.

[0091] At operation 710, the data analysis and visualization component 136 may compare the difference obtained at operation 708 with a second threshold to determine that the difference is less than the second threshold. For example, the second threshold may also be predetermined, available as part of the normative data stored in the screening database 144, and may indicate the maximum difference expected between the pupil image and a normative image of a normal, healthy eye. If the difference is greater than or equal to the second threshold (operation 710—Yes), then the output generation component 138 may generate an output reporting the abnormality at operation 716, as described in more detail below.

[0092] In operation 712 (operation 710-No), the data analysis and visualization component 136 may determine differences between the patient's pupil image and a pupil image of the same eye captured during a previous vision screening test. The patient's pupil image captured during a previous vision screening test (e.g., a previous annual screening test) may be stored in a database (e.g., screening database 144) as part of the patient data 140. The data analysis and visualization component 136 may access the pupil image from the previous screening test from the database and / or may load the image onto the computer-readable medium 130 prior to the start of the current vision screening test.

[0093] In operation 714 (operation 712—No), the data analysis and visualization component 136 may compare the difference obtained in operation 712 to a third threshold and determine that the difference is less than the third threshold. For example, the third threshold may also be predetermined, available as part of normative data that may be stored in the screening database 144, and may indicate the maximum difference expected between pupil images of the same eye captured after a period of time. If the difference is greater than or equal to the third threshold (operation 714—Yes), the output generation component 138 may generate an output reporting an abnormality in operation 716. For example, the output may indicate a change in the patient's eye, which may be due to a progressive eye disease such as cataracts or macular degeneration, a new disease state not present in previous screening tests, etc.

[0094] At operation 716, the output generation component 138 may generate an output reporting the abnormality. As described above, if any of the differences determined in operations 704, 708, and 712 are equal to or greater than their respective thresholds, the data analysis and visualization component 136 or the output generation component 138 may determine that an ocular abnormality and / or disease state may exist. The output may include a diagnosis based on the differences that triggered the reporting of the abnormality. As described with reference to FIG. 4, certain diseases may exhibit characteristic differences in the appearance of the pupil image, and some diseases, such as retinoblastoma, cataracts, and corneal scars, may be identified from differences detected during a vision screening test. The output generation component 138 may store the output in a database, such as the screening database 144, as part of the patient data 140. The output generation component 138 may also display the output to an operator (e.g., operator 102) on a display screen (e.g., display screen 122, 220, 334).

[0095] If no differences meet or exceed the respective thresholds, then in operation 718, the output generation component 138 may generate an output reporting a normal screening, such as, for example, that the patient's eyes were determined to be normal in the vision screening test. The output generation component 138 may store the output in a database, such as the screening database 144, as part of the patient data 140 and / or display the output to an operator of the vision screening device (e.g., operator 102) on the display screen 122, 220, 334.

[0096] 8 illustrates an exemplary method 800 for vision screening according to some embodiments of the present disclosure, where the vision screening includes one or more separate vision screening tests, such as screening for different eye conditions. Various operations of method 800 may be substantially similar or identical to those described with reference to FIGS. 6 and 7. As noted above, the operations of process 800 are described as being performed by processor 128 of vision screening device 104, even though these operations may alternatively or additionally be performed by processor 146 of remote vision screening system 110.

[0097] In operation 802, the patient screening component 132 may select a vision screening test to administer to a patient participating in a vision screening session. As described with reference to FIG. 1, the patient screening component 132 may determine a list of vision screening tests to administer to the patient based at least in part on the patient's test category stored in the patient data 140 (e.g., the patient's age group or medical history). The vision screening test selected in operation 802 may be the next uncompleted vision screening test in the list of vision screening tests to be administered to the patient. In some examples, in 802, the patient screening component 132 may select the vision screening test based on input received from a clinician or operator administering the screening test.

[0098] In operation 804, the data analysis and visualization component 136 may receive images of the patient's eye undergoing vision screening from the image capture control component 134. For example, these images may include one or more of grayscale images captured under NIR radiation illumination, color images captured under white light illumination, color or grayscale video, and / or composite images. The images may be captured as described with reference to the exemplary method 600 shown in FIG. 6. In some examples, the images may have been previously captured, such as during administration of a previously selected vision screening test, and no new images may be captured.

