In 2009, research scientist Justin Ryan learned firsthand the limits of a Microsoft Excel spreadsheet-based quality management system (QMS). At the time, he worked in Phoenix Children’s Hospital’s 3D printing lab.
“We were trying to track everything on the patient side of things, and then we tried to track the 3D print consumables and the 3D print itself and the maintenance on the machine,” he recently told attendees at the 2021 MD&M West conference in Anaheim, Calif. “So, we started to scale up. And we had an Excel sheet with 100-plus columns and it wasn’t pragmatic anymore.”
Ryan commented as a panelist for a discussion on AM for end-use device development. He’s moved on from the Arizona hospital and is currently director of the Helen and Will Webster Foundation 3D Innovations Lab at Rady Children’s Hospital in San Diego.
More than a decade after the unwieldy, spreadsheet-based QMS at Phoenix Children’s, Ryan said direction and support for the quality aspect of his business still is lacking from regulators and software vendors.
“We are looking to see what does the FDA says in that conceptual framework when it becomes something more than just a concept, what types of QMS do we need to specifically put in place?” he said. “That might still be homegrown because there’s still this lag of software addressing our specific needs as a hospital-based, point-of-care manufacturer.”
The U.S. Food & Drug Administration issued a guidance document in 2017 on 3D printing, but it didn’t include point-of-care manufacturing like that done in hospitals, FDA research scientist Matthew DiPrima said in a separate interview.
Sean McEligot, who is section head of medical device research and development at the Mayo Clinic and who also spoke on the MD&M West conference panel, said his site also home grew its QMS because the off-the-shelf software is tailored to large-scale manufacturing and not the “onesies” and “twosies” they do.
That’s expected to change, though, since Mayo acquired a titanium printer. Its intent is to move beyond making anatomic models and lab instruments to fabricate patient-specific, 3D-printed titanium implants. The U.S. FDA told Mayo it would be a medical device manufacturer if that happens.
“As we move forward looking at becoming a registered device manufacturer and at the likelihood of audits, we’re going to have to up our game, and really focus more on making sure that everything, every ‘i’ is dotted and every ‘t’ is crossed,” he said.
Quality management is just one of two gaps where software vendors can help end-use device developers.
They also need more options for FDA-certified segmentation software.
These applications enable a user to segment the data in a CT or MRI scan DICOM file—separating surrounding tissue from bone, for example—and choose only the information needed for a 3D printing STL file.
“There are very few players in the game right now,” said panelist Cambre Kelly, VP of research and technology at device maker Restor3d. “So, I’m hopeful that we will start to see new packages and new features coming out in the next couple years.”
In the meantime, Restor3d has started looking at writing some of the software in-house for its very specific workflows, she said.
Moderator Gaurav Manchanda, director of medical market development at 3D printer maker Formlabs, said, “I think there are three vendors with [FDA] cleared packages for diagnostic use medical models—Materialise, 3D Systems, Synopsis is the latest, and there are a few more coming.”
Ryan agreed with Kelly on the need for more segmentation software vendors.
“Certainly, from a hospital perspective, I hope we see more software that produces more competition and we see some prices drop,” he said. “Because right now, it is very costly for a nonprofit hospital to buy FDA-cleared software.”
The panel turned to the topic of finding the right people to use the software and the printers to create new options for limb salvage and reconstruction, in Restor3d’s case, and separating conjoined twins, which McEligot and colleagues supported at Mayo, among other more frequent uses.
“I actually went onto Reddit and tried to hire a software engineer that I could teach 3D printing to,” said Ryan. “I wanted a software engineer to learn how to run machines but deliver some new resources to us,” for example, virtual and augmented reality.
He said the desired skills for potential hires include 3D printing and basic engineering design principles, including CAD and segmentation.
“It would be great coming out of a radiology tech school knowing how to do 3D segmentation in a very high-level, efficient way,” he said. “We’re seeing some steps towards that space.”
Kelly said it’s a huge plus if candidates have had some exposure to the different software packages that Restor3d uses for anatomic modeling.
“Just even knowing what tools are out there and available to them to be able to dig their teeth into,” she said. “Same thing on the printing.”
McEligot said his lab at Mayo is fortunate to be partnered with the Radiology department at Mayo, where the biomedical engineers are expert in using segmentation software. His challenge lies with hiring with an eye toward being a full-blown medical device company.
“I’d love a dual major in mechanical and biomedical [engineering],” he said. “They’d come with those SolidWorks (CAD-CAE software from Dassault Systemes) skills or something equivalent, and then, as with medical device companies, it’s on-the-job training.”
That OTJ training includes biocompatibility, sterilization, 3D printing, image analysis, and segmentation software.
Right now, one of McEligot’s best employees is an intern.
“He ran a 3D printing lab at the University of Wisconsin-Madison and has taken over the AM facility with a storm, and it is bringing all kinds of new and good ideas to us,” he said. “So, talent comes from everywhere.”
Connect With Us
