While 3D printers were once huge, expensive devices only available to the industrial elite, they have rapidly gained popularity over the last decade with everyday consumers. I enjoy printing a myriad of objects at the Duke Colab ranging from the Elder Wand to laptop stands.

One of the most important recent applications of 3D printing is in the medical industry. Customized implants and prosthetics, medical models and equipment, and synthetic skin are just a few of the prints that have begun to revolutionize health care.

3D printed prosthetic leg: “customizable, affordable and beautiful.”

Katie Albanese is a student in the Medical Physics Graduate Program who has been 3D printing breasts, abdominal skeletons, and lungs to test the coherent scatter x-ray imaging system she developed. Over spring break, I had the opportunity to talk with Katie about her work and experience. She uses the scatter x-ray imaging system to identify the different kinds of tissue, including tumors, within the breast. When she isn’t busy printing 3D human-sized breasts to determine if the system works within the confines of normal breast geometries, Katie enjoys tennis, running, napping and watching documentaries in her spare time. Below is the transcript of the interview.

How did you get interested in your project?

When I came to Duke in 2014, I had no idea what research lab I wanted to join within the Medical Physics program. After hearing a lot of research talks from faculty within my program, I ultimately chose my lab based on how well I got along with my current advisor, Anuj Kapadia in the Radiology department. He had an x-ray project in the works with the hope of using coherent scatter in tissue imaging, but the system had yet to be used on human-sized objects.

Could you tell me more about the scatter x-ray imaging system you’ve developed?

Normally, scatter in a medical image is actively removed because it doesn’t contribute to diagnostic image quality in conventional x-ray. However, due to the unique inter-atomic spacing of every material – and Bragg’s law – every material has a unique scatter signature. So, using the scattered radiation from a sample (instead of the primary x-ray beam that is transmitted through the sample), we can identify the inter-atomic spacing of that material and trace that back to what the material actually is to a library of known inter-atomic spacings.

Bragg diffraction: Two beams with identical wavelength and phase approach a crystalline solid and are scattered off two different atoms within it.

How do you use this method with the 3D printed body parts?

One of the first things we did with the system was see if it could identify the different types of human tissue (ex. fat, muscle, tumor). The breast has all of these tissues within a relatively small piece of anatomy, so that is where the focus began. We were able to show that the system could discern different tissue types within a small sample, such as a piece of excised human tissue. However, in order to use any system in-vivo, which is ideally the aim, you have to determine whether or not it works on a normal human geometry. Another professor in our department built a dedicated breast CT system, so we used patient scans from that machine to model and print an accurate breast, both in anatomy and physical size.


What are the three biggest benefits of the x-ray imaging system for future research? 

Main breast phantom used and a mammogram of that phantom with tissue samples in it

Main breast phantom used and a mammogram of that phantom with tissue samples in it

Coherent scatter imaging is gaining momentum as an imaging field. At the SPIE Medical Imaging Conference a few weeks ago in San Diego, there was a dedicated section on the use of scatter imaging (and our group had 3 out of 5 talks on the topic!). One major benefit is that it is noninvasive. There is always a need for a noninvasive diagnostic step in the medical field. One thing we foresee this technology being used for could be a replacement for certain biopsy procedures. For instance, if a radiologist finds something suspicious in a mammogram, a repeat scan of that area could be taken on a scatter imaging system to determine whether or not the suspicious lesion is malignant or not. It has the potential to reduce the number of unnecessary invasive (and painful!) biopsies done in cancer diagnosis.

Another thing we envision, and work has been done on this in our group, is using this imaging technique for intra-operative margin detection. When a patient gets a lumpectomy or mastectomy, the excised tissue is sent to pathology to make sure all the cancer has been removed from the patient. This is done by assessing whether or not there is cancer on the outer margins of the sample and can often take several days. If there is cancerous tissue in the margin, then it is likely that the extent of the cancer was not removed from the patient and a repeat surgery is required. Our imaging system has the potential to scan the entirety of the tissue sample while the patient is still open in the operating room. With further refinement of system parameters and scanning technique, this could be a reality and help to prevent additional surgeries and the complications that could arise from that.

What was the hardest or most frustrating part of working on the project? 

We use a coded aperture within the x-ray beam, which is basically a mask that allows us to have a depth-resolved image. The aperture is what tells us where the source of the scatter came from so that we can reconstruct. The location of this aperture relative to the other apparatus within our setup is carefully calibrated, down to the sub-millimeter range. If any part of the system is moved, everything must be recalibrated within the code, which is very time-consuming and frustrating. So basically every time we wanted to move something in our setup to make things better or more efficient, it was like we were redesigning the system from scratch.

 What is your workspace like?

Katie and the team at the AAPM (American Association of Physicists in Medicine) conference from this past summer in Anaheim, CA where she presented in a special session on breast imaging. From left to right: Robert Morris (also in the research lab and getting his degree in MedPhys), Katie, Dr. James Dobbins III (former program director and current Associate Vice Provost for DKU) and Dr. Anuj Kapadia, my advisor and current director of graduate studies in the program

Katie presented in a special session on breast imaging at the American Association of Physicists in Medicine conference this past summer in Anaheim, CA. From left to right: Robert Morris, also working in the lab; Katie; Dr. James Dobbins III, former program director and current Associate Vice Provost for Duke-Kunshan University; and Dr. Anuj Kapadia, Katie’s advisor and current director of graduate studies.

We have a working experimental lab within the hospital. It looks like any other physics lab you might come across- messy, full of wires and strange electronics. It is unique from other labs within the Medical Physics department because a lot of research that is done there focuses on image processing or radiation therapy treatment planning and can be done on just a computer. This lab is very hands-on in that we need to engineer the system ourselves. It is not uncommon for us to be using power tools or soldering or welding.

What do you like best about 3D printing? 

3D printing has become such a great community for creativity. One of my favorite websites now, called Thingiverse, is basically a haven for 3D printable files of anything you could ever dream of, with comments on the best printing settings, printers and inks. You can really print anything you want — I’ve printed everything from breasts, lungs and spines to small animal models and even Harry Potter memorabilia to add to my collection. If you can dream it, you can print it in three dimensions, and I think that’s amazing.


Anika_RD_hed100_2By Anika Radiya-Dixit