Bryan Shaw and Greg Hamerly

When Bryan Shaw’s son, Noah, was diagnosed with pediatric eye cancer at four months old, it sparked in Shaw a mission to help other families avoid the challenges his faced. On this Baylor Connections, Dr. Bryan Shaw, associate professor of chemistry, and Dr. Greg Hamerly, associate professor of computer science, share the moving story behind the White Eye Detector app, an effective screening tool that analyzes photos in the phone to more quickly catch pediatric eye disease. Retinoblastoma claimed Noah’s eye, but today he is a healthy 11-year old who knows his experiences are helping others.
Transcript
DEREK SMITH:
Hello and welcome to Baylor Connections, a conversation series with the people shaping our future. Each week we go in depth with Baylor leaders, professors and more, discussing important topics in higher education, research and student life. I'm Derek Smith and our guests today are Dr. Bryan Shaw and Dr. Greg Hamerly. Dr. Shaw serves as an associate professor of chemistry and biochemistry at Baylor and Dr. Hamerly is an associate professor of computer science. Together they developed an app to help families detect early signs of various eye diseases including the pediatric eye cancer, retinoblastoma. Shaw's son, Noah, developed retinoblastoma as an infant. Noah, now a healthy 11 year old, lost an eye to the disease, an experience that led Shaw to partner with Hamerly to help other families detect disease more quickly. Together, they developed an app called CRADLE, that searches family photos for signs of white eye that can signal the presence of disease. And the prototype app was recently featured in the journal Science Advances, with a study finding it to be an effective tool to augment clinical screenings. It's been a busy stretch for both of you, but it's great to have you on the program. Thanks so much for joining us today.
GREG HAMERLY:
Thank you for having us.
BRYAN SHAW:
Thanks Derek.
DEREK SMITH:
Well, we're visiting with you, you visited with CBS this morning, not too long ago for a story they are working on. So it's exciting to see that national attention coming to work that we've been able to see here at Baylor for a little while now. And let's dive right into it. Let's start with a more, I gave a brief description of CRADLE and what we're talking about. But can you take us a little deeper in that and how it works for families?
BRYAN SHAW:
Well, let's start off with the name. So it is called CRADLE, Computer Assisted Detector of Leukocoria, but Apple kind of wanted us to change the name a little to something a bit more descriptive. So it's also called the White Eye Detector. So if you're searching for it, use White Eye Detector and you'll always be able to find it.
GREG HAMERLY:
And you can find it in the Apple app store or in the Google Play Store. So it's for Android or iOS.
DEREK SMITH:
Get it for both. Would either of you have imagined, Dr. Shaw, you're a Chemist, Dr. Hamerly, Computer Scientist. Would you have imagined sitting here talking about what we're about to talk about, and not only from your own personal experience, but just as academics, the way you've come together to create something like this, an app that helps families detect disease?
BRYAN SHAW:
You know, there's a lot of things that happened to me in relation to this project I would have never anticipated. Having a four month old diagnosed with tumors in both of his eyes was the beginning, but after that, all the other stuff wasn't too surprising.
GREG HAMERLY:
Well, as an academic, you always hope to make an impact in the real world, but it doesn't always happen, right? You publish research and it goes into the ether and this really has a real world impact, which has been really satisfying.
DEREK SMITH:
Well, obviously a pretty incredible personal story and an incredible young man in Noah, as we mentioned now, a healthy 11 year old, but Dr. Shaw, you and your family experienced some real challenges. On this side of the ledger first, what does it mean to you and to now be on the side of things, helping other families, hopefully a catch this early?
BRYAN SHAW:
Well it's good, it warms your heart. You know, as a father, having Noah as a child, it's now the benefits are for him to enjoy too. He's getting older and he knows what's going on and now I can show him the other children that have been helped because of the ordeal he went through. And that's more important now than it's ever been for him. So that's something new.
DEREK SMITH:
Could you take us inside those early experiences? What it was like for you and your family in those early days?
BRYAN SHAW:
Well, it was rough. It was the best of times and the worst of times. I was just getting ready to go on the job market and my wife noticed these white pupils in all of his pictures and she finally told the pediatrician, and within four hours we were referred to an ophthalmologist and he was diagnosed with tumors in both of his eyes. Right as he sat on my lap, the ophthalmologist made the diagnosis in minutes and it was just one nightmare after another.
DEREK SMITH:
Wow.
BRYAN SHAW:
It was awful, but we got through it.
DEREK SMITH:
How common is this disease? And as you were working through it with Noah, what did you learn about it?
