Looking through the heart

You could insert a stent into a valve and simulate what will happen as blood tries to flow through — is the stent too big, tearing the tissue around it, or too small, causing blood to leak around its edges? You could simulate the effects of a heart attack or a medical procedure. Or you could use the simulation technology to create a unique, 3-D model of an individual patient’s heart, using data from a CT scan or M.R.I.

Those are some of the goals of a research project and simulation software created by a company better known for creating 3-D models of cars and jets.

Dassault Systèmes, a French company, creates complex 3-D simulations and models for industries including aerospace, architecture, consumer goods and finance — and, most recently, life sciences. About six months ago, the company introduced the Living Heart Project, and in late August released a beta version of its highly realistic 3-D simulation of the human heart. The company announced a five-year research agreement with the Food and Drug Administration that will focus on using the 3-D technology to simulate the reliability of pacemaker leads — the wires that deliver electrical stimulation from the pacemaker to the muscle of the heart — and try to determine how well they will perform.

“Thirty per cent of all people die from heart disease,” said Steve Levine, chief strategy officer at Dassault and director of the Living Heart Project. “We’re not winning the war. It’s getting worse. There’s a lot of cool stuff happening, but it’s not translating into better health and longer life.”

Problems with leads are among the most common issues with pacemaker surgery, either because the leads are placed in the wrong spot or because they become dislodged. In either case, more surgeries are needed to correct the issues. “Leads fail too often,” Steve said in an interview, “But the FDA has thrown their hands up and said, we don’t have any technology to figure this out. If plane wings failed as much as pacemaker leads, you can bet there would be a lot of attention to that.” He said the research project could evaluate a new lead design from the company by virtually sticking it inside a simulated heart to see what happens. “But all of the work gets done on the computer,” he said.

The company cites its experience in modelling complex systems like jet planes, cars and buildings, and Steve said Dassault was in a unique position to pull together members of the medical, manufacturing and regulatory communities to work on better computational modelling of heart issues. A small group of researchers,
clinicians and cardiac device makers have been experimenting with the simulation software, hoping to use it to create accurate models of heart problems, treatments and behaviours. Dassault Systèmes plans to eventually release commercial software that universities or cardiac device makers can use to develop better treatment methods for heart problems.

“This is the generic baseline heart; the next step is to be able to personalise,” Steve said. “Each heart has its own geometry and its own behaviour. A number of participants know how to do that. One colleague has made five personalised hearts and published them, from a research perspective.” There’s no guarantee that the device will offer a
perfect solution. Medical simulations have been tried before, including the Physiome Project, which aims to create an entire virtual version of the human body. But simulating a heart’s behaviour and scientifically validating that simulation — that is, proving that a device will behave the way the software says it is going to behave — can be a difficult leap.

Alan S Louie, research director at IDC Health Insights, part of the global research firm IDC, said he was briefed on the project about a year ago and thinks the Dassault heart model has potential because it is a more precise representation of a moving heart than existing cardiac simulations. “Validation is more of a question of how it’s used,” Dr Louie said. “If you want to use it as a training model for medical students to understand the heart, that’s fairly straightforward, because at the end of the day you don’t kill anybody.”

But he said he expected the specific applications to expand as more people try simulation. And computational modeling is considered a promising development in health sciences. The biggest hope is that it could shorten the time between a proposed health solution and when it becomes available to people who are actually sick.

“The overall goal of what they’re trying to do is a big deal, and there’s been a lot of work done in this field over the past 20 years,” said Thomas J Hund, a professor at the Davis Heart and Lung Research Institute at Ohio State University.
“Hopefully, this effort would be able to address some of the technical challenges: how long it takes to develop and optimise these models and how well we’re able to close the loop between our model and the actual intervention.” Steve, whose daughter was born with  a congenital heart defect that has left her with a pacemaker since she was 2,
said he hoped the Living Heart Project would evolve into something akin to the Human Genome Project. “We have the opportunity here to get a hybrid that works,” he said. “We know that if you simulate it here, it’ll be safe, but you can take that and do whatever you want from here.”