Virtual heart may hold key to better treatment
Jack and his parents live on the east coast of England. Every now and then they come to London, trying to make the most of their day out. But the reason for their regular visits beats inside Jack’s little chest.
“We found out about that condition when I was about 28 weeks pregnant – they told us unfortunately that’s hypoplastic left heart syndrome, and it was very complex, and said it would involve a three-stage palliative surgery. So it was quite a shock,” says Jack’s mother Sarah.
After three open heart operations, Jack needs routine check-ups just to make sure there have been no negative developments. Here, at the Evelina Children’s Hospital, Doctor Razavi monitors Jack using sophisticated diagnostic equipment.
“When he puts the stickers on, it feels like I’m doing nothing, I’m just relaxing!… It does this: boom boom boom boom… But I don’t know how it does that thing.”
Ultrasounds can be used to make an image of the heart, painted in real-time by tissue and blood reflecting the sound waves. Echocardiography can provide a wealth of helpful information, giving therapists a quick look at how well a patient’s cardiovascular system is functioning. It’s a vital tool, considering the severity of the condition, says the Head of the Imaging Sciences Division at King’s College London, Dr. Reza Rezavi: “Congenital heart problems altogether affect about one in 100 children, and heart disease in the general population is of course very common: it’s the most common cause of illness and death, certainly in the Western world, and it’s becoming the most common cause of illness and death in the developing world.”
However useful, echocardiography and other imaging methods have their limits. Overcoming them is the goal of a European research project
“The model that’s now on the screen is just a picture so you can see how big the blood vessels are, how things are connected to each other, but it doesn’t actually tell us how the heart’s beating in this particular model. And we hope, if we have a model where we can input anatomical data and some data on how the heart itself is contracting, we’ll be able to better tell which children’s heart is going to run into trouble sooner, and also we’ll be able to try some medications on a virtual model, rather than just trying medications on children,” says Hannah Bellsham-Revell, clinical research Fellow in Paediatric Cardiology at the Rayne Institute.
“What we want to be able to do is to see what happens after treatment, particularly if the treatment is quite difficult to give, or very expensive – we want to know before we give it whether it’s going to work or not. And for that, computer models give us great versatility – they allow us to try any treatments and to see what will be the result, without having to really go through that in the first place, just doing it ‘in silico’ – on the computer,” adds Reza Rezavi.
So, a realistic model should integrate all the relevant medical data. But first, how do we collect it? Jürgen Weese is research Fellow in cardiovascular interventional solutions at Phillips research.
“We have different scanners that you can use to make images of the heart: what you see here is one that uses a magnetic field. What is specific about this scanner is that you can get information about the heart motion, and information about the heart muscle’s tissue properties, and you can also image the blood flow in the heart.”
Medical scanners produce multiple digital images of thin slices of the heart. With the right software, it is pretty easy to assemble a three-dimensional object from a series of cross-sections stacked together. Some of the modern scanners can do that with a single press of a button.
“What you see here is the CT scanner: with this type of scanner you use x-ray radiation to make an image of the heart, so you make x-ray projections from all different sides and compose a 3D image out of all the single images. What is particular about this scanner is that you get a very high resolution, very good images of the coronary, of the heart,” says Weese.
To make the step from visualization to modelling, scans need to be interpreted by a computer. Algorithms are being developed here, at Eindhoven’s High Tech Campus. Every heart is unique, so the program measures all its peculiarities, accurately mapping the individual patient’s organ. But at this point it’s just a draft, not a working model. The aim is to go further, says Weese: “This technology allows us to only extract geometry from the images. It doesn’t predict how the heart will beat in the future if you change something. If you want to predict how the heart will react if you prescribe a treatment or if you change the blood flow in some part, then you need other models that describe the biophysics of the different parts of the heart. Only then can you predict how the heart will work after therapy or decide between different treatment options.”
This is where engineering lends a helping hand. The heart is basically a very efficient pump. Its geometrical map is a kind of blueprint of a complex machine, and we already have the knowledge and experience of testing machines and other artificial objects in computer simulations. Nic Smith is Professor of Computational Physiology at Oxford University: “These techniques have been used to analyse bridges and structures in the 1950s, 60s and 1970s, and about that time we started to have the computing power that allowed us to move into locomotive and aerospace, and airplane design. And what really happened most recently is the same kind of techniques are now being applied to some of the most complex problems that we have in biology, physiology and medicine.”
The result is a virtual heart that works just like its real counterpart, showing, for example, cells that don’t perform well enough, or allowing the study of the propagation of the electrical wave in someone’s heart muscle before implanting a pacemaker. The result is not only visual says Smith: “What we’re achieving is the ability to predict patient behaviour to the therapy and to select patients based on what we see in their particular anatomy for more appropriate therapies. I think that’s within a wider goal of moving medicine from a practice which is based on presentation and trial and error, to one which is based on science, where we understand what the mechanisms are.”
Years of research and clinical validation are still needed before a virtual heart can become a reliable prognostic tool, giving new hope to those who need it the most. That’s not going to be soon enough for some regrets Dr. Rezavi: “Children like Jack, although they’re quite well as children, when they reach adulthood they’re likely to have quite a lot of difficulties, and their life expectancy is much less than in general population. We’re of course trying to find new ways of treating them to improve their life expectancy, and this project substantially helps us along that way.”
The last word, of course, should go to Jack himself: “I want to become a doctor when I grow up! Because it’s a big fun. He can see all other people’s hearts.”
But it is by letting others look into his that Jack is already, aged six, making a vital contribution to the medicine of the future.