BRYAN SHAW:
Yeah, so with retinoblastoma, there's about 8,000 children in the world that get it each year and half of them will die. Most of those deaths are going to be in resource limited settings because the tumor is diagnosed late and it's already crawled down the optic nerve back to the brain. So the health disparity that I learned about was pretty shocking. There were all sorts of other disorders that present with this white pupil or this Leukocoria as the doctors refer to it. And some of those are common like refractive error in myopia, and strabismus, and amblyopia and others are rare, Coats disease.
DEREK SMITH:
You know, you mentioned the stark statistics. What were some of those moments like? And what were the moments like as you realized there was going to be the opportunity, to now that you've found it, to treat it?
BRYAN SHAW:
Well, whenever you have a kid with bilateral retinoblastoma tumors in both of the eyes, you also have a five to 10% chance of developing retinoblastoma on your third eye, your pineal gland. You actually have a gland in the center of your head with photo receptors, you could call it your third eye. And if it develops there, it is not good. So you get MRIs every few months. And there was a time where it looked like maybe he had developed that and it was on a Friday. And they had to do a spinal tap without anesthesia, because we'd already used the anesthesia for the MRI and cytology couldn't examine the results til Monday. So you know, all weekend you're thinking your boy has brain cancer and he may not survive and I was in Cambridge, Massachusetts, I was a postdoc at Harvard, I didn't even live there. I didn't know where would I bury him, it was just... You just think about these awful things and it's rough.
DEREK SMITH:
And he was how old when this process began?
BRYAN SHAW:
Four months.
DEREK SMITH:
Four months.
BRYAN SHAW:
He was diagnosed on exactly four months, at his four month Well-baby visit.
DEREK SMITH:
Visiting with Dr. Bryan Shaw and Dr. Greg Hamerly. And you mentioned that your wife noticed white eye in the pictures. What is the significance of that? Which obviously, as we talk about the app later on, will take on even greater significance in your work.
BRYAN SHAW:
So if you see a white pupil in a picture of your child, it can mean a lot of things. It can mean that you're just reflecting light off the optic disc, the little 1.8 to two millimeter white disc in the back of your eye, and that is not always a problem. It could mean you have refractor error, or lazy eye, or something like that. But you can also have white pupils because light's reflecting off the tumor, or light's reflecting off a cataract in your lens, or a cholesterol deposit in the back of your eye, if you have Coats disease, or if your retina's damaged, retinopathy of prematurity, ROP, can cause Leukocoria. So it's a symptom of a lot of different disorders. And this app, we were quite surprised to find how many of these disorders the app's able to catch.
GREG HAMERLY:
And to be clear, people are used to seeing red eye pupils in photos and that's not an issue. And that's the kind of thing that we're looking for, except white. And if the reflection is off the surface of the eye, you get what we call a specular reflection at this point source, off the surface of the eye, that's not a problem either. But if the pupil is filled with a white light, that's the type of symptom that we're looking for.
DEREK SMITH:
And Dr. Hamerly, as we hear Dr. Shaw tell his story and his son, Noah's story, certainly, I think for those of us hearing it from him the first time it's quite poignant. But as you've got to know him and his family as colleagues, and as researchers, and as friends, what stands out to you about the Shaw family and what they faced, and that just desire to channel it for something greater?
GREG HAMERLY:
Well, I mean, I find it difficult to imagine what they've gone through. Even knowing them for a while and hearing their story and knowing them, it's just hard. I'm a father of two boys who have similar ages to Bryan's two boys and I can't imagine going through that. So just wanting to support them, but also I see in their family a ton of energy and desire to help other people. I've seen Bryan and his family go time after time in support of, virtual support usually, or calling people, emailing people, trying to get the word out about these diseases, trying to get people the support that they need, the help that they need for people who have heard about our project, written to us, written to Bryan and needed support, moral support. And he's just always been there for them. So I'd say their desire to be a good word about what can happen with this disease, but also how you can get support and what people need to be on the lookout for, is just, they're indefatigable, that's the word I'm looking for, is indefatigable. They're tireless in their efforts here. And it's been really an admirable thing.
DEREK SMITH:
Dr. Shaw, Noah, these were mentioned. He's a healthy, 11 year old. The disease did claim an eye. How old was he when that decision was made, and that realization was this, this is how we treat it? And is it something that with early detection maybe could have been avoided?
