Understanding the Physics of Neutrons Can Be Key to Improving Hip Replacement Surgeries

Joyoni Dey, assistant professor in the LSU Department of Physics & Astronomy, works on ways to make medical imaging better. She will now apply methods she’d developed for X-rays to neutron beams, thanks to a $227,680 EPSCoR Research Infrastructure Improvement Track 4 grant from the National Science Foundation. Neutrons have an important advantage over X-rays in viewing bone and metal, especially where bone and metal meet. Her research aims to increase the success rate of hip replacement surgeries, the fastest growing elective surgery in the nation:

  

 

Dr. Dey in her office

Dr. Dey in her office.

Photo: Elsa Hahne/LSU

“First, let me tell you about X-rays. Ordinary X-ray machines will only find attenuation, or gradual absorption, of X-rays as they pass through the body. So, all you see is that difference in attenuation contrast. But as the X-rays move through the body, they’re also going to phase-shift and scatter. For X-ray scatter we mainly consider elastic small angle scatter (SAXS) and in-elastic Compton scatter. The Compton scatter in the context of X-ray interferometry basically just degrades the image contrast. The small-angle scatter, on the other hand, can give us a lot of information, and it would be great if we could capture all of this useful information—including the phase-shift and small-angle-scatter—and not just the attenuation.
 
My work here at LSU has been on developing these extra two modalities, phase-shift and small-angle scatter, and a novel phase-contrast X-ray system using special grating-based interferometry. We will potentially obtain these multiple-contrast (attenuation, phase-shift, and small-angle scatter) images using the same X-ray dose.
 
There is a big difference between X-ray and X-ray-interferometric imaging. If a patient stands in front of a regular X-ray machine, you’ll see “shadows” of the internal organs in their body. But it is different for grating-based X-ray interferometry. What you’ll see instead are interference patterns, which are attenuated, shifted, and locally distorted by the three phenomena of absorption, phase-shift, and small-angle scatter of X-rays as they travel through a patient’s body. And then, from those interference patterns in comparison with a blank reference scan without a patient, we can derive the attenuation, phase-shift, and scatter images.
 
In traditional X-ray attenuation images, a radiologist may not see much difference between a healthy patient and another with emphysema or cystic fibrosis. But if you look at a small-angle scatter image, the diseases are well-differentiated. These scatter images can also detect lung tumors much better than attenuation images. So, having both scatter and attenuation images will help. That’s where we’re heading. Hopefully, future X-ray machines will have all three of these complementary modalities.

 

If you toss two pebbles into a pool, you’ll see waves generated from each; constructive and destructive patterns on the surface. That’s interference. If you add a grating—sort of like a sieve pattern—in front of the neutron (or X-ray) beam, you get a corresponding expected pattern, depending on the distance. When you add an object, the pattern will change. It will attenuate, phase-shift, and locally distort (analogous to X-rays) the interference pattern. Again, (analogous to X-rays) we can derive attenuation, phase-shift, and elastic scatter from the object by comparing and analyzing the interference pattern with and without the object. We can then infer information about the object, and very accurately, at high resolutions. 

 

Now, to our main topic—neutrons. Neutrons can be defined as particles as well as waves, and you have similar absorption, phase-shifting, and elastic and in-elastic scattering effects. We’re mostly interested in the elastic scatter, as well as the attenuation and phase.
 
Neutrons have already been studied as particles, including how much they attenuate and scatter. Several computer simulation packages exist for simulated neutron particle beams. But we’re going to add the phase-shift to that mix, adding a phase analysis after a random or Monte-Carlo-based absorption and scatter analysis. So far, very few researchers have looked at the phase-shift of neutrons together with absorption and scatter. To do this, we first propose to make the neutron particle beam coherent, and then for each particle, calculate the path-lengths between each interaction and use these path-lengths to compute the phase shifts of the waves.
 
To summarize, my team’s contribution here will be to add the phase-shifting to the already known simulators for attenuation and scatter. This will be useful for the National Institute of Standards and Technology (NIST), Oak Ridge National Laboratory, and neutron researchers at other places.
 
If you toss two pebbles into a pool, you’ll see waves generated from each; constructive and destructive patterns on the surface. That’s interference. If you add a grating—sort of like a sieve pattern—in front of the neutron (or X-ray) beam, you get a corresponding expected pattern, depending on the distance. When you add an object, the pattern will change. It will attenuate, phase-shift, and locally distort (analogous to X-rays) the interference pattern. Again, (analogous to X-rays) we can derive attenuation, phase-shift, and elastic scatter from the object by comparing and analyzing the interference pattern with and without the object. We can then infer information about the object, and very accurately, at high resolutions.
 
With this new EPSCoR grant, we will collaborate with NIST and their Center for Neutron Research in Maryland. I will be going up there soon (March/April) and then for the whole summer, and one of my students will also work on the project and travel there. Hopefully, it will amount to his PhD work.
 
So, why do we want to study neutrons? They probe deeper. They interact relatively weakly with metal compared to X-rays. They interact strongly with hydrogen or oxygen. For medical applications, this is useful, as neutrons will be better for imaging bone-metal joints, where X-rays would lead to strong metal artifacts. Neutrons give high-quality information about surfaces, and they are very useful for looking at bone-metal implants, such as hip replacement joints. We hope to get samples from the LSU Vet School. Now, I have to be clear—you cannot use neutrons for imaging patients due to high radiation concerns. It will not be used for in-vivo imaging anytime soon. But materials scientists can use neutrons and neutron interferometry to look at tissues and metal-tissue interfaces ex vivo; and this could teach us a lot about metal implants in the body.
 
There are three key benefits of the work we’re doing: One is determining the grating and detector distances that will bring the best visibility and sensitivity; when we build a simulator, this will help other researchers who work in this field. Number two, we will explore different dual gratings, some borrowed from Professor Butler’s lab at LSU and see what results we get. Third, we hope our bone-metal imaging techniques can help improve the success rate of hip replacement surgeries around the world, and contribute to the knowledge of materials scientists who make hip replacements. Hip replacements are one of the most popular and fastest growing elective surgeries. But about five percent fail, and this is something we hope to help scientists reduce. If we get a sample from a failed surgery, we can see what went wrong; for example, if it cracked, how, where, etc. If we can help scientists make the interface between metal and bone more stable and more secure for the rest of a patient’s life, they don’t have to return to the hospital or do the surgery again.”

 

 

Elsa Hahne
LSU Office of Research & Economic Development
225-578-4774
ehahne@lsu.edu