Several engineers and doctors at MIT and Boston Children's Hospital have been working on a method that strings together medical imaging, image recognition software and 3-D printing to give doctors vivid models of human hearts, particularly those that have been shaped by congenital defects.
The underlying assumption that the team is trying to test is whether surgeons might find planning surgeries a bit easier if they have a physical model of the organ they can hold in their hands, or even cut into.
The process has three steps. First, clinicians take MRI or CT scans of the patient's organ. Then they convert those images into a set of blueprints a 3-D printer can read, and finally the printer builds the model.
The method could even be used to develop a virtual model of a heart that would display either on a computer screen, or holographically, allowing the surgeon to perform a kind of virtual "practice run," to test different surgical approaches.
The researchers have focused a lot of energy on tackling one of the biggest bottlenecks in the process of converting MRI or CT scans into a model: Shrinking the amount of time it takes to convert images into a model that a 3-D printer can read.
An MRI or CT scan is actually a three-dimensional image that is a stack of a hundred or more two-dimensional slices. Surgeons use MRIs all the time to plan surgeries. They typically have to look at the images and reconstruct a three-dimensional model of the organ in their heads—not an easy task even for the most experienced practitioners.
"No surgeon is bad at doing this, but some may be better at it than others," said Andrew Powell, a pediatric cardiologist at Boston Children's.
The heart is an especially tricky organ with which to work. Even normal human hearts are complex organs, with different chambers and valves and a lot of fine detail. Hearts with congenital defects are not only different from normal hearts, but they are different from each other. Even hearts diagnosed with the same condition often have differences among them, Powell said.
This is a problem for surgeons attempting to choose how best to go about operating, and how the operation might affect the patient.
It is even tougher for computers, which are generally not as good at image recognition as the human brain is. They have a hard time recognizing the boundaries between different objects in a picture—such as the border between one chamber in the heart and another.
This means that a person has to sit down at a computer and trace the outlines of different areas, called "segments" on an image—in this case on one of the hundred or more slices that make up an MRI scan or CT image. It would take several hours at least for a person to "segment" a single MRI scan.
An algorithm can do this much faster, but it would be far less accurate.
Danielle Pace and her adviser, Polina Golland at MIT, have been working on that piece of the process. They developed an algorithm that works with a user—a clinician sits in front of the computer, segments a portion of all of the slices in an MRI scan, and then the algorithm looks at what the user has done and uses the information to segment the rest.
Pace's algorithm can produce an image that is 90 percent accurate using only 14 user-annotated slices out of 100 or 200 total slices. Even manually segmenting three slices produced a model that was 80 percent accurate.
Pace said they have managed to get the amount of time for the process of converting the images from MRI to 3-D blueprint down to about two hours—about one hour for the user to manually segment the slices, and another for the algorithm to do the rest. The goal is to cut those two hours down to 30 minutes.
That shaves anywhere from two to eight hours off the total time, depending on the complexity of the job and the experience of the user.
It means that the process can create a 3-D replica of a heart within 24 hours, making something that could be practical in a hospital setting, Pace said.
The team will present its results at the International Conference in Medical Image Computing and Computer Assisted Interventions in early October in Munich. They are also planning a clinical trial in which several surgeons at Boston Children's will use the process and determine if it helps them better plan out their surgeries.
Once the process is perfected for the heart, it could be used on any part of the body, said Dr. Mehdi Hedjazi Moghari, a cardiovascular magnetic resonance scientist at Boston Children's who is also working on the project.
"I think we have already addressed the most challenging part of the body to image, which is the heart," Moghari said. The heart beats regularly, he said, meaning it is almost always in motion. "Other organs are static. So it is very easy to acquire images for other organs such as the brain and segment them, because they don't move the way the heart does."
This group is far from being the only one exploring the applicability of 3-D printing to medicine.
"It would be unfair to say this is the first effort of its kind," said Dr. Andrew Powell, a pediatric cardiologist at Boston Children's. Other research groups have been working on using 3-D printing to build organs, he noted, and Powell said some are exploring the possibility of printing implants.
But this team wants to turn something that is possible in a lab into something doctors can use in the clinic, when time is short.
"We are trying to make it more practical, and streamlined so we can do it for more and more and more patients," he said.
Powell has said that most of the surgeons to whom they have introduced the idea seem excited, while reserving what Powell says is "an appropriate amount of skepticism."
"This is no question that this will eventually be very helpful," said Gosta Pettersson, vice chair of thoracic and cardiovascular surgery at the Cleveland Clinic. "The more information we can get about the heart, the better. There are so many kids that are born with complicated anomalies, the more help we can get the better operationwe can perform."
Pettersson noted that medical imaging technology has improved a lot since he began practicing medicine and has recently been introduced to a number of technologies that take imaging to the three-dimensional level. One company demonstrated a program that could holographically render images of the heart that surgeons could then manipulate in the same room.
He said that he and his colleagues will definitely be interested in getting access to these tools, which he thinks may be available in the near future. He said the fact that the team has been able to reduce the processing time is impressive, but the appeal of these tools will ultimately come down to how precise they are.