Space, Bones & Breakthroughs: Dr. Thomas Lang on NASA Research & AI Imaging | Stars Launch Pod

Susan Wise: Welcome to Stars Launch Pod, Space Technology and Research Sciences, brought to you by starsciences.org, where we connect, collaborate, and accelerate. I’m your host, Susan Wise, and joining us today is Dr. Thomas F. Lang, Professor in the Department of Radiology and Biomedical Imaging at the University of California, San Francisco.
Hi, and welcome to the program.

Thomas Lang: Oh, thank you so much for having me.

Susan Wise: Well, I’m excited to talk with you today and learn more about your work. I’d like to hear more about what you’ve done in the past and what’s happening now.

Thomas Lang: I’m a professor at the University of California, San Francisco. If we go back to the mid-1990s, I was just starting out as an assistant professor and getting my group’s research program underway. What we developed was a quantitative imaging approach using hospital-type clinical CT scanners to quantify bone density. I’m sure you’re familiar with osteoporosis and bone densitometry.

Susan Wise: Yes! I’ve lost an inch and a half, and I want to know why.

Thomas Lang: Okay.
Well, the standard method for quantifying bone density uses bone densitometers. They’re basically X-ray scanners that create a two-dimensional quantitative image. What they can do is look at different regions of interest in images of the spine and hip, and quantify how much bone mineral mass you have.
They divide that by the projected area of the bone and get what they call areal bone mineral density, in grams per square centimeter. That technique had already been clinically developed when I started, but another development occurred around the same time—the emergence of new CT scanners that allowed for rapid acquisition of large volumes of data. Those were called helical CTs.

So, my group’s research focus became developing quantitative imaging analogous to standard bone densitometry, but in three dimensions, using this new generation of helical volumetric CT scanners. The beauty of this technique is that it allows you to take a 3D picture of the bone.

If a bone is undergoing age-related loss—or, in the case of astronauts, exposed to weightlessness in space—the bone loss varies depending on where you are in the bone, three-dimensionally. In the hip and long bones, for example, you have a thick envelope of dense cortical bone, and inside that is spongy bone marrow—an almost gossamer network of thin rods and struts.

Those compartments—cortical and trabecular (spongy)—change at different rates. What CT allows you to do is look in detail at where these changes are taking place.

When I started this work, my vision was to apply it to osteoporosis, but there was a wonderful opportunity that emerged around the same time. NASA had begun building the first parts of the International Space Station. I had the chance to apply, with a collaborator at Baylor College of Medicine in Houston, Dr. Adrian LeBlanc, for a grant to use this approach to study astronauts.

At that time, American astronauts had flown on Mir, and Russian cosmonauts had long-term missions there. NASA had installed a standard bone densitometer at Star City—Russia’s equivalent of Johnson Space Center. What we proposed was something similar, but using a CT scanner at Baylor College of Medicine. We imaged astronauts before they went up, immediately after they returned, and again a year later.

The idea was to look in detail at bone loss—examining trabecular bone in the interior of the hip, the surrounding cortical bone, and similar measurements for the spine—to understand how bone loss evolves, how much is recovered, and what three-dimensional changes occur during that process.

What we were able to do was standardize this approach so that it could be used for a wide range of studies. My involvement with NASA’s internal operations lasted from about 1999 to 2019—a long collaboration with multiple projects using this technique.

Susan Wise: As we focus on going back to the Moon and spending more time in space—and hopefully, someday, even Mars—this becomes even more important. Spending more time up there will have an even greater impact on bone health.

Thomas Lang: That’s correct.

Susan Wise: Yeah.

Thomas Lang: One of the rewarding aspects of this work was that it supported NASA’s efforts to develop countermeasures for bone loss. One of the first discoveries we made was that, with the first generation of exercise equipment sent to the space station, astronauts were still experiencing significant bone loss—even though they were working out almost every day.

Those early machines were interim solutions. Eventually, NASA flew a much more advanced exercise gym called ARED—the Advanced Resistive Exercise Device. It allows astronauts to perform high-impact exercises such as squats and deadlifts with heavy loads.

Using our technique, we found that ARED dramatically reduced the rate of bone loss. We were also involved in another study in which astronauts took an anti-osteoporosis drug.

Susan Wise: I was going to ask about that.

Thomas Lang: Yes—they took Fosamax, one of the first major anti-osteoporosis drugs on the market. We did what I’d call a pilot study—Dr. Adrian LeBlanc was the principal investigator—and what we found was that across all measures, the crew who took this intervention did not experience significant bone loss on average during their six-month missions.

Our technique helped NASA understand the risks of bone loss and evaluate the effects of different countermeasures—exercise and pharmaceutical.

Susan Wise: It sounds like you’ve had several “aha” moments over this time. What has been your biggest challenge, and how did you overcome it?

