3D Printed Medical Models: There Is a Growing Need for Quality Assurance in Production

Infographic overview of 3D printed model / part verification

Researchers from around the US have convened to analyze the uses of 3D printed medical models further. In their recently published paper, ‘Methods for verification of 3D printed anatomic model accuracy using cardiac models as an example,’ authors Mohammad Odeh, Dmitry Levin, Jim Inziello, Fluvio Lobo Fenoglietto, Moses Mathur, Joshua Hermsen, Jack Stubbs, and Beth Ripley appreciate the usefulness of 3D printed medical models, but they also realize a growing need for quality assurance in such products—and especially as popularity for their use continues.

The authors research different methods for examining quality of medical models through:

• Physical measurements
• Digital photographic measurements
• Surface scanning
• Photogrammetry
• Computed tomography (CT) scans

In verifying quality assurance, models can be tested for accuracy in demonstrating patient-specific health conditions, along with ensuring they will serve their intended purpose for the patient. The authors also state that there can be challenges in verification, to include difficulty in measurements and obtaining the desired dimensions.

The models were created at the University of Washington School of Medicine to evaluate the usefulness of cardiac models in pre-surgery planning. Some of the models were subjected to additional QA testing to help set standards for verifying such models. Data was exported from Mimics Medical into 3-Matic Medical Software (Version 13.0) for editing, and then sent back to Mimics software for final verification. All medical models were 3D printed on a Form2 in clear, white, and gray resins.

Physical measurements of 3D printed cardiac models and cadaveric hearts. a-c Three examples of cadaveric hearts and their corresponding 3D printed models, used for measuring features of the aortic valve (a, b) and the mitral valve (b, c). d-e Measurement of perimeters was accomplished using a malleable metal wire which was marked with a needle driver and then straightened and measured against a ruler. f Organic shapes present challenges to measurement with linear rulers or calipers

The authors measured for discrete features on the models themselves, then on the .stl files, and then compared measurement styles. They also explored photogrammetry and surface scanning of 3D printed models, CT scanning of the models, and alignment of patient DICOM and 3D model DICOM datasets. While there were some challenges in attaining physical measurements, as the researchers state, it takes years to establish a good quality assurance program, and this study just focused on part verification.

Digital photographic measurement technique. a-d Four discrete 3D model features (a Feature 1, b Feature 2, c Feature 3, and d Feature 4) were chosen for measurement, based on their accessibility on an exterior surface and the fact that they had distinguishing features that would allow for repeat measurements. All digital photographs were calibrated before measurement. This was achieved by placing a marker of known length at the same height and angle as the feature of interest before photographing the model. Once the photograph was imported into an image analysis software (in this case, ImageJ), the length of the known marker was measured in pixels, and a conversion factor was calculated for pixels to millimeters. Of note, insufficient illumination for feature 3 (panel c) caused blending of edges, most pronounced along the inferior aspect of the photograph. This affected accuracy of measurements for this feature

As the researchers learned, 3D printing medical models can introduce errors at ‘each step in the process.’ Their solution is to monitor the process, creating checkpoints in the design process and printing. Each form of measurement has both pros and cons, and the researchers suggest that verification methods should be fitted to the patient specific requirements of the model being verified.

“The choice of which method to adopt into a quality assurance program is multifactorial and will depend on the type of 3D printed models being created, the training of personnel, and what resources are available within a 3D printed laboratory,” concluded the authors.

Medical models are changing the face of medicine for everyone involved, causing a positive trickle down effect as medical professionals convert data into models to be 3D printed, doctors and surgeons are able to make more accurate diagnoses, establish treatment plans, and train for some procedures that may be very rare, medical students are able to learn, and patients and their families are able to be better educated on what is happening through such a progressive visual aid. Over time, medical models have helped with improving patient care further, helping doctors fight cancer, and even veterinary care. What do you think of this news? Let us know your thoughts! Join the discussion of this and other 3D printing topics at 3DPrintBoard.com.

Digital photographic measurement technique. a-d Four discrete 3D model features (a Feature 1, b Feature 2, c Feature 3, and d Feature 4) were chosen for measurement, based on their accessibility on an exterior surface and the fact that they had distinguishing features that would allow for repeat measurements. All digital photographs were calibrated before measurement. This was achieved by placing a marker of known length at the same height and angle as the feature of interest before photographing the model. Once the photograph was imported into an image analysis software (in this case, ImageJ), the length of the known marker was measured in pixels, and a conversion factor was calculated for pixels to millimeters. Of note, insufficient illumination for feature 3 (panel c) caused blending of edges, most pronounced along the inferior aspect of the photograph. This affected accuracy of measurements for this feature