Radiology Research in a Rapidly Changing World

The world is changing at an accelerating rate, creating growing levels of anxiety in the U.S. and across the globe. An aging population, the climate crisis, almost impossibly fast advances in artificial intelligence (AI), the COVID-19 pandemic and demographic shifts tearing at the social fabric of people's lives. All of these and more not only weigh heavily on the minds of many, but they also impact human health in various ways.

Subscribe to the latest updates from Radiology Advances in Motion

Error: Please enter a valid email address.

Email Address Submit

Radiology research can play an important role in responding to the changes underway. Even now, investigators in the Radiology Department at Massachusetts General Hospital are studying the healthcare implications of many of the changes and, where appropriate, exploring avenues through which to mitigate their detrimental effects. Here are four ways in which they are currently doing this.

Caring for Our Aging Population

The aging population in the U.S. is a growing concern, not least because it brings with it considerably higher rates of neurodegenerative disease. As the prevalence increases, radiology researchers at Mass General continue to lead the way in studying the diseases and developing new tools for diagnosis and treatment.

Heidi Jacobs, PhD, a clinical neuropsychologist and a neuroscientist in the Athinoula A. Martinos Center for Biomedical Imaging, focuses on Alzheimer's, the most prevalent of the neurodegenerative diseases. She and her team are working toward earlier detection and, ultimately, the introduction of interventions that could help stave off the most devastating effects of the disease.

"My grandmother died of late-onset Alzheimer's," she says. "Experiencing up close how this disease robs an individual of their memories, and ultimately of who they are as a person, motivated me to figure out how we can halt the disease."

Much of her work centers on a pair of toxic proteins in the brain: beta-amyloid and tau. Currently, the earliest known stages of Alzheimer's disease are characterized by the presence of these proteins in the cortex, the outermost layer of the brain, in individuals who are not yet showing clinical symptoms. Dr. Jacobs and her team are looking to move the detection of Alzheimer's to an even earlier stage in its progression — before the proteins begin to affect the brain — by pioneering methods using existing technologies that allow them to image tau-related processes in the brain as early as early adulthood.

They have now achieved this and more. They have shown that not only can they visualize tau-related processes in the structure using these methods, but they can also predict cognitive decline earlier and more precisely than was previously possible.

Dr. Jacobs believes that, with further development, the methods could transform our approach to detecting and managing Alzheimer's.

"I envision that in the future we will image the locus coeruleus of healthy individuals during their midlife years and that this can move clinical work toward healthcare — instead of disease-care," she says. "Including the results of these scans in the standard clinical evaluation could help clinicians provide personalized lifestyle advice or enroll individuals in prevention trials to maintain a healthy brain."

Keeping Pace With the Rise of AI

Researchers are developing AI models for applications across the field of radiology, from improving image registration in MRI studies to predicting the risk of heart attack and stroke based on a single chest X-ray. In recent years, they have introduced seemingly countless new tools that promise to transform healthcare by wielding the unprecedented power of AI.

As with any new medical technology, though, successful demonstration of new AI models is only the first step in getting the tools into the hands of clinicians. The models must also undergo other steps in the translational process: regulatory approval, clinical validation, and more. If AI is to achieve the same kind of penetration in healthcare as in other application areas, developers of the tools need to find new and innovative ways to complete these steps.

Mass General Brigham researchers have long sought to facilitate the translation of AI advances into the healthcare arena. They started collaborating with industry nearly a decade ago, initially developing AI tools and handing them off to companies for commercialization, and gradually taking on additional steps in the translation process (regulatory approval, etc.). More recently, they introduced an AI business entity, allowing them not only to drive commercial revenue, but also to drive revenue in ways that are fully aligned with the Mass General Brigham mission.

Further supporting the integration of AI healthcare solutions into clinical practice, in November 2024, Mass General Brigham AI announced the launch of the Healthcare AI Challenge, a series of events that will enable healthcare professionals from across the U.S. to assess the effectiveness of the solutions in simulated environments. The feedback they provide will fuel publicly available insights and analytics, essentially crowdsourcing assessments of the medical and societal benefits of the solutions.

"The velocity of AI innovations and breadth of their healthcare applications continues to increase," says Keith Dreyer, DO, