1 00:00:00,000 --> 00:00:12,083 [MUSIC] 2 00:00:12,083 --> 00:00:17,270 The back surface of the eye contains layers of neural tissue. 3 00:00:17,270 --> 00:00:19,260 So here's a schematic. 4 00:00:19,260 --> 00:00:23,650 We're looking at this surface here, it's known as the retina. 5 00:00:23,650 --> 00:00:29,970 One of the layers of cells here is a layer of light sensitive cells. 6 00:00:29,970 --> 00:00:33,150 These are the rods and cone cells. 7 00:00:33,150 --> 00:00:37,480 The rods are very sensitive indeed, they allow us to see in dim light levels. 8 00:00:37,480 --> 00:00:41,680 The cones are less sensitive, they allow us to see in daylight levels. 9 00:00:41,680 --> 00:00:46,130 In the retina there are approximately 6 million cone cells. 10 00:00:46,130 --> 00:00:49,920 Most of them are concentrated in the central region of the retina here 11 00:00:49,920 --> 00:00:51,510 where our vision is best. 12 00:00:51,510 --> 00:00:53,320 The cone cells are very small. 13 00:00:53,320 --> 00:00:56,000 They're between two and eight microns in diameter. 14 00:00:56,000 --> 00:01:00,280 That's 2 thousandths of a millimeter or 8 thousandths of a millimeter. 15 00:01:00,280 --> 00:01:04,630 And to see individual cells of that size, we would normally use a microscope. 16 00:01:04,630 --> 00:01:08,700 A sample of tissue would be put on a microscope slide, and we'd 17 00:01:08,700 --> 00:01:13,270 use the microscope to magnify the tissue so that we could see individual cells. 18 00:01:13,270 --> 00:01:16,030 We want to image cells in the living human eye. 19 00:01:16,030 --> 00:01:21,020 We want to be able to see individual cone cells by looking into the eye. 20 00:01:21,020 --> 00:01:23,100 >> Here in the lab we've built a special microscope for 21 00:01:23,100 --> 00:01:25,390 looking in the back of a participant's eye. 22 00:01:25,390 --> 00:01:27,100 But unlike a traditional camera, 23 00:01:27,100 --> 00:01:30,520 we don't a picture of the entire retina all at once. 24 00:01:30,520 --> 00:01:34,800 Instead what we do is focus a laser beam into the eye. 25 00:01:34,800 --> 00:01:39,610 So the laser is directed into the eye via these lenses and mirrors and 26 00:01:39,610 --> 00:01:43,480 via these two special mirrors that scan the beam. 27 00:01:43,480 --> 00:01:47,800 So one moves the beam horizontally, the other moves it vertically, and 28 00:01:47,800 --> 00:01:52,540 together, they trace a rectangular patch on the retina. 29 00:01:52,540 --> 00:01:57,440 Then, some extra mirrors relay the light into the participant's eye. 30 00:01:57,440 --> 00:02:02,640 Some of that light reflects back out and through these lenses and mirrors again. 31 00:02:02,640 --> 00:02:06,750 We collect some of that light and use it to measure the optical distortions in 32 00:02:06,750 --> 00:02:10,020 the eye, which we can correct with this mirror. 33 00:02:10,020 --> 00:02:13,495 Most of the light goes onto this very sensitive detector. 34 00:02:13,495 --> 00:02:18,130 And what we can do is use that to build up an image by measuring 35 00:02:18,130 --> 00:02:23,290 the amount of light at each position as it is scanned. 36 00:02:23,290 --> 00:02:28,860 And using some very clever software, we can reconstruct an image pixel by pixel. 37 00:02:28,860 --> 00:02:33,100 >> We're borrowing techniques from astronomy to be able to do this and 38 00:02:33,100 --> 00:02:35,690 we've worked in collaboration with 39 00:02:35,690 --> 00:02:40,160 researchers in the instrumentation group at Durham University. 40 00:02:40,160 --> 00:02:45,050 The final lens in our microscope is actually the eyes lens, so 41 00:02:45,050 --> 00:02:48,040 it's the lens of a living human eye. 42 00:02:48,040 --> 00:02:49,630 Light has to pass through here so 43 00:02:49,630 --> 00:02:52,850 that we can image the retina at the back of the eye. 44 00:02:52,850 --> 00:02:56,680 That introduces a few challenges, so even for an observer who's got 45 00:02:56,680 --> 00:03:01,520 perfect eyesight, the optical quality of these components is not perfect. 46 00:03:01,520 --> 00:03:06,160 And secondly, if we ask our participants to try to keep their eyes still, 47 00:03:06,160 --> 00:03:07,370 it's not possible to do that. 48 00:03:07,370 --> 00:03:10,800 The eye always makes minute eye movements. 49 00:03:10,800 --> 00:03:14,860 >> One of the key components of this system is this deformable mirror. 50 00:03:14,860 --> 00:03:19,670 This is what allows us to see these really tiny cells at the back of the eye. 51 00:03:19,670 --> 00:03:24,240 This mirror has lots of small actuators behind it's surface, and 52 00:03:24,240 --> 00:03:29,860 these actuators push and pull to change the shape of the mirror surface. 53 00:03:29,860 --> 00:03:34,330 By measuring the optical distortions in the eye, we can calculate the shape 54 00:03:34,330 --> 00:03:39,300 that this mirror needs to have to cancel out those optical distortions. 55 00:03:39,300 --> 00:03:45,040 Allowing us to recover a very detailed picture of the back of the eye. 56 00:03:45,040 --> 00:03:47,540 Okay, Anna, we're going to take a picture of the back of your eye. 57 00:03:47,540 --> 00:03:51,300 So if you'd like to come forward and bite onto the dental impression. 