GB2496005A

GB2496005A - Method for detecting strabismus in images of the eyes

Info

Publication number: GB2496005A
Application number: GB1212093.7A
Authority: GB
Inventors: Ron Uriel Maor; Nigel Andrew Simon Barnard; Yuval Yashiv
Original assignee: IRISS MEDICAL TECHNOLOGIES Ltd
Current assignee: IRISS MEDICAL TECHNOLOGIES Ltd
Priority date: 2012-07-06
Filing date: 2012-07-06
Publication date: 2013-05-01
Anticipated expiration: 2032-07-06
Also published as: GB2496005B; GB201212093D0

Abstract

A method for detecting strabismus in at least one image comprising a subject's eyes, wherein the image has been obtained using an apparatus comprising a light source 102 and a target 104 offset from the light source, the method comprising, for each of the subject's eyes: determining from the image a reflection offset distance, being the distance between a reference point of the eye 302 and a reflection of the light source on the eye 220 (see figure 3); determining a difference between the reflection offset distance and a reference reflection offset distance, wherein the reference reflection offset distance has been determined based on relative positions of the light source, the target and the subject's head; and determining whether the difference is less than a difference threshold value for determining whether the subject has a strabismus. The reference point of the eye 302 may be the pupil centre. The target may have a device to get the user's attention.

Description

STRABISMUS DETECTION

Technical Field

The invention relates to a method, apparatus and system for detecting strabismus in an image of a subject's eyes.

Background

Detection of strabismus in subjects is typically performed manually. Under such circumstances, a trained clinician assesses a subject's eyes using the Hirschberg test", which is a well-established, but rough, test for checking if the pupils of the eyes are symmetrical when in focus and fixating on a target placed directly in front of them.

Broadly speaking, the Hirschberg test is carried out by shining a light at the eyes of a subject from a light source positioned directly in front of the eyes, getting the patient to focus on the light or an object right next to it, and seeing whether the distance between the reflection of the light and the centre of the pupil appears to be equal in magnitude and direction in both eyes.

Small differences in each of the eyes of a subject are difficult to determine manually and are therefore easy to miss, even by experienced clinicians. Typically, differences of anything less than about 1 mm are too small to see reliably. Difficulties with the test are exacerbated when the subject is a small child, as it is hard to get a child to fixate and focus on a target for long enough to assess each eye. Therefore, the decision of whether the eyes are symmetrical is often made quickly and without complete information. It is also a test that requires considerable skill, and so is impractical for many testing scenarios, such as testing all pre-school children in a given territory.

Summary

According to the invention in a first aspect, there is provided a method for detecting strabismus in at least one image comprising a subject's eyes. The at least one image has been obtained using an apparatus comprising a light source and a target offset from the light source. For each of the subject's left and right eyes, a reflection offset distance, being the distance between a reference point of the eye and a reflection of a light source on the eye, is determined. A difference between the reflection offset distance and a reference reflection offset distance for the corresponding left or right eye is determined. The reference reflection offset distance has been determined based on relative positions of the light source, the target and the subject's head. It is determined whether the difference is less than a difference threshold value for determining whether the subject has a strabismus.

Optionally, the image has been obtained using an apparatus in which the light source and the target are positioned in a device plane perpendicular to an axis defining the direction that the subject is facing.

Optionally, the image has been obtained using an apparatus in which one of the light source and the target is positioned in front of the subject on the axis defining the direction that the subject is facing, such that the target, the light source and the subject's head form a right angled triangle.

Optionally, the image has been obtained using an apparatus in which the light source is positioned on the axis defining the direction that the subject is facing.

Optionally, the method further comprises obtaining the image.

Optionally, the target comprises an attention grabber, and obtaining the image comprises operating the attention grabber to attract the subject's attention such that the subject's eyes fixate on the target.

Optionally, the image is obtained while the attention grabber is operating.

Optionally, the attention grabber comprises one or more of alight, a flashing light and a means for generating a sound.

Optionally, the method further comprises instructing the subject to fixate their eyes on the target.

Optionally, the reference reflection offset distance has been determined based on a distance from the light source to the target and a distance from the light source or the target to the subjects head.

Optionally, the reference reflection offset distance has been determined based on the distance from the device plane to the subject's head along the axis defining the direction that the subject is facing.

Optionally, the method further comprises determining the reference reflection offset distance.

Optionally, the reference point of the eye comprises the centre of the pupil of the eye.

Optionally, the method further comprises determining that no strabisrrius is present in the image of the subject's eyes if the difference is less than the difference threshold.

Optionally, the at least one image comprises a plurality of images, and each of the plurality of images has been obtained with the target in a different location relative to the light source.

Optionally, the plurality of images has been obtained with the target at a plurality of angularly spaced locations relative to the light source.

Optionally, the plurality of angularly spaced locations are equally angularly spaced.

Optionally, the method further comprises determining whether the number of images in which the difference is less than the difference threshold value is greater than an image threshold value.

