Researchers at Georgia Tech and Emory University have engineered an innovative computer program to help cardiac surgeons optimize surgical procedures before they enter the operating room. The image-based surgical-planning software lets surgeons manipulate a three-dimensional computer model of a patient’s heart. Once the doctor has altered the model so that it includes the desired vascular configuration he or she wants to create during surgery, the program uses computational fluid dynamics to run a blood-flow simulation displaying how well the modified heart will perform.
The aim of the project is to develop a complete system that addresses the demanding needs of cardiovascular surgery planning and assessment, says Ajit Yoganathan, the principal investigator on the project and associate chair of the Department of Biomedical Engineering at Georgia Tech and Emory University. The program was built to test one of the most common and complex congenital heart problems known: a single ventricle defect.
Children with this condition have only one heart ventricle (the left), instead of two, for pumping oxygenated and deoxygenated (dirty) blood throughout the body. In healthy patients the right ventricle pumps the deoxygenated blood through the arteries to the lungs, while the left ventricle receives the oxygenated blood through the veins and shoots it off to every organ in the body.
“The mixture of blood in the single ventricle greatly compromises circulation throughout the body,” says Shiva Sharma, a private pediatric cardiologist. “The surgeon’s job is to separate the circulation, meaning rerouting the deoxygenated blood directly and evenly to the lungs while minimizing resistance to flow, an operation called Fontan repair.”
Designing the best connection is essential because too much resistance can increase blood pressure and cause a variety of life-threatening complications, explains Pedro del Nido, chief of cardiac surgery at Children’s Hospital of Boston. “Surgical procedures are based on a surgeon’s personal experience, experimentation, and, frankly, a lot of trial and error. There is no direct way of knowing whether or not we have made things better or if a slight variation in our technique will make a slight difference or not.”
Before image-based surgical planning, surgeons worked somewhat like “freehand artists” in that they would look at the anatomy and then sketch out a plan for surgery, adds Sharma.
The program developed by Yoganathan and his colleagues works by creating a three-dimensional computer model of the heart using data from the child’s magnetic resonance imaging (MRI) scans at different times in the cardiac cycle. After viewing the images and devising a few plans, the surgeon sits down at a computer and manipulates the model using “input devices” that look like scalpels, explains Yoganathan.
“What we have developed is a system that lets you grab the geometries as if you were holding them in your hands and then twist, rotate, and move them as three-dimensional models in front of you,” says Jarek Rossignac, one of the system’s designers and a professor at the College of Computing at Georgia Tech. “The result is a new three-dimensional model that reflects a particular shape that a surgeon envisions for the operation.”
The new anatomically modified three-dimensional model is seamlessly exported and meshed for a computational fluid dynamics analysis (CFD). Using CFD creates a simulation of blood flow in the newly configured heart that can be viewed by the surgeon on the screen. After multiple mock models are designed and tested, the surgeon can decide which operation proved optimal for that particular patient. Thus far, the hearts of five patients have been designed and tested for surgery using the three-dimensional model.
At the moment, the system is only being used by a small group of surgeons involved in the research. Yoganathan says the technology is three to five years from being ready for general use, and there are still some challenges to overcome. The flow dynamics, he says, are very computer intensive and involve complex formulas. Converting the geometries back into computational meshes is “painstakingly slow.”
Right now, the computer engineers have the surgeons draw out the design and then the engineers manually enter the geometries. It’s a “very cumbersome process,” says Yoganathan. “We are working on developing tools that, once the geometry is drawn, the computational mesh for analysis would be done automatically so there is no engineer involvement.”
There are also no precise mathematical formulas for anatomical shapes, which are organic and have interesting material-property issues, so mimicking how they might evolve represents a new set of challenges, explains Rossignac.
The researchers want to provide “user-friendly” human-shape interface technology, which would let surgeons manipulate shapes in an intuitive, efficient way; right now it takes the surgeons two to three hours to manipulate a patient’s heart to the configuration they visualize. According to del Nido, however, these shape-editing technologies “are pretty simple to use and intuitive for anyone who has done computer games.”
Eventually, the researchers want the software to provide the optimal solution for the cardiac problem.
A group at Stanford University, led by Charles Taylor, is also working on image-based surgical-planning systems. Recently, his lab opened the Center for Simulation in Medicine at Stanford Hospital to focus on surgical planning for cardiovascular interventions for children with congenital heart disease and adults with atherosclerosis and aneurysms.
According to Taylor, simulation-based planning of cardiovascular treatments could lead to lower morbidity and mortality, reduced reoperative rates, and reduced time in the operating room.
Ultimately, image-based surgical planning will have an immense impact on surgical procedures, improving the quality of life of not just children but adults as well, says del Nido.
DeepMind’s cofounder: Generative AI is just a phase. What’s next is interactive AI.
“This is a profound moment in the history of technology,” says Mustafa Suleyman.
What to know about this autumn’s covid vaccines
New variants will pose a challenge, but early signs suggest the shots will still boost antibody responses.
Human-plus-AI solutions mitigate security threats
With the right human oversight, emerging technologies like artificial intelligence can help keep business and customer data secure
Next slide, please: A brief history of the corporate presentation
From million-dollar slide shows to Steve Jobs’s introduction of the iPhone, a bit of show business never hurt plain old business.
Get the latest updates from
MIT Technology Review
Discover special offers, top stories, upcoming events, and more.