Andrew Powell, MD, Senior Associate in Cardiology and Advanced Cardiac Imaging
Arguably, one of the most difficult tasks facing a cardiac surgeon is visualizing what a patient’s diseased heart will look like once it is finally exposed in the operating room. Relying on two-dimensional image data from echocardiograms, x-ray angiograms, CTs and MRIs, they must mentally construct a 3D model to imagine how the heart looks and functions in the physical world.
Ideally, surgeons would be able to work with an actual three-dimensional model to prepare for a complex procedure. Technology now makes this scenario a reality.
Andrew Powell, MD, Senior Associate in Cardiology and Advanced Cardiac Imaging, says that over the past ten years, significant progress has been made in 3D heart model production. However, creating either a virtual or physical 3D model of a patient’s heart is still quite cumbersome and time-consuming. The process has not yet been simplified enough for widespread adoption in clinical practices.
Powell and his research collaborator Mehdi Hedjazi Moghari, PhD, are working to streamline the system without sacrificing accuracy. “Eventually, we want to be able to do this on a rapid and routine basis,” says Powell.
Constructing a 3D heart model involves three steps:
Step 1: Obtain the 3D image data. This can be done with MRI, CT and 3D echocardiography.
Step 2: Identify the borders of the important heart structures (“segmentation”). This step involves separating out the myocardium, blood pool and great vessels.
Step 3: Render a virtual heart model and print it using a 3D printer.
Currently, Step 2 can take many hours, particularly when accuracy is at a premium. Powell and Moghari want to shorten that time to under an hour without compromising accuracy. Their goal is to make it possible for doctors to perform an imaging study and then have the 3D printed model available by the next day.
In the first stage of their research, Powell and Moghari are focusing on the diagnosis double outlet right ventricle. This disease is a good candidate for 3D modeling because it has anatomic variability that impacts the surgical strategy for repair. A 3D heart model would allow the surgeon to trial various repair strategies before the surgery, which could potentially lead to better decision-making and patient outcomes.
Powell also sees enormous potential in using 3D imaging technology as an education tool. He currently instructs Cardiology fellows and knows how important it is to be able to conceptualize the heart’s structure when it comes to complex congenital defects. The ability to show trainees cardiac anatomy in 3D without having to rely on pathology specimens should be extremely helpful.