Optionally, the method further comprises, if the number of images is greater then the image threshold value, determining that the subject does not have a concomitant squint.

Optionally, the difference threshold value is a lower threshold value, and the method further comprises, if the difference is greater than the lower threshold value, determining whether the difference is greater than an upper threshold value.

Optionally, the method further comprises determining that the subject is not looking at the target if the difference is greater then the upper threshold.

According to the invention in a second aspect, there is provided a computer program product comprising computer program code configured, when run on a computer, to undertake the method described herein.

According to the invention in a third aspect, there is provided an apparatus comprising a processor configured to carry out the method described herein.

Optionally, the apparatus further comprises a camera and a light source, each in electrical communication with the processor, wherein the camera is configured to obtain an image while the light source is emitting light.

According to the invention in a fourth aspect, there is provided a system for detecting strabismus in an image comprising a subject's eyes, the system comprising: a camera; a light source; a target offset from the light source; and a processor in electrical communication with the camera and the light source, wherein the camera is configured to capture an image while the light source is emitting light, and wherein the processor is configured to carry out the method described herein.

According to the invention in a fifth aspect, there is provided a kit of parts comprising: an apparatus as described herein; and a target.

According to the invention in a sixth aspect, there is provided a device for detecting strabismus in an image comprising a subject's eyes, the image obtained using an apparatus comprising a light source and a target offset from the light source, the device comprising: an input interface configured to receive image data from a camera; a memory for storing image data received from the camera; and a processor configured to process the image data, for each of the subject's left and right eyes, by determining, in the image data, a reflection offset distance, being the distance between a reference point of the eye and a reflection of a light source on the eye, determining a difference between the reflection offset distance and a reference reflection offset distance for the corresponding left or right eye, wherein the reference reflection offset distance has been determined based on a relative position of the light source, the target and the subject's head, and determining whether the difference is less than a difference threshold value for determining whether the subject has a strabismus.

Optionally, the apparatus further comprises a camera configured to obtain the image data and transmit the image data to the input interface.

Optionally, the apparatus further comprises a light source, wherein the camera is configured to obtain the image data while the light source is emitting light.

Optionally, the apparatus further comprises a display, configured to display to a user information relating to whether the subject has a strabismus.

Brief description of the drawings

Exemplary embodiments of the invention will now be described with reference to the accompanying drawings, in which: Figure 1A is a schematic representation of a system for detecting strabismus in a subject; Figure 1 B is a block schematic representation of a system for detecting strabismus in a subject; Figure 2 is a schematic of a layout of a system for detecting strabismus in a subject; Figure 3 is an enlarged schematic of the left eye of Figure 2; Figure 4 is a flow chart showing a method for detecting strabismus in a subject; and Figure 5 shows a system for detecting strabismus in a subject.

Description

Generally, disclosed herein is a method, apparatus and system for detecting strabismus in an image of a subject's eyes, the image being captured by a camera using a light source and a target that is offset from the light source. The offset target provides a focal point for a subject, who may be instructed (or attracted by a light and/or a sound) to look at the target when an image is captured. Through knowledge of the relative position of the subject, the light source and the target, reference data may be determined that may be compared with data determined from the captured image for detecting strabismus.

During a manual Hirschberg test, a subject focuses on a light source located directly to their front and a clinician assesses the relative separation between the reflection of the light source and the centre of the pupil in each eye. This process may be automated by capturing an image and performing image processing techniques. The inventors have appreciated that a problem exists with this type of strabismus detection if a subject relaxes their eyes during the test. Subjects, children in particular, normally do not fixate and focus on a given location for long periods so, when the test is undertaken, their eyes are often relaxed and "staring into the middle distance". If a Hirschberg test is undertaken in an automated system, without a skilled clinician, this causes a problem as the automated system is unable to distinguish between eyes focused and fixated on a correct location and eyes that are relaxed. When a subject's accommodation (focus) is relaxed, a strabismus that would otherwise be evident may not be seen. Additionally, when a subject's eyes do not fixate, they fall to a position looking directly to the front. Therefore, using a typical Hirschberg test, if a subject relaxes their eyes, e.g. when they are day dreaming", they may appear to be looking directly ahead and to have no strabismus, which leads to an increased number of false negatives resulting from the test.

Figure 1A shows schematically a system 100 for detecting strabismus. The system comprises a light source 102 and a target 104. The light source 102 may, for example be a flash used in a camera assembly. In the exemplary system 100 of Figure 1A, the light source 102 is directly in front of a subject's head 106. That is, the subject's head 106 faces the light source 102 to define an axis 108. The axis 108 extends through the centre of the subject's head in a forward direction from the face of the subject and passes through the centre of the light source 102.

The target 104 is offset from the light source 102 in a direction perpendicular to the axis 108 by a distance 110. In exemplary systems, the target may be offset from the light source by a distance in the range from 5cm to 10 cm.

