FDA’s initial thoughts on 3D printed medical devices

The U.S. Food and Drug Administration (FDA) has recently released a draft guidance for the 3-D printed medical devices sector. The agency took into consideration inputs from device manufacturers, 3-D printing companies, and academics who testified at a 2014 hearing. According to QMED, the document covers device design, manufacturing, and design testing and for the purposes of the draft, FDA identified four main types of 3-D printing—powder fusion, stereolithography, fused filament fabrication, and liquid-based extrusion.

Source: QMED

3D printed medical device producers would have to “clearly identify every step in the 3-D printing process, and might need to submit a ‘high-level summary of each critical manufacturing process step,’” the guidance says. They would also have to record each step’s risk, and describe how they would lessen those risks. “The type of testing data needed would depend upon whether the device is an implant, load-bearing, and available in standard sizes or custom-made for each patient, or as FDA put it ‘patient-matched.’”

The implications for 3-D printed devices are enormous – therefore the draft guidance is the agency’s initial thoughts on the technical considerations surrounding the design, manufacture and testing of 3-D medical devices, which have few precedents. “While this draft guidance includes manufacturing considerations, it is not intended to comprehensively address all considerations or regulatory requirements to establish a quality system for the manufacturing of your device,” the agency said.

The draft guidance “provides a solid basis for medtech innovators to understand what is needed to prove safety, efficacy, and consistency to growing on-demand components for the human body,” said Derek Mathers, an adjunct professor of 3-D printing at the University of Minnesota and business development manager at Worrell Design in Minneapolis.

     “Standard-sized 3-D printed devices are offered in discrete sizes, and include features that are too complex to be manufactured with traditional processes like machining and molding,” he explained. “Patient-matched 3-D printed devices are devices that are digitally scaled (manually or by using an algorithm) to match a patient’s specific anatomical features. The FDA identifies that these bespoke devices will require significantly more validation work across every step of the ‘scan-to-fit’ design process.”

Artificial intelligence beating humans in detecting illness

    According to a recent report in Kurzweil Accelerating Intelligence, Samsung Medison has updated its RS80A ultrasound-imaging machine with a feature called S-Detect for Breast that analyzes breast lesions. The system uses big data collected from breast-exam cases and suggests whether the lesion is benign or malignant.

Source: QMED

     "We saw a high level of conformity from analyzing and detecting lesion in various cases by using the S-Detect," said professor Han Boo Kyung, a radiologist at Samsung Medical Center. "Users can reduce taking unnecessary biopsies and doctors-in-training will likely have more reliable support in accurately detecting malignant and suspicious lesions."

   Researchers at the Regenstrief Institute and Indiana University School of Informatics have also noticed that computers are better than humans in detecting cancer. According to their research, existing algorithms and open-source, machine-learning tools are as good or even better than human when detecting cancer cases using data from free-text pathology reports. The electronic approach was also faster and used fewer resources than people, according to Regenstreif.

    "We think that it’s no longer necessary for humans to spend time reviewing text reports to determine if cancer is present or not," said author Shaun Grannis, MD, interim center director. "We have come to the point in time that technology can handle this. A human's time is better spent helping other humans by providing them with better clinical care."