58 00:03:51,300 --> 00:03:54,120 I'm now going to align your eye lens to the system so 59 00:03:54,120 --> 00:03:57,690 that we can get the best possible image that we can. 60 00:03:57,690 --> 00:04:00,520 I'm now going to put a fixation cross on the screen in front of you. 61 00:04:00,520 --> 00:04:06,040 If you could just keep your eye fixated on that while I take some images. 62 00:04:06,040 --> 00:04:10,680 >> So the ability to caption high resolution images of the living eye 63 00:04:10,680 --> 00:04:13,300 has opened up new avenues of research. 64 00:04:13,300 --> 00:04:16,140 Labs around the world are exploiting this ability to 65 00:04:16,140 --> 00:04:21,000 see individual cells in the living human eye. 66 00:04:21,000 --> 00:04:25,610 We can obviously start to analyze cone statistics. 67 00:04:25,610 --> 00:04:29,397 So the density and the packing arrangement of cells within the retina. 68 00:04:29,397 --> 00:04:34,250 And we can think about how that impacts our ability to see. 69 00:04:34,250 --> 00:04:39,170 Each of these blobs is an individual cone cell. 70 00:04:39,170 --> 00:04:40,950 But the cones are not all equal. 71 00:04:40,950 --> 00:04:44,400 So there are three different types of cone cell. 72 00:04:44,400 --> 00:04:48,770 Cones that are sensitive to different ranges of wavelengths. 73 00:04:48,770 --> 00:04:52,930 Some of them are sensitive to short wavelengths, some to middle wavelengths, 74 00:04:52,930 --> 00:04:54,530 and some to long wavelengths. 75 00:04:54,530 --> 00:04:59,170 And its comparisons between these three classes of cone signals that allow us 76 00:04:59,170 --> 00:05:06,530 to see color, to extract wavelength information about the external world. 77 00:05:06,530 --> 00:05:12,410 So it's possible to identify the type of cone of each individual cell here. 78 00:05:12,410 --> 00:05:16,540 They're either short, middle, or long wavelength sensitive. 79 00:05:16,540 --> 00:05:21,840 We can also use the system to image the retina in health and in disease. 80 00:05:21,840 --> 00:05:27,060 And we can look at minuscule changes that occur within disease, 81 00:05:27,060 --> 00:05:32,100 opening up the possibility to track these changes at a very fine scale, 82 00:05:32,100 --> 00:05:37,320 give us early warning signals about change in the damaged eye and to identify 83 00:05:37,320 --> 00:05:41,370 possible candidates for treatment, for example in gene therapy trials. 84 00:05:41,370 --> 00:05:45,140 By adjusting the depth of focus of our microscope, 85 00:05:45,140 --> 00:05:49,940 of our imaging system, we can also look at different layers in the retina. 86 00:05:49,940 --> 00:05:56,940 And here's an example where we're focusing on the blood vessel layer. 87 00:05:56,940 --> 00:06:00,310 And, because we have such high resolution images, 88 00:06:00,310 --> 00:06:05,670 we can see blood flow of individual blood cells moving through these vessels. 89 00:06:05,670 --> 00:06:08,290 We can also focus into different depth planes, and 90 00:06:08,290 --> 00:06:10,090 look at other structures within the retina. 91 00:06:10,090 --> 00:06:14,530 So here's an example where we're focused on the nerve fiber layer. 92 00:06:14,530 --> 00:06:17,620 This is the layer of axons of all those neurons in the retina 93 00:06:17,620 --> 00:06:20,870 before they leave the retina via the optic nerve. 94 00:06:20,870 --> 00:06:25,600 The ability to image at this scale, at this microscopic scale, 95 00:06:25,600 --> 00:06:29,460 also opens up the possibility of stimulating the retina at that scale. 96 00:06:29,460 --> 00:06:33,800 So it's possible to send light to an individual cone cell and 97 00:06:33,800 --> 00:06:40,140 to track the ability to use that signal to elicit percepts, for example. 98 00:06:40,140 --> 00:06:44,240 So we can stimulate the retina at the single cone level. 99 00:06:44,240 --> 00:06:47,570 We mentioned one of the challenges of this sort of imaging being 100 00:06:47,570 --> 00:06:52,160 the fact that the human eye is always in motion. 101 00:06:52,160 --> 00:06:56,480 As well as that being a challenge, it's also an open question in research. 102 00:06:56,480 --> 00:06:58,860 So why does the eye move all the time and 103 00:06:58,860 --> 00:07:03,310 in what way does that impact our ability to see? 104 00:07:03,310 --> 00:07:07,220 One of the fundamental things we know about vision is that if we suppress these 105 00:07:07,220 --> 00:07:13,060 movements, if we stabilize the image on the retina, then vision fades to nothing. 106 00:07:13,060 --> 00:07:16,400 So it's fundamentally important that the eye moves in order for 107 00:07:16,400 --> 00:07:18,120 us to be able to see it all. 108 00:07:18,120 --> 00:07:20,915 Being able to measure eye movements at this scale, 109 00:07:20,915 --> 00:07:24,231 opens up exciting new avenues of research to understand more 110 00:07:24,231 --> 00:07:28,327 about the link between eye movements, the stimulation of the retina, and 111 00:07:28,327 --> 00:07:40,840 the biological properties of the cells in the retina.