The inventors have appreciated that the problem defined above may be solved if the subject focuses on the offset target during the test. In the system 100 of Figure 1 A, the subject's head 106 remains aligned with the axis 108, while the subject's eyes are rotated to a line of sight shown by dashed line 112. In this way, the problem of relaxation of a subject's eyes is overcome as relaxed eyes always fall to a position looking directly ahead, which will be immediately obvious during the test. That is, if the subject begins to daydream, their eyes will relax to look along the axis 108.

The system also includes a camera (114 in Figure 1B), which may be collocated with the light source 102, and a processor (122 in Figure 1B). The camera may alternatively be located at other positions in other exemplary systems. In addition, it is noted that in other exemplary systems, the subject's head may face a target and the eyes may be rotated to focus on a light source. Indeed, in alternative systems, a subject's head, a light source, a target and a camera may have any relative positions so long as those relative positions are known or measured and may be defined by one or more distances and/or angles.

A block schematic of the system 100 is shown in Figure 1 B. The light source 102, the target 104 and a camera 114 are in electrical communication with a device 116. More specifically, the light source 102, the target 104 and a camera 114 are in electrical communication with an input interface 118 of the device 116. The target 104 is shown in Figure 1 B in electrical communication with the device 116 via the input interface 118 as the target may transmit and receive data to and from the device 116. For example, the target 104 may provide an attention grabber, such as a flashing light or image or a sound, to encourage a subject to focus on the target, which attention grabber may be controlled by the device 116. In addition, the target 104 may include one or more sensors configured to collect data and transmit that data to the device 116. However, it is noted that the target 104 is not required to be in electrical communication with the device 116 and can be any object offset from the light source 102.

The device 116 also comprises a memory 120 for storing data received from the light source 102, the target 104 or the camera 114. To this end, the memory 120 is in electrical communication with the input interface 118.

The device 116 also comprises a processor 122 for processing data stored in the memory 120. In particular, the processor 122 is configured to process image data received from the camera 114 and stored in the memory 120. The operation of the processor 122 in processing the image data is discussed in greater detail below.

The device also comprises a display 124 for displaying information to a user of the device 116. The display 124 may, for example, be configured to display the results of a test for determining whether a subject has a strabismus as set out below. In certain embodiments, the display 124 may also be configured to receive inputs from a user via a touch surface. Other devices may allow input by a user through other means, such as a keyboard.

The electrical communications are shown as physical links in Figure 1 B, but it is noted that the electrical communications may be provided by other means, such as a wireless communications system or a network.

In the exemplary system described herein, the light source 102 is collocated with the camera 114 in a single device 116. In addition, the processor 122 is collocated with the light source 102 and the camera 114 in the same device 116. The device 116 may be a camera, a smart-phone, a FDA, a laptop, ophthalmoscope or any other device capable of capturing images and comprising a processor capability. However, it is noted that, in other exemplary systems, each of the light source 102, the camera 114 and the processor 122 may be located in separate devices. In a particular exemplary system, the light source 102 and the camera 114 may be included in a single device, further configured to store captured images. The captured images may then be transferred to a separate device comprising a processor, which is configured to process the captured images as disclosed herein.

Figure 2 shows an exemplary system 100 for detecting strabismus. The system comprises a device 200 comprising the camera 114, the light source 102 and the processor 122. The target 104 is offset from the device 200 by the distance, dir, 110.

The device 200 and the target 104 are each located in a device plane defined by the dashed line 202, which is perpendiculai to the axis 108 and parallel to a face plane defined by dashed line 204. The face plane 204 is the plane defined by the face of the subject, shown schematically in Figure 2 to be across the front of a right eye 206 and a left eye 208 of a subject.

The right eye 206 and the left eye 208 are separated by a distance, d093, 210. The device 200, and more specifically the light source 102 and the camera 114, are centred between the right eye 206 and the left eye 208 and positioned a distance, ddf, 212 away from the plane 204. In practice, the distance, ddf, 212 may not be defined as accurately as depicted in Figure 2 and may, for example, be measured simply as the distance from the light source to the subject's head. However, the distance, ddf, 212 is typically in the range from 0.5 metres to 3 metres and so inaccuracies in the measurement of the distance, ddf, 212 will not impact greatly on the determination of the reference reflection offset (described below). In other exemplary embodiments, the distance ddt 212 may be measured by the camera 114.

The light source 102 is configured to emit light toward the eyes 206, 208 along lines 214 and 216. This results in light source reflections seen at the positions 218, 220 on the eyes 206, 208 where the emitted light 214, 216 contacts the eyes 206, 208. The light source is reflected from the cornea of the eyes 206, 208.

The eyes 206, 208 view the target 104 using the fovea 222, 224, the central part of the macula at the back of the eyes 206, 208. The fovea 222, 224 view the target 104 through the pupils of the eye (shown in Figure 3) along a lines of sight 226, 228.

Figure 3 shows a close up schematic view of the left eye of Figure 2. Corresponding features of Figure 3 that also appear in Figure 2 are given the same reference numerals.