[0099] In operation 806, the data analysis and visualization component 136 may perform the selected vision screening test, such as by using the method 700 described with reference to FIG. 7 . Each vision screening test may have a set of predetermined image requirements associated with it that indicate the images best suited to detecting the condition being screened for. For example, the image requirements may include the type of image (color, grayscale, synthetic, video, etc.), the lighting angle or gaze direction, the lighting type / level, etc. The data analysis and visualization component 136 may identify images to analyze based on the image requirements, such as by matching the image requirements to metadata associated with the captured images, and perform the selected vision screening test based on an analysis of the identified images. For example, a screening test for the presence of ocular cataracts may require synthetic images, while a screening test for the presence of ocular cataracts may require color images as input. In some examples, images captured at different lighting angles or gaze directions may be extracted from the video, such as by identifying frames associated with the required lighting angle or gaze direction.

[0100] At operation 808, the patient screening component 132 may determine whether the vision screening session is complete, e.g., whether the vision screening test performed at operation 806 was last on the list of vision screening tests determined by the patient screening component 132. If the patient screening component 132 determines at 808 that the vision screening session is not complete (operation 808—No), the patient screening component 132 may proceed to operation 802 to select the next vision screening test to perform. On the other hand, if the patient screening component 132 determines at 808 that the vision screening session is complete (operation 808—Yes), the output generation component 138 may generate a report at operation 810 that includes the results of the vision screening tests performed during the vision screening session. For example, the report may include the output generated at operations 716, 718 of method 700 for each vision screening test performed during the vision screening session.

[0101] Based at least on the description herein, it will be understood that the vision screening device and associated systems and methods of the present disclosure can be used to assist in conducting one or more vision screening tests, including tests for screening a patient's eye for eye diseases and / or abnormalities. The components of the vision screening device described herein may be configured to generate radiation of different wavelengths in addition to white light, illuminate the eye of a patient undergoing vision screening, capture images of the eye under different lighting conditions, generate visualizations that aid in diagnosing a disease state, determine differences between pupil images, and determine an output indicative of a diagnosis, recommendation, or screening test result. An exemplary vision screening device may include a radiation source for generating radiation of different wavelengths, a sensor for capturing radiation reflected from the patient's eye, a white light source, a camera configured to capture color images of the patient's eye under white light illumination, and a display screen for displaying the output to an operator of the vision screening device. Because the devices described herein can be used to screen patients for eye diseases and abnormalities without requiring input or feedback from the patient and without the need for eye magnification, the devices can be used to screen very young, very elderly, incapacitated, or uncooperative patients.

[0102] The foregoing is merely illustrative of the principles of the present disclosure, and various modifications may be made by those skilled in the art without departing from the scope of the present disclosure. The foregoing examples are presented for purposes of illustration and not limitation. Furthermore, the present disclosure may take many forms other than those expressly described herein. It is therefore emphasized that the present disclosure is not limited to the methods, systems, and apparatus expressly disclosed, but is intended to cover such variations and modifications as come within the spirit and scope of the following claims.

[0103] As yet another example, it is possible to vary the device or process limitations (e.g., dimensions, configuration, components, order of process steps, etc.) to further optimize the provided structures, devices, and methods as shown and described herein. Regardless, the structures and devices and related methods described herein have many applications. Thus, the disclosed subject matter should not be limited to any single example described herein, but rather its breadth and scope should be construed according to the appended claims.

Claims

1. 1. A visual screening device comprising: a radiation source configured to emit radiation at a first wavelength; a sensor configured to capture radiation reflected from the patient's eye; A white light source; a camera configured to capture a color image of the eye of the patient; a processor operatively connected to the radiation source, the sensor, the white light source, and the camera; A memory, When executed by the processor, the processor: causing the radiation source to emit radiation at the first wavelength for a first period of time; causing the sensor to capture a portion of the radiation reflected from the eye of the patient during the first time period; causing the white light source to illuminate the eye of the patient for a second time period after the first time period; causing the camera to capture a color image of the eye of the patient during the second time period; generating a composite image of the eye based at least in part on the grayscale image indicative of the captured portion of the radiation and the color image; determining a difference between a value associated with the eye and an expected value based on the composite image; generating an output indicative of a condition related to the eye based at least in part on the difference; the memory storing the a display unit disposed on a first side of the visual screening device and configured to display the output to an operator of the visual screening device; the radiation source, the sensor, the white light source, and the camera are disposed on a second side of the vision screening device opposite the first side; the white light source includes an array of light emitting diodes (LEDs) configured to direct light onto the patient's eye from a first angle and a second angle different from the first angle relative to an optical axis associated with the patient's eye; the composite image includes a first plurality of pixels representing the grayscale image and a second plurality of pixels representing the color image; Vision screening device.