BRYAN SHAW:
Yeah. So we caught it at three months old, and then we reported it at four months old, and he was diagnosed at four months old. And you give them chemo, systemic chemo, to try and shrink the tumors and sometimes the tumors respond well and they just almost dissolve, chemo reduction it's called, but his didn't. And then they started to break up and when that happens you can have seeding and metastasis. So we fought it for about five months and then at nine months he lost the eye. And the tumor was so big in the eye that it began to wander. The eye had begun to wander and he probably didn't have much vision in it anyway, we could have saved it may be with radiation, but the dose of radiation we would have had to have given him would have been so large, and for an eye that you don't have much vision in. If we would've caught it when the white eye emerged, looking back in photos, the white eye began to emerge at 12 days old, we just didn't notice it. If we had caught it then, his ophthalmologist predicted he would have just got laser photo ablation therapy. The tumors would have been so small, millimeter, sub-millimeter, they would have just blasted them with the laser. No chemo, no radiation, no eye removal. Or this is one of the cancers where early detection totally improves the prognosis and it improves the vision, however big the tumor is, that's how big your blind spot's going to be.
DEREK SMITH:
This is Baylor Connections. We are visiting with Dr. Bryan Shaw and Dr. Greg Hamerly, and as you got some distance from this end, you began to think about ways to help other families dealing with this and help detect this earlier. What was the genesis of this app, that idea formation? And how did you both come to work together?
BRYAN SHAW:
For me, the idea that a parent could be helped by technology that searched through their photos, popped in the brain. It just, the light bulb went off when I saw a picture of my son with white eye at 12 days old, months before it was diagnosed. And after his doctor told me what the prognosis would have been if we had brought him in earlier, I thought, oh boy, we need some sort of technology. And then I moved to Baylor and I met Greg.
DEREK SMITH:
Greg, what were your thoughts when this was presented to you?
GREG HAMERLY:
So my research area is in machine learning and it seemed like a natural fit to my research area because machine learning has been used to tackle a lot of the, if you take this problem of taking a natural photo, and when I say natural, I mean not clinical, and trying to identify symptoms of disease, you need to break that down into parts. And those parts are like where are their faces in this photo? And where are the eyes in this photo? And do those eyes look diseased? And when I sort of heard Bryan's story and he described where we want to go with this, I immediately thought, well you can break that problem down into these parts, and a lot of those, if not all of those, can be answered with machine learning approaches. And so the immediate idea was what it became. I mean, the immediate idea was, let's take these approaches and that's sort of what we did. The actual getting from idea to the app itself took a while, and at the time when we first met, smart phones existed, but they were not nearly what they are today. So we had a few false starts at the beginning, but the path ahead seemed clear. It was just getting there and finding the right people like Ryan Henning to work on this project with us, who was my former graduate student.
DEREK SMITH:
I know there's been a lot of great people, past and current who have worked with you on this, to get it to this point. I think you mentioned a little of the challenges in terms of, whether it's finding a location in the photo of the eye, or it's a non-professional photo, but what were some of the, whatever you want to call them, hurdles or just problems that needed to be solved along the way? What were some of the biggest areas that you had to apply your expertise to?
GREG HAMERLY:
Well, the first thing is, for any machine learning project you have to have sort of examples of what you're looking for. And we needed examples of what looked like diseased eyes and what looked like normal eyes. And finding normal eyes is really easy, pictures of normal eyes. You just go out on the internet, look for creative, commons images and there are millions of them. Finding examples of the diseased eyes that we were looking for is quite a bit more difficult because this doesn't affect as many people as... Well, in order to collect enough information, you need a lot of images. And so we collected initially images from Noah and then we also, as the story got around, we also were able to collect many more images. But initially we just had a few images. So collecting data was one of the first problems. Another major problem is just the robustness of the detector itself. So you want it to make sure that it's actually detecting faces and not hubcaps, which we've seen. We want to make sure that it's actually triggering on the right kind of disease presentation and is not giving you false positive triggers. So we don't want the app to scare people and we don't want it to scare parents. We want it to only trigger in cases where there really is white eye presentation. And so getting it to be robust with respect to in different lighting conditions, different face orientations, people being far away in the picture, people being close in the picture, all these things are just variables that you don't control when you don't control who's taking the picture and under what conditions. And so the other major problem was just getting it to behave well in all different situations.
BRYAN SHAW:
And you know, all the disadvantages of diverse photography that parents take in non-clinical settings, all these challenges that Greg just described, there's one huge benefit to it and that is parents take pictures of children at multiple optic axes. And so that increases, when a doctor does the red reflex test, when they shine a light in your kid's eye and at a pediatric pediatric visit, undulated, they're only seeing 2% of the retina. But mom, dad, grandma, grandpa, whoever, they take pictures of all of these different angles-
GREG HAMERLY:
And they're doing it all the time.
BRYAN SHAW:
And they're doing it all the time. And if they're doing it in the holidays, it might be dark light, pupil may be open up. And so you increase the probability that if there is a lesion, light from the flash will reflect off of it and reflect back to the detector in your camera. So the types of photos that parents take are a curse and blessing for us, I would say.