An angle, kappa, 300 is defined between the line of sight 228 and a line taken from the fovea 224 and the centre of the pupil 302. The angle kappa is a known angle within the field of ophthalmology and is therefore known to the skilled person. Typically, the angle kappa is approximately three degrees for a majority of subjects. In exemplary apparatuses and methods disclosed herein, the angle kappa may be assumed to be three degrees, or some other average value of kappa for a particular population. In other exemplary apparatuses and methods, the angle kappa may have been measured previously for a subject and that measured value may be used. In addition, an angle, alpha, 304 may be defined at a centre 306 of the eye 208 between a notional mid-point 308 of the eye 208 and the light source reflection point 220. Alpha 304 is also shown in Figure 2 between the axis 108 and the line of sight 216. Alpha 304 shown in Figure 2 has the same value as alpha 304 shown in Figure 3 as they are corresponding angles. The notional mid-point 308 of the eye 208 is the point at which a line taken from the centre point 306 and parallel to the axis 108 intersects the surface of the eye 208. This notional mid-point 308 aids in the calculation of reference data as set out below.

Referring to Figures 2 and 3, reference data may be determined representing the expected reflection offset distance 310 in a non-strabismic eye. The reflection offset distance 310 being the distance between the centre of the pupil 302 and the light source reflection 220. An exemplary method of calculating a reference reflection offset distance 310 is set out below for the left eye 208. A similar method may be arrived at for the right eye 206.

Alpha 304 is determined by: a = taif' This assumes that the distance ddt is equal to the distance from the device plane 202 to the centre 306 of the eye 208, which is a safe assumption if ddf is much greater than the radius of the eye 208. The straight line distance, dr-rn, between the light source reflection 220 and the notional mid-point 308 is then determined by: dr_rn = tan(a)r where r is the radius of the eye 208, i.e. the distance from the centre 306 to the notional mid-point 308.

An angle, beta, may be defined between the line of sight 228 and a straight line parallel to the axis 108 and connecting the left fovea 224 to the device plane 202. Assuming that the fovea 224 is directly behind the centre 306 of the eye 208, which is a safe assumption if ddf is large relative to r and the angle of the line of sight is sufficiently small, the angle beta may be determined by: 1evn/ fl=tan' dff Now, the distance between the notional mid-point 308 on the surface of the eye 208 and the centre of the pupil 302 may be determined by: = sin(fl+ icki where d is the diameter of the eye 208.

The reference reflection offset distance is therefore determined by: d mfrnet = + dmp The above describes one exemplary method for determining the reference reflection offset. It is noted that many other methods may be employed using basic trigonometric functions. The method described above makes certain assumptions, which improve the speed of the calculation without having a significant impact on the determined reference reflection offset value. The assumptions are based on the fact that the distance, ddf, 212 is much greater than the distances in the area of the subject's face, e.g. the radius of the eye, and that the angles of the lines of sight 226. 228 to the axis 108 are small. However, more accurate methods of determining the reference reflection offset may be used if such assumptions become unsafe, i.e. if the relative magnitude of the distance, ddf, 212 and the distances in the area of the subject's face, e.g. the radius of the eye, and the angles of the lines of sight 226, 228 to the axis 108 begin to affect significantly the accuracy of the reference reflection offset distance.

When conducting a test to detect strabismus, a subject is instructed to face the light source 102 and direct their eyes 206, 208 to and focus on the target 104. An image is then taken by the camera 114.

Referring to Figure 4, an exemplary method of processing the image for the detection of strabismus is shown.

Reference data, which may form part of a look-up table, is provided that includes data indicative of pixel values (e.g. intensities and colours) associated with different features within a subject's face, and data indicative of pixel values associated with different features within a subjects eye region. For example, the reference data may include ranges for ROB values typical for various skin types, ranges for ROB values typical for sclera pixels, ranges for ROB values typical for pupil pixels, etc. Image data is received 400 by the processor 122 from the memory 120. The processor 122 pre-processes 402 the image data to determine whether it comprises both the left and right eyes 206, 208 of the subject and to determine the location of the eyes 206, 208. To this end, the processor 122 uses the reference data described above to determine whether the pixels of the image data comprise pixels relating to left and right eye regions of the subject.

Once it is determined that the image data includes the eyes 206, 208 of the subject, the accurate locations of the eye features are determined 404. In order to determine accurate locations for the eye features, the processor 122 re-classifies the pixels within the image data and processes and analyses this data. More specifically, in a non-limiting example of the invention, relative brightness of the pixels is used to determine, for each eye 206, 208, accurate parameter values for the centre of the light source reflection 218, 220 and the centre of the pupil 302. In exemplary systems, the processor 122 is configured to locate the centre of the light source reflection 218, 220 and the centre of the pupil 302 for each of the left and right eyes 206, 208.

