Prof Clare Elwell
Clare Elwell is a professor of medical physics at UCL. Her research involves the development and application of novel optical methods for investigating the human brain, particularly in infants.
One of the challenges for medical physics is to measure something without changing it and nowhere I think is that more true than when you’re looking at the brain.
On becoming interested in science
My mum’s a nurse, so I was around hospitals a lot and I always had a fascination with medical things and I quite liked the smell of a hospital. But I remember a book that was on my parents’ bookshelf which was treating the body as a machine, so you treated each of the different systems in the body as mechanical type processes, and that’s when I really got an interest in thinking about how can we treat the body, not just from a biology point of view but from a physics point of view, and then I thought, well, perhaps that means I really want to be a doctor, but then I really didn’t like chemistry very much. So when I was at school I liked maths and physics and I just went into a lecture about medical physics one day when I was 17 and a lightbulb moment happened and I thought, that’s it, this is what I want to do, and I’d never heard of medical physics before then but once I’d been in that lecture I realised that’s what I wanted to do.
On the area of medical physics
So, medical physics is about the application of science and technology and engineering to medical problems. So you’ve all heard of X-ray, ultrasound, MRI, those technologies are developed by medical physicists. And, actually, the birth of medical physics was the discovery of the X-ray. How can we use X-rays, which we understand from a physics point of view, to interrogate the body and get images. And it’s that interface which medical physics lies at.
On advancements in brain imaging
So, MRI has revolutionised the way we can look inside the brain; so from a structural point of view it provides exquisite images of anatomy, so it can tell us about what the brain structure is and, importantly, what shouldn’t be in the brain, for example a tumour. But it can also tell us a huge amount about function and it’s been used to navigate our way around the processes of the brain. And this enables us to understand how the brain is connected together, what type of activities are going on in the brain and potentially to have a look at conditions such as Alzheimer’s.
One of the limitations of current conventional brain imaging technologies is that most of them involve large scanners and they’re expensive and they will only suit a certain type of subject or patient. And, critically, one type of subject or patient that can’t be monitored in these big scanners like MRI are infants, and there’s a huge amount of untapped information in the developing brain. So the technologies that I’ve been developing using light which don’t involve big scanners are enabling us to really work out the brain development of normal and potentially abnormal infants.
On current research
So one of my current projects is looking at using our optical brain imaging systems to understand how the development of the brain processes social information, and we’ve shown in a recent study that there are differences between infants at low risk of autism and high risk of autism in the way in which they process both visual and sound information that may be either social or non-social.
So the early work of this project started with us using optical brain imaging systems to look at babies who were in intensive care to find out whether they’ve suffered some kind of brain injury. We’ve been talking about this work and publishing this work and as a result we were contacted by a group of psychologists who were interested in autism and the developing brain. And they said, ‘Is there any way you can translate the system you’re using in intensive care to measure healthy four month old infants that we might want to investigate for autism?’
So the procedure started with us having lots of initial discussions to understand exactly what they needed from our optical imaging systems. We then went back to our engineering labs and we custom built a brain imaging system that was designed specifically to measure four month old infants. We then took this system back to the psychology lab and we piloted initially in completely healthy four month infants.
And so the protocol would be that the infant is presented with a range of different visual stimuli and all the time that they’re looking at these stimuli, we’re recording what’s going on in their brain. And this gave us the baseline information about how does a normal infant respond to social stimuli: when they see another human interacting with them, which part of their brain lights up. This gave us confidence first of all that our imaging technology was working in this group but also it was providing really useful information about the brain.
So we then upped the ambition level and said, ‘OK, this group of psychologists are interested not just in the normally developing brain, but also whether there are early indicators of when the brain may not be developing normally’. So we then did a study where we compared infants who were at low risk of autism with infants at high risk of autism. Now autism isn’t diagnosed until the second or third year of life so how would you know at four months of age whether you have an infant at high risk of autism? Well, the answer is that we selected infants who have a sibling with autism because they have an increased risk themselves of having autism.
So we repeated the study with these two groups of infants, we gave them identical visual stimuli and then we looked at the differences in the brain imaging data and we showed that infants with a high risk of autism had an absent response to these human, interactive, visual stimuli compared to the low risk infants.
Great achievements in understanding the brain
I think some of the greatest achievements in the last 50 years in terms of brain imaging have been the capability of looking at the brain almost when the brain doesn’t know it’s being imaged. So the idea that one of the challenges for medical physics is to measure something without changing it, and nowhere I think is that more true than when you’re looking at the brain. So the idea of looking at consciousness, the idea of looking at natural processes that we may not even be aware that we’re performing, which has enabled us to delve really into the inner workings of the brain in a way that no other imaging techniques have really been able to do.
Once you understand more about the human brain, it’s a ridiculous quest that you enter into because you…it’s so tantalising that there’s so much you’ll never know, but it’s only your own brain that’ll work it out. I find that circularity very interesting.
Now, the greatest challenges are really broadening the application of brain imaging technology to a much wider range of patient and subject groups. Currently things like MRI scanners are quite expensive to install and also to maintain but we can get lots of useful information from more portable systems like our optical imaging system. A really good example of that is a study that I’ve become involved in in the last year which has been funded by the Bill and Melinda Gates Foundation, and they’re interested in looking at brain development in children in developing countries who may have suffered malnutrition. There is a link between malnutrition and a deficit in brain development.
But there’s no way that we can put these babies into expensive scanners, they don’t exist in those settings. So we’ve taken one of our optical imaging systems and flown it out to a rural African field station and have made measurements of these babies’ brain development using this type of technology. And I think that’s a really interesting avenue of broadening the applicability of these brain imaging technologies.
I think one of the most startling images I have in my mind is the difference in the data that we got between these low and high risk infants. I must admit that I thought it may have been beyond the technology to pick up a subtlety as early as four months of age in these infants, differences in which they were processing the world around them, and it was extraordinarily exciting to see that data and to understand that all the engineering and the psychology in the interface in these disciplines had come together to enable us to get what was a really important insight into early markers of autism.
So one of the reasons you might be interested in consciousness is if you were undergoing an anesthetic. So what actually happens when you undergo an anesthetic to your consciousness? How do we capture the biological process that you’re undergoing when that anesthetic starts to take effect? How can we better develop anesthesia and how it’s applied to patients? Well, maybe, some of the answers may lie in us understanding consciousness in the active state and then we can know more about how to control it when we want to subdue it.