Thomas Lang: First, let me say what my biggest aha moment was. That came when we started getting our first data on bone recovery.

We were taking both DXA (standard bone densitometry) and CT measurements, which allowed us to look at macro-architectural features of the hip. What we found was that with DXA, astronauts would typically lose a significant amount of bone during their missions, but when we measured recovery after returning, it looked like they nearly regained it all—not 100%, but close.

However, the CT scans told a more complicated story. We saw severe bone loss in the interior trabecular compartment of the hip, and that area showed very limited recovery. Most astronauts did not regain bone in that compartment. Instead, their bones became larger as a way to adapt when reloading after returning to Earth.

So, they didn’t necessarily grow denser bone—they grew bigger bones. DXA shows that as a recovery of bone density, but in CT, you see that it’s mainly an increase in size, not density.

The aha moment was realizing that without proper intervention, there are irreversible changes in bone architecture. Those irreversible changes are similar to what happens with aging—people lose bone but their bones get wider. We were seeing, over six months in space, the same pattern that takes decades to occur on Earth.

That told us prevention is crucial—if you don’t prevent the loss up front, recovery afterward can’t fully fix it.

Susan Wise: Wow, that’s amazing. And what about your challenges along the way—one of your biggest challenges, and how did you overcome it?

Thomas Lang: My biggest challenge is probably one that any engineer or technology developer faces when trying to move something from the lab to production.

In our case, my group developed this three-dimensional imaging approach, and my goal wasn’t necessarily to make it a standard clinical tool used everywhere. I wanted it to be used in research—particularly epidemiological studies—to understand how the size, shape, and density of bones like the hip and spine relate to fracture risk, aging, or spaceflight effects.

To do that, you have to analyze a lot of images, which means you need algorithms that run smoothly, produce consistent results, and can handle large datasets.

Based on my experience, the biggest challenge is making a technology like this work reliably over time across many cases. Anyone who has taken something from a university lab to broader application knows that’s not easy.

In the end, you overcome it through persistence—addressing problems as they come up, solving them, and moving forward.

Susan Wise: What one piece of advice would you like to share with our listeners today?

Thomas Lang: The main piece of advice I’d offer is: always have a long-term vision for yourself, and work toward it. Ask yourself—where do I want to be in five years? Ten years? What impact do I want my work to have?

If you have that long-term vision, it helps you navigate the ups and downs of day-to-day challenges.

I remember around 2005 or 2006, I hit a really rough grant cycle. My group was entirely funded by NIH and NASA grants, so I had to keep a constant stream of proposals going. At one point, nothing I submitted was getting funded. I was mid-career, running out of money, juggling accounts and staff, and worrying constantly.

Then I realized I was just reacting day to day. I needed to step back and think long-term—what do I want to accomplish? What capabilities do I want to develop? I wrote out a plan, and that process led to a new round of successful grants. That experience really reinforced the value of thinking long-term.

Susan Wise: That support is so important for you to continue your research.

Thomas Lang: Right.

Susan Wise: For the long term.

Thomas Lang: Exactly. I thought, “I could give up and do something else,” but then I realized—I don’t want to do anything else. I want to do this. So, I asked myself: what do I need to do to keep doing it? And that mindset worked for me.

Susan Wise: I really appreciate your story and you sharing it with us. It’s such important work—not only for NASA and our astronauts, but also for all of us here on Earth. I hope your work continues. Before we wrap up, Thomas, do you have a past or current passion project you’d like to share?

Thomas Lang: Yes, I do, and I’m working on it now. Since retiring, I’ve had a chance to reflect on my “sins” as an image-processing developer. I’ve gone back and rewritten all my software in Python, which offers so many resources for making it more reliable, user-friendly, and easier to use.

Some colleagues and I recently received a small internal grant to combine that work with AI techniques to automate large portions of the process.

The goal is to apply our approach to clinical CTs stored in hospital PACS—picture archival systems. With the appropriate IRB permissions, these images are an incredible scientific resource. We want to use AI to analyze them automatically, allowing large-scale studies without needing huge research labs. Researchers would just need to monitor the AI and ensure quality control.

Susan Wise: AI has really helped in so many ways.

Thomas Lang: I’ve always felt like an “old beast,” left behind by newer developments. So this gives me a chance to catch up—and feel a bit more relevant again.

Susan Wise: (laughs) Thomas Lang, thank you so much for joining us today and for sharing your story. We’re looking forward to seeing what’s next for you and will definitely be following along. We appreciate you being here.

Thomas Lang: It’s been my pleasure. Thank you so much for inviting me.

Susan Wise: Be sure to join us as we connect, collaborate, and accelerate. Visit starsciences.org to receive a free one-year membership—just enter the promo code FRIENDS321. If you enjoyed this episode, please leave us a review and subscribe. Until next time—keep looking up.