A rough location of each of the eyes 206, 208 is determined, by grouping and counting adjacent pixels associated with different parts/features of each eye region, e.g. pupil and sclera zones. A weighted centre of these pixels is calculated, which is taken to be the rough centre of the eye 206, 208. A rectangle is defined around each weighted centre having a width four times larger than its height and covering an area four times the number of pixels in each eye region.

The pixels are re-classified within the rectangle using a reclassification technique based on the ROB values for pixels of different features in the eye region. More specifically, in each eye region, a count is performed of the pixels by their ROB values considering that in a typical eye region there are pixels related to sclera, having ROB values of about 200,200,200, and pixels related to eyelashes or pupil, having ROB values of about 30,15,15. It should be understood that these values, and similar values mentioned below, are not limiting examples of the reference data.

Also, the brightest pixels are identified within each rectangle (eye region) of the left and right eyes 206, 208 in order to locate the light source reflection 218, 220. These pixels may have ROB values of approximately 254,252,250.

The processor 122 counts the number of the brightest pixels, which relate to the light source reflection 218, 220, and calculates rough weighted centres of the light source reflections 218, 220 in each eye. Then, a rough radius of the light source reflection 218, 220 (being large enough to cover the light source reflection 218, 220) is determined as a square root of the number of the light source reflection pixels and a bounding square region is defined around each light source reflection 218, 220 having sides twice the size of the radius.

The rough values of the centre and radius parameters of the light source reflection 218, 220 are used to arrive at more accurate values, as follows.

Within each bounding square, pixels are re-classified according to the corresponding reference data about typical values of light source reflection pixels, light source reflection boundary pixels typically having values intermediate between very bright and very dark, and other pixels. A more precise weighted light source reflection 218, 220 centre is calculated for each eye 206, 208 using the brightness of the light source reflection pixels as their weight. When calculating the weighted centre, a bright pixel counts as one pixel and a boundary pixel counts as less than one pixel, in proportion to how bright it is. Thus, a pixel that is halfway between fully bright and fully dark will be counted as half a pixel. The newly calculated centre in pixel units may or may not be an integer.

Then, the centre of the pupil zone of each eye is located in the rectangle region. The light source reflection 218, 220 is removed from the rectangle. The processor 122 therefore makes a copy of the rectangle region and processes that copied image to remove the light source reflection 218, 220 therefrom and replace the respective pixels by pixels related to the pupil. The pixels associated with the area covered by the light source reflection 218, 220 are now assigned with ROB values being a weighted average of other eye-part pixels nearby. Each of the replaced pixels is assigned with the ROB value of the nearest pixel outside the light source reflection 218, 220, resulting in a rough replacement of the light source reflection 220 by the pupil and iris zones.

The replacement pixels within the removed light source reflection 218, 220 are re-classified as corresponding to pupil, iris, sclera and other (including skin, eye lashes, etc.) as indicated above utilizing the corresponding reference data.

More specifically, dark pixels in each eye region are classified as pupil-related pixels having RGB values of about 12, 12, 12, and the number of these pixels is determined.

Then, a weighted centre of these dark pixels is determined and considered as a rough centre of the pupil. A rough radius of the pupil, being large enough to cover the pupil zone, is determined as a square root of the number of pupil associated pixels and a square having a side of twice the radius is defined as surrounding the rough centre.

The pixels in the bounding square are reclassified as "pupil pixels" (typically very dark), "pupil border pixels" (typically having slightly brighter values) and other pixels". A weighted centre is calculated using this classification where the pupil border pixels are given a weight proportional to their darkness. The result is considered as an accurate centre of the pupil 302 for each eye, which may or may not be an integer in pixel units.

The centre of the pupil 302 of each eye is used as a reference point within the eye to determine 406, for each eye, the reflection offset distance 310, i.e. the distance between the light source reflection 218, 220 and the centre of the pupil 302.

The determined reflection offset distance 310 is compared to the reference reflection offset distance described above to calculate a difference between the two 408.

It is then determined 410 whether the difference is less than a difference threshold value, which can be used to determine whether the subject has a strabismus. In particular, if the difference is less than the difference threshold, the system 100 determines that the subject does not have a strabismus.

Exemplary difference threshold values may be in the range from 0.07mm to 0.13 mm.

A particular exemplary difference threshold value is 0.1 mm.

The result of the determination 410 is output to the user via the display 124.

If the difference is determined to be greater than the difference threshold value, this may indicate either that the subject is not looking at the target, or that the subject is strabismic and is looking at the target. Therefore, if the difference is greater than the difference threshold value, the system 100 may output an indication to the user that the test must be conducted again.

Referring to Figure 5, an arrangement of a system is shown with the target 104 in a plurality of locations with respect to the light source 102. In Figure 5, a plurality of targets 104a to 104h are shown at different locations in the device plane 202.

However, it is noted that the system may comprise only a single target 104 that is movable to each of the different locations. The device plane 202 is parallel to the sheet of paper on which Figure 5 is drawn.