2. 10. The vision screening device of claim 1, wherein the first wavelength is in the near-infrared (NIR) band of the electromagnetic spectrum.

3. 3. The vision screening device of claim 2, wherein the radiation source comprises an array of NIR light emitting diodes (LEDs) arranged so that the NIR LEDs have a common axis.

4. The instructions to the processor further include: causing the white light source to illuminate the eye of the patient from the first angle during a first portion of the second time period; causing the camera to capture a first color image of the eye of the patient during the first portion of the second time period; causing the white light source to illuminate the eye of the patient from the second angle during a second portion of the second time period; causing the camera to capture a second color image of the eye of the patient during the second portion of the second time period; generating a series of animated images of the eye based on the first color image and the second color image; 2. The visual screening device of claim 1, wherein the sequence of animated images is displayed on the display unit.

5. The instructions to the processor further include: causing the camera to capture a first color image of the patient's eye corresponding to a first gaze direction of the patient; causing the camera to capture a second color image of the patient's eye corresponding to a second gaze direction of the patient; 10. The vision screening device of claim 1, wherein an image of the retina of the eye is generated based on the first color image and the second color image.

6. applying a light to the patient's eye for a first period of time; capturing a grayscale image of the eye during the first time period; applying light to the eye for a second period of time distinct from the first period of time; capturing a color image of the eye during the second time period; generating a composite image of the eye based at least in part on the grayscale image and the color image, deriving a first plurality of pixel values ​​from the grayscale image and a second plurality of pixel values ​​from the color image; determining a difference between a color value associated with the eye and an expected color value based on at least one of the color image or the composite image; the eye is a first eye of the patient, and the expected color value is one of a second color value associated with a second eye of the patient or a third color value obtained from standard data associated with a human eye; generating an output related to the patient based at least in part on the difference.

7. 7. The method of claim 6, wherein the directing light for a first period of time comprises directing radiation to the eye in the near-infrared (NIR) band of the electromagnetic spectrum.

8. determining a first region of the grayscale image corresponding to a pupil of the eye; determining a second region of the color image corresponding to the pupil of the eye; The method of claim 6 , wherein the difference is based at least in part on the first region or the second region.

9. the color value is a first color value; The method comprises: obtaining patient data associated with the patient; determining the second color value as the expected color value; The method of claim 6 , wherein the patient data includes a second color value associated with the eye.

10. providing the grayscale image or the color image as an input to a trained machine learning model; The method of claim 6 , further comprising receiving the output from the trained machine learning model.

11. determining that the difference is greater than or equal to a threshold value; The method of claim 6 , wherein the output is based at least in part on a determination that the difference is greater than or equal to the threshold value.

12. determining portions of the composite image where the difference is greater than or equal to the threshold; The method of claim 11 , further comprising generating a graphic illustrating the portion of the composite image, wherein the output includes the composite image having the graphic.

13. providing the composite image having the graphic via a graphical user interface (GUI); The method of claim 12 , wherein the GUI allows a user to turn the graphics on or off.

14. 7. The method of claim 6, wherein the output indicates at least one of a normal eye screening, required additional screening, or a disease associated with the eye.

15. Memory and a processor; computer-executable instructions stored in the memory and executable by the processor, causing the radiation source to emit near-infrared (NIR) radiation for a first period of time; causing a sensor to capture a portion of the NIR radiation reflected by the patient's eye during the first period of time; directing a white light source at the eye for a second time period separate from the first time period; causing a camera to capture a color image of the eye during the second time period; determining a difference between the eye-related value and an expected value based on the color image and the portion of the NIR radiation; determining that the difference is equal to or greater than a threshold value; generating an output indicative of the eye condition based at least in part on determining that the difference is greater than or equal to the threshold; causing the white light source to illuminate the eye of the patient from a plurality of angles during the second time period; causing the camera to capture a plurality of color images of the eye of the patient, each color image corresponding to an angle of the plurality of angles; generating an image based on the plurality of color images, wherein each frame of the image corresponds to a color image among the plurality of color images; and the instructions for performing operations including displaying the video on a display screen.

16. The operation is generating a grayscale image indicative of the captured portion of the NIR radiation; generating a composite image based on the grayscale image and the color image; The system of claim 15 , wherein the composite image includes a first plurality of pixel values ​​from the grayscale image and a second plurality of pixel values ​​from the color image.

Images