HMI World

Welcome

Based on PHMI inputs, new "health city" in plans for Australia. Also: LAU and PHMI host national program on medical education

Feature Stories

In Conversation: MGH radiology chief discusses the big picture

As Radiologist-in-Chief at Massachusetts General Hospital, Dr. James Thrall helps oversee the movement of major innovations in medical technology from concept to practice. Here, in an interview with HMI WORLD, Dr. Thrall goes back to basics in his ever-changing field, and charts the course of radiology as it moves forward in the coming years.

HMI WORLD: How can we take a step back from the modalities we see in practice and learn how to think about the concept of imaging?

Dr. Thrall: Every medical imaging procedure involves the interaction of some kind of energy with the tissues of the body. So depending on which tissue we want to examine, we may choose one energy source or another, such as X-ray to look for a broken bone or ultrasound to look at a fetus inside the womb.

HMI WORLD: Before we move into the future of radiology, can you help us understand where we are now, in terms of present capabilities and the clear opportunities for the future?

Dr. Thrall: The history of radiology has been at the level of organs and organ systems—one of the first things that X-rays were used for was to diagnose broken bones. In the lung, we use X-rays and CAT scans to diagnose tumors and pneumonia, and in fact we use X-rays in almost every part of the body to diagnose tumors and infectious disease.

The future is going to take us to smaller dimensions. Just as there are many areas of science where people are very excited about nanotechnology, we are very excited in imaging with the ability to now image not just at the level of the whole organ, but at the level of molecules and cells. The wonderful new term “molecular imaging” is used to describe this trend.

The other thing that we’ve been able to do is take advantage of the change in the function of an organ in order to watch organs process signals from the outside world. This is most important in the brain, where we are able to actually watch the brain think by observing changes in the energy metabolism of the brain, and changes in the blood flow to different parts of the brain as it processes a thought. It’s really quite astonishing because there is a tight coupling between energy utilization, blood flow, and the amount of work that the brain is doing. So if we expose a subject to a visual stimulus, the parts of the brain that process the signals in the eyes will light up on the scan. Or if we play a musical note, a different part of the brain that’s involved in auditory processing lights up.

HMI WORLD: Molecular imaging is a new term but maybe not such a new concept, right? There are things we have been using for years which can be considered molecular imaging.

Dr. Thrall: Yes. Molecular imaging got its start 50 years ago or more in nuclear medicine. But we didn’t conceptualize what we were doing as molecular imaging. Today we would call what we did 50 years ago molecular imaging. In the meantime, the number of different applications has simply skyrocketed. So it’s no longer just nuclear medicine—it’s now ultrasound, MRI, and even visible light sources or near-infrared light sources used as the energy for molecular imaging.

HMI WORLD: A lot of focus in imaging is on the machines, but the science is a complicated mix of physics, engineering, chemistry/biology, and now IT. What is the process of how advances in imaging science become part of imaging practice?

Dr. Thrall: It is a complex process. To shorten the story, think of some other thing that we use in every day life, like a television set, and consider how it has been improved over the years: the advent of high-definition recording, flat panel displays, and other innovations. The same thing is happening in imaging, where every component of everything we do is simply getting better, allowing us to see finer detail with higher resolution. We can capture changes in organ function more rapidly, and we can use pharmaceuticals as enhancement agents to light up certain parts of the body that are involved in a disease process.

HMI WORLD: What is the interplay between academic medical centers and industry that brings these sorts of advances into practical application?

Dr. Thrall: Typically industry does the engineering and physics and comes to us and we do what’s called translational research with every new product to demonstrate its usefulness clinically. In fact, this is required by the U.S. Food & Drug Administration. So there’s necessary partnership between industry and the academic world because the only place you can do that kind of clinical research is in a hospital. You cannot do that in the manufacturer’s research laboratory.

HMI WORLD: So that would probably go hand in hand with training and education. Students come up through academic medical centers where new technologies are being tested.

Dr. Thrall: Absolutely. In fact, people train in academic centers where all the cutting-edge devices and concepts are being tested. Often when they leave those centers, for example to go into private practice, it’s like stepping back in time, sometimes to their surprise.

HMI WORLD: Today magnetic resonance imaging (MRI) is thought of mainly as a modality for diagnosing neurological and musculoskeletal problems. Where do you see the use of MRI expanding?

Dr. Thrall: MRI is now being used in cancer diagnosis. For example, in prostate cancer it’s possible to obtain highly detailed spectroscopy images to distinguish benign from malignant prostate tissue. This will be extremely valuable. There are some MRI enhancement agents—so-called contrast enhancement agents—that light up specific tissues that are targeted to important structures like lymph nodes. This is bringing MRI to the threshold of being used to determine whether a cancer has spread from where it started, say in the breast or the prostate, and to determine whether it has spread to the lymph nodes in the area. These are going to be extremely important applications of MRI.