In the exemplary system of Figure 5, eight targets 104a to 104h are positioned equally angularly spaced with respect to the light source 102. In addition, the distance from the light source 102 to each of the targets 104a to 104h is the same. As a result, the targets 1 04a to 1 04h are located on the circumference of a notional circle with the light source 102 at the centre. It is noted that alternative exemplary systems may have other numbers of targets and/or other target arrangements.

When determining whether a subject has a strabismus using the system of Figure 5, eight images are obtained and are each processed as disclosed above. In each of the eight images, a subject fixates their eyes on a different one of the targets 1 04a to 1 04h.

In the exemplary system of Figure 5, the same reference reflection offset distance may be used for each captured image as the distance between the light source 102 and each of the targets 1 04a to 1 04h is the same. Alternatively, a different reference reflection offset distance may be determined for each image.

In particular, the determination of different reference reflection offset distances may be required in systems in which the distance between the light source 102 and each of the targets lO4ato 104h are not the same. In such cases, the principles of determining the reference reflection offset that are desciibed above apply for each image with values changed, as necessary.

The use of the system of Figure 5 allows the detection of a concomitant squint in a plurality of images. A concomitant squint is one in which an affected eye has a squint in every direction that the subject looks. All the eye muscles are working properly, and therefore it is also called a non-paralytic squint. With a concomitant squint the subject's eyes always face in different directions. As the system of Figure 5 obtains a plurality of images, each with the subject looking in a different direction, the existence of a concomitant squint may be determined.

In exemplary systems and methods, an image threshold may be used whereby, if a number of images in which the difference is less than the lower difference threshold is greater than the image threshold, it is determined that the subject does not have a concomitant squint.

In the exemplary system of Figure 5, the image threshold may be five images. In other exemplary systems, the image threshold may be greater than 50% of the number captured images.

Exemplary systems and methods may comprise a lower difference threshold (defined above) and an upper difference threshold value that is higher than the lower threshold value mentioned above. The upper threshold value is for determining whether the subject is looking at the target or not. That is, if the difference between the determined reflection offset distance and the reference reflection offset difference is so great that it is greater than the upper threshold, this indicates that the subject must not have been looking at the target. Therefore, if the difference is greater than the upper threshold value, the user is instructed to undertake another test as described above and the results of the current test are discarded.

In exemplary systems, the processor 122 may determine the reference reflection offset distance 310 based on the relative positions of the light source 102, the target 104 and the subject. This information may be provided by a user of the system, who may measure the respective distances between the light source 102, the target 104 and the subject and input those distances to the processor. The processor may then determine the reference reflection offset distance 310 as part of the method of Figure 4.

However, it is noted that it is not an essential feature of the systems and methods disclosed herein that the processor 122 determines the reference reflection offset distance 310. In exemplary systems and methods, images may be captured while the system is arranged to a predetermined layout with respect to the subject. That is, the distance between the light source 102 and the target 104 and the distance between the face plane 202 and the device plane 204 may be predefined and the system 100 may be arranged to that predefined layout. In such cases, the reference reflection offset distance 310 may be calculated in advance and supplied to the processor.

Exemplary systems may comprise a device that includes a camera part configured to take pictures in conjunction with a light source, such as a camera flash, and a fixation target, such as a colourful flashing LED light. In exemplary systems the fixation target does form part of on the camera part itself, but is offset to one side by a distance, typically in the range from 5 to 10 centimetres. The camera part may be equidistant from each eye, typically facing the patient straight on.

It is noted that the taking of an image of a subject is not an essential feature of the systems and methods disclosed herein. In exemplary systems, images may have been previously and at locations remote from the processor and may be transmitted to the processor for detecting strabismus.

The skilled person will be able to envisage further embodiments of the invention without departing from the scope of the invention as defined in the appended claims.

In particular, although the invention has been described with reference to determining strabismus, it will be appreciated that it may be possible to apply it to other eye conditions.