GREG HAMERLY:
But we're playing a numbers game here, because like Bryan is saying, a pediatrician will screen your child once every few months. We get tons of information about that child's eyes into the app at every opportunity. And so we can afford to be very conservative with when the app triggers, because if there is a symptom that's going to present itself, it's going to present itself many times over. And if we can just catch a small percentage of that, then we can alert appropriately the parents when that happens.
BRYAN SHAW:
Yeah. So if a parent has, five, 10, 15 white eye picks on their phone, we don't need to detect everyone, we just need to detect one to alert them and they can investigate.
DEREK SMITH:
We are visiting with Dr. Bryan Shaw and Dr. Greg Hamerly, and as we head into the final few moments on the program, you've painted a really good picture of how this works and the benefits of it. You recently passed a pretty important step on the way with the Science Advances study that talked about it being a very good tool that can, I guess, in some cases help parents. Obviously it doesn't replace the doctor, but it can replace the necessity of getting a screening early on. What does that study mean for the app going forward?
BRYAN SHAW:
Well, it, it really proves for the first time that casual photographs that parents take can be useful for this app. It shows that this app is capable of detecting Leukocoria in casual sets of photographs and that it can detect Leukocoria or white eye in the pictures long before the child was eventually diagnosed by a doctor.
GREG HAMERLY:
Yeah. Until this point we had our beliefs that this worked well, but this is a validation of the approach.
BRYAN SHAW:
It's the first time anyone's ever even looked at the frequency of white eye pictures, how often they pop up in casual sets of photographs that parents take of kids with all of these different eye disorders. Nobody had ever even looked at how often white eye shows up in pictures that parents take of their child from birth through diagnosis, treatment, remission, in the case of cancer. So that was nice.
DEREK SMITH:
Dr. Hamerly mentioned they can get it on iTunes in the Google store, and again it's called the CRADLE app, but how best can families find this?
GREG HAMERLY:
Yeah, they can search for the White Eye Detector or White Eye app. I believe it might also come up under CRADLE, but the best option is to search for White Eye Detector.
DEREK SMITH:
White Eye Detector.
BRYAN SHAW:
I think it comes up, for the Android version, it comes up whether you search CRADLE or White Eye Detector, but for Apple, I think you have to search CRADLE. And the Android and the Apple version work a little differently. The Android version, once you turn it on, it will scan in the background.
GREG HAMERLY:
Every photo.
BRYAN SHAW:
Constantly, forever until you turn it off. And if there's a problem, it'll tell you. The iPhone version, Apple doesn't like apps running in the background because of the battery drain issues. So you have to turn it on and search your photos each time you wish.
GREG HAMERLY:
But the Android version will not drain your battery. I run it on my phone, it's on.
DEREK SMITH:
Dr. Shaw, it would be remiss if we didn't ask you that, what might we find Noah doing today? What are some of his talents and gifts that you get a lot of joy out of watching him do?
BRYAN SHAW:
Well, Noah, he has some real gifts. He's a precocious reader. He started reading very early and we've learned that despite having slight hearing damage from the carboplatin, despite having hearing aids,, we've had some music instructors, two of them actually tell us he has perfect pitch. So he can call out notes and he's a pretty good bass player. So he has some musical gifts and so he plays a lot of music. He crafts a lot. He's always building stuff with paper and tape, he's having a good time in life. I mean, you can look back and say, "Gosh, why did this happen to such an innocent child? Why did such a bad thing happened to him?" But if you count up his curses and his blessings, way, way, way more blessings.
DEREK SMITH:
That's fantastic. Well, quite an impressive young man and I'm glad to hear you said he's starting to understand how what happened to him is going to help other people. And I'm sure as he gets older he'll even see that even more.
BRYAN SHAW:
One last thing, we always need more pictures. If your child has amblyopia, strabismus, refractive error, cataract, even the rare serious disorders like retinoblastoma, Coats disease, and you want to donate your images to our research team, they'll help us out. And you can obviously remain confidential.
DEREK SMITH:
Right. How can they get that to you?
BRYAN SHAW:
You can email me at bryan_shaw@baylor.edu, bryan_shaw@baylor.edu.
DEREK SMITH:
Bryan with a Y.
BRYAN SHAW:
Bryan with a Y. Or find Bryan Shaw or Greg Hamerly at computer science or chemistry at Baylor.
DEREK SMITH:
So email those and they will take you up on those. Well, thank you both very much.
GREG HAMERLY:
Thank you.
DEREK SMITH:
Thank you. Again, we hope people will take the opportunity. Young parents, maybe if they're expecting to go ahead and look this up so they have it at the ready. White Eye Detector and the White Eye App. Our guests today, Dr. Bryan Shaw and Dr. Greg Hamerly here on Baylor Connections. I'm Derek Smith. Reminder, you can hear this and other programs online, baylor.edu/connections.