HMI WORLD: What about PET/CT, which wasn’t on the radar screen a decade ago, and now is the gold standard for oncology? Where do you see its use expanding in the coming years?

Dr. Thrall: It’s terrific, because positronic emission tomography and computed tomography combines metabolic information about what’s occurring in a particular tissue with high-resolution anatomic information. Before we could put those two together, we couldn’t say with a high degree of certainty that a particular metabolic abnormality was matched up with a specific structure in the body. With PET/CT we can do that, and it allows us to be more accurate in our diagnoses.

HMI WORLD: What role is imaging now playing in the diagnosis of cardiac problems?

Dr. Thrall: This is the area of coronary computed tomography. First, a step back: if you look at the natural history of everything done in medicine, it starts out as very invasive, and as we understand it better, and as technology gets better, it becomes less invasive. In fact, one might think of diagnostic medical imaging as non-destructive testing.

So right now the standard for studying the coronary arteries is the catheterization laboratory: a catheter is inserted into an artery and then pushed up to the heart. Of course this is less invasive than cutting open the chest in an operating room, but it is much more invasive than putting a needle in the patient’s vein and injecting contrast material, followed by a CT scan to visualize the same arteries that the cardiologist is looking at in the cardiac catheterization laboratory. This is going to benefit patients, first of all, because currently 30 percent of patients who go to the catheterization laboratory turn out to not have disease. If we had known that before they went to the laboratory, then they never would have had to go. Therefore, if we do less invasive testing first, and get the diagnostic answer, we can use the cardiac cath lab more efficiently, by only sending patients there who actually need treatment to open up the blocked arteries. In fact, if you talk with cardiologists, even though they make their living doing these procedures, no one really wants to do a study that in retrospect is not necessary, because not only does it expose the patient to discomfort, radiation, and the risk of the pharmaceutical, but it’s also expensive. And society simply cannot afford to do things that aren’t effective.

HMI WORLD: Will we soon see a PET/MRI modality, and if so, what will that allow us to do?

Dr. Thrall: We will. In fact, at MGH we are installing a PET/MRI device in one of our research laboratories. We’ll be able to meld the functional imaging capacity of MRI with the metabolic information coming from PET scanning. This has tremendous potential and tremendous implications, mostly directed at continuing to understand how the brain functions. I am less certain what the impact will be for day-to-day clinical practice, but in terms of mapping the brain and determining how it’s wired, and what causes it to behave one way or another, PET/MRI is going to be unbelievable.

HMI WORLD: Finally, I want to ask you about one of the other major advances that has impacted imaging: the move from analog to digital image recording. How is digitization changing what we are able to learn about patients from imaging?

Dr. Thrall: The switch from analog recording of images to digital recording has been pivotal in the modern age of imaging. When we recorded the image on a piece of X-ray film, we could not process or manipulate that image after the fact. We could measure a few things, but we could not re-project the tumor in three dimensions to allow a surgeon to study it and understand how best to approach it in the operating room. We could not do tissue segmentation, which allows the separation of different kinds of tissue using digital image processing. We could not use computer-aided detection algorithms, which many of us believe are going to be very effective. So the ability to extract quantitative parameters from the data and to create 3-D representations of the data, and the ability to do computer-aided detection all are contingent on having the data in digital form.

HMI WORLD: What challenges does this present for the hospital? Are we moving closer to the idea of a virtual radiology department?

Dr. Thrall: That’s right. Working in a digital environment is akin to mobilizing all of the information as soon as it’s in digital form, whether it’s a medical image or any other kind of information. This enables us to send it anywhere in the world in order to collaborate, seek second opinions, and balance the load of work.

But I might say that the digital age came late to all of medicine. Banking, entertainment, and business were all way ahead of imaging. In fact, when mankind landed on the moon in 1969, an event predicated on the use of computers, there was not a single important application of computers in diagnostic radiology. Three years later the CT scanner was announced to the world.

(Photo used courtesy of MGH REMS)

Search

Featured HMI Website Content

Academic Initiatives by PHMI Team Members: Visit the PHMI Main Website to learn more about the academic activities undertaken by our faculty and staff, including publications, resaearch, and teaching around the world.

HMI Reader Survey

What global health topic would you like to learn more about?

Around Harvard

A new toolkit for analyzing drug effects on the mitochondria, an aerosal vaccine for tuberculosis, and an alternate pathway to growing blood vessels when body parts are jeopardized by poor circulation. Also: folic acid, hypnosis, and other news and research from the Harvard medical community.