Claims

<claim-text>CLAIMS: 1. A method for detecting strabismus in at least one image comprising a subject's eyes, wherein the at least one image has been obtained using an apparatus comprising a light source and a target offset from the light source, the method comprising, for each of the subject's left and right eyes: determining, from the image, a reflection offset distance, being the distance between a reference point of the eye and a reflection of a light source on the eye; determining a difference between the reflection offset distance and a reference reflection offset distance for the corresponding left or right eye, wherein the reference reflection offset distance has been determined based on relative positions of the light source, the target and the subject's head; and determining whether the difference is less than a difference threshold value for determining whether the subject has a strabismus.</claim-text> <claim-text>
2. A method according to claim 3, wherein the image has been obtained using an apparatus in which the light source and the target are positioned in a device plane perpendicular to an axis defining the direction that the subject is facing.</claim-text> <claim-text>
3. A method according to claim 2, wherein the image has been obtained using an apparatus in which one of the light source and the target is positioned in front of the subject on the axis defining the direction that the subject is facing, such that the target, the light source and the subject's head form a right angled triangle.</claim-text> <claim-text>
4. A method according to claim 3, wherein the image has been obtained using an apparatus in which the light source is positioned on the axis defining the direction that the subject is facing.</claim-text> <claim-text>
5. A method according to any preceding claim, further comprising obtaining the image.</claim-text> <claim-text>
6. A method according to claim 5, wherein the target comprises an attention grabber, and wherein obtaining the image comprises operating the attention grabber to attract the subject's attention such that the subject's eyes fixate on the target.</claim-text> <claim-text>
7. A method according to claim 6, wherein the image is obtained while the attention grabber is operating.</claim-text> <claim-text>
8. A method according to claim 6 or 7, wherein the attention grabber comprises one or more of a light, a flashing light and a means for generating a sound.</claim-text> <claim-text>
9. A method according to any of claims 5 to 8, further comprising instructing the subject to fixate their eyes on the target.</claim-text> <claim-text>
10. A method according to any preceding claim, wherein the reference reflection offset distance has been determined based on a distance from the light source to the target and a distance from the light source or the target to the subject's head.</claim-text> <claim-text>
11. A method according to any of claims 2 to 10, wherein the reference reflection offset distance has been determined based on the distance from the device plane to the subject's head along the axis defining the direction that the subject is facing.</claim-text> <claim-text>
12. A method according to any preceding claim, further comprising determining the reference reflection offset distance.</claim-text> <claim-text>
13. A method according to any preceding claim, wherein the reference point of the eye comprises the centre of the pupil of the eye.</claim-text> <claim-text>
14. A method according to any preceding claim, further comprising determining that no strabismus is present in the image of the subject's eyes if the difference is less than the difference threshold.</claim-text> <claim-text>
15. A method according to any preceding claim, wherein the at least one image comprises a plurality of images, and wherein each of the plurality of images has been obtained with the target in a different location relative to the light source.</claim-text> <claim-text>
16. A method according to claim 15, wherein the plurality of images has been obtained with the target at a plurality of angularly spaced locations relative to the light source.</claim-text> <claim-text>
17. A method according to claim 16, wherein the plurality of angularly spaced locations are equally angularly spaced.</claim-text> <claim-text>
18. A method according to any of claims 15 to 17, further comprising determining whether the number of images in which the difference is less than the difference threshold value is greater than an image threshold value.</claim-text> <claim-text>
19. A method according to claim 18, further comprising, if the number of images is greater then the image threshold value, determining that the subject does not have a concomitant squint.</claim-text> <claim-text>
20. A method according to any preceding claim, wherein the difference threshold value is a lower threshold value, the method further comprising: if the difference is greater than the lower threshold value, determining whether the difference is greater than an upper threshold value.</claim-text> <claim-text>21. A method according to claim 15, further comprising determining that the subject is not looking at the target if the difference is greater then the upper threshold.22. A computer program product comprising computer program code configured, when run on a computer, to undertake the method of any of claims 1 to 16.18. An apparatus comprising: a processor configured to carry out the method according to any of claims 1 to 18.19. The apparatus of claim 18, further comprising a camera and a light source, each in electrical communication with the processor, wherein the camera is configured to obtain an image while the light source is emitting light.20. A system for detecting strabismus in an image comprising a subject's eyes, the system comprising: a camera; a light source; a target offset from the light source; and a processor in electrical communication with the camera and the light source, wherein the camera is configured to capture an image while the light source is emitting light, and wherein the processor is configured to carry out the method of any of claims 11016.21. A kit of parts comprising: an apparatus according to claim 19 or 20; and a target.22. A device for detecting strabismus in an image comprising a subject's eyes, the image obtained using an apparatus comprising a light source and a target offset from the light source, the device comprising: an input interface configured to receive image data from a camera; a memory for storing image data received from the camera; and a processor configured to process the image data, for each of the subject's left and right eyes, by determining, in the image data, a reflection offset distance, being the distance between a reference point of the eye and a reflection of a light source on the eye, determining a difference between the reflection offset distance and a reference reflection offset distance for the corresponding left or right eye, wherein the reference reflection offset distance has been determined based on a relative position of the light source, the target and the subject's head, and determining whether the difference is loss than a difference threshold value for determining whether the subject has a strabismus.23. An apparatus according to claim 22, further comprising a camera configured to obtain the image data and transmit the image data to the input interface.24. An apparatus according to claim 23, further comprising a light source, wherein the camera is configured to obtain the image data while the light source is emitting light.25. An apparatus according to any of claims 22 to 24, further comprising a display, configured to display to a user information relating to whether the subject has a strabismus.26. A method substantially as herein described with reference to the accompanying drawings.27. An apparatus substantially as herein described with reference to the accompanying drawings.28. A system substantially as herein described with reference to the accompanying drawings.AMENDMENTS TO THE CLAIMS HAVE BEEN FILED AS FOLLOWSCLAIMS: 1. A method of processing at least one image for detecting strabismus, wherein the at least one image has been obtained using an apparatus comprising a light source and a target offset from the light source, the method comprising, for each of the subject's left and right eyes: determining, from the image, a reflection offset distance, being the distance between a reference point of the eye and a reflection of a light source on the eye; determining a difference between the reflection offset distance and a reference reflection offset distance for the corresponding left or right eye, wherein the reference reflection offset distance has been determined based on relative positions of the light source, the target and the subject's head; and determining whether the difference is less than a difference threshold value.2. A method according to claim 1, wherein the image has been obtained using an (Y) apparatus in which the light source and the target are positioned in a device plane perpendicular to an axis defining the direction that the subject is facing.0 3. A method according to claim 2, wherein the image has been obtained using an 00 20 apparatus in which one of the light source and the target is positioned in front of the subject on the axis defining the direction that the subject is facing, such that the target, the light source and the subject's head form a right angled triangle.4. A method according to claim 3, wherein the image has been obtained using an apparatus in which the light source is positioned on the axis defining the direction that the subject is facing.5. A method according to any preceding claim, further comprising obtaining the image.6. A method according to claim 5, wherein the target comprises an attention grabber, and wherein obtaining the image comprises operating the attention grabber to attract the subject's attention such that the subject's eyes fixate on the target.7. A method according to claim 6, wherein the image is obtained while the attention grabber is operating.8. A method according to claim 6 or 7, wherein the attention grabber comprises one or more of a light, a flashing light and a means for generating a sound.9. A method according to any of claims 5 to 8, further comprising instructing the subject to fixate their eyes on the target.10. A method according to any preceding claim, wherein the reference reflection offset distance has been determined based on a distance from the light source to the target and a distance from the light source or the target to the subject's head.11. A method according to any of claims 2 to 10, wherein the reference reflection offset distance has been determined based on the distance from the device plane to C?) the subject's head along the axis defining the direction that the subject is facing.12. A method according to any preceding claim, further comprising determining the reference reflection offset distance.13. A method according to any preceding claim, wherein the reference point of the eye comprises the centre of the pupil of the eye.14. A method according to any preceding claim, further comprising determining that no strabismus is present in the image of the subject's eyes if the difference is less than the difference threshold.15. A method according to any preceding claim, wherein the at least one image comprises a plurality of images, and wherein each of the plurality of images has been obtained with the target in a different location relative to the light source.16. A method according to claim 15, wherein the plurality of images has been obtained with the target at a plurality of angularly spaced locations relative to the light source.17. A method according to claim 16, wherein the plurality of angularly spaced locations are equally angularly spaced.18. A method according to any of claims 15 to 17, further comprising determining whether the number of images in which the difference is less than the difference threshold value is greater than an image threshold value.19. A method according to claim 18, further comprising, if the number of images is greater then the image threshold value, determining that the subject does not have a concomitant squint.20. A method according to any preceding claim, wherein the difference threshold value is a lower threshold value, the method further comprising: if the difference is greater than the lower threshold value, determining whether the difference is greater than an upper threshold value.r. . .
21. A method according to claim 20, further comprising determining that the subject is not looking at the target if the difference is greater then the upper threshold.00 20
22. A computer program product comprising computer program code configured, when run on a computer, to undertake the method of any of claims 1 to 21.
23. An apparatus comprising: a processor configured to carry out the method according to any of claims 1 to 21.
24. The apparatus of claim 23, further comprising a camera and a light source, each in electrical communication with the processor, wherein the camera is configured to obtain an image while the light source is emitting light.
25. A system for processing at least one image for detecting strabismus, the system comprising: a camera; a light source; a target offset from the light source; and a processor in electrical communication with the camera and the light source, wherein the camera is configured to capture an image while the light source is emitting light, and wherein the processor is configured to carry out the method of any of claims 11021.
26. A kit of parts comprising: an apparatus according to claim 24; and a target.
27. A device for processing at least one image for detecting strabismus, the image obtained using an apparatus comprising a light source and a target offset from the light source, the device comprising: an input interface configured to receive image data from a camera; a memory for storing image data received from the camera; and a processor configured to process the image data, for each of the subject's left and right eyes, by determining, in the image data, a reflection offset distance, being the distance between a reference point of the eye and a reflection of a light source 00 20 on the eye, determining a difference between the reflection offset distance and a reference reflection offset distance for the corresponding left or right eye, wherein the reference reflection offset distance has been determined based on a relative position of the light source, the target and the subject's head, and determining whether the difference is loss than a difference threshold value.
28. A device according to claim 27, further comprising a camera configured to obtain the image data and transmit the image data to the input interface.
29. A device according to claim 28, further comprising a light source, wherein the camera is configured to obtain the image data while the light source is emitting light.
30. An apparatus according to any of claims 27 to 29, further comprising a display, configured to display to a user information relating to whether the subject has a strabismus.
31. A method substantially as herein described with reference to the accompanying drawings.
32. An apparatus substantially as herein described with reference to the accompanying drawings.
33. A system substantially as herein described with reference to the accompanying drawings. r Co</claim-text>