Accurately detecting heart defects in unborn children has long been a challenge for doctors, even in the most advanced hospitals. But British researchers have developed an open source computer programme which can turn ordinary MRI scans into a detailed 3D image of the heart and blood vessels within a foetus.
The 3D model allows doctors to accurately map out and assess the severity of foetal heart problems, usually first indicated by an ultrasound.
Its development also highlights to policymakers the advantages of investing in healthcare engineering and health tech — showing how innovation can help patients with complex conditions and reduce the pressure on busy health systems.
Matters of the heart
Identifying anomalies, such as narrow blood vessels, in a foetus’s heart or cardiovascular system prior to birth can greatly improve the efficiency of a medical response.
But the ultrasound scans typically used to observe foetuses do not provide results detailed enough to properly study these parts of the anatomy, which measure just millimetres and move unpredictably with the baby.
MRI scans of the cardiovascular system require electrodes to be attached to the chest, which is impossible for unborn children.
But researchers from King’s College London and the Evelina London Children’s Hospital at Guy’s and St Thomas’ NHS Trust developed computer software in response to this problem.
The program can take a “stack” of MRI scans and put them together to form a three- dimensional image to be reviewed by specialists.
“You’re getting information of the same bits [of the scan], multiple times from different views,” Professor Reza Razavi, Vice President and Vice-Principal (Research) at King’s College London, said.
After being processed by the software “you end up with even a better resolution than the images from the stacks you had,” says Razavi. “The software builds a 3D shaped model from that information.”
This means doctors can then “be certain about diagnosis” of a foetal congenital heart condition and have a much greater understanding of its severity, says Razavi.
In some cases, the 3D imaging also led to new discoveries of heart disease that previous scans hadn’t picked up on.
The project was trialled on 85 expectant mothers — the procedure has been used more than 100 times since — and results were published in the medical journal The Lancet.
Razavi says the procedure will soon become a locally commissioned clinical service, with plans to make it more widely available across the country. There are also plans to train doctors from Canada and the Netherlands.
Boosting child health
While a relatively simple concept, the development has significant benefits for both patients and the healthcare systems that treat them.
“It’s about making a shorter stay in intensive care and improving the smoothness of the journey through,” says Razavi.
Planning for surgery within the hospital is made easier, reducing delays which can significantly impact the health of the new-born.
The babies who need surgery will also be better prepared for it. If doctors know in advance what a specific problem is, they can provide correct treatment immediately after birth, instead of waiting for clarification, according to Razavi.
“If they’re in good condition when they go into surgery, they have less complications and they have better outcomes,” Razavi adds.
In addition, doctors are able to brief families more fully on their child’s condition and whether it might need an operation. “That’s really important, you shouldn’t underestimate how much anxiety and uncertainty a family has if their baby has a problem it might die
from,” he says.
The benefits of the 3D imaging programme also demonstrate the wider advantages of healthcare engineering – the field which makes use of data, devices and software.
According to Razavi, this is primarily done in two ways, both can reduce resource pressures on healthcare systems.
Firstly, the technology improves the speed and accuracy of identifying patients who need treatment.
“If you make that decision early and quickly and in a way that maximises outcomes and value to patients, that can take out a whole bunch of costs and uncertainty and worry,” he says.
Secondly, it can help inform patients of the right treatment for them, reducing wastage, and deliver it in a more efficient way, such as using robots during operations.
“The impact can be just as good as a drug,” adds Razavi.
The relatively fast implementation of healthcare engineering — this study began in 2015 — is a stark contrast to drug development, which can take decades, as scientists try to understand new substances through numerous rounds of testing.
But healthcare engineering can be done in a “very effective, efficient way” when hospitals and universities collaborate, says Razavi.
In this field, problems are not defined by new engineering techniques – unlike discovering a new drug – but by clinical need which can be tackled by collaborative teams working through faster rounds of testing and refinement. There’s often less risk to patients as the techniques are not usually affecting their physiology, unlike a drug.
“You just get people working on it, then you get a solution, you test it, you improve it, and then you’re applying it to patients straight away,” Razavi says. “It’s a very different cycle of development.”
Razavi praises the policies encouraging “integration as much as possible between universities and hospitals” which led to the creation of academic health science centers. These institutions can have a “major impact” on healthcare innovation, says Razavi. “It’s a good story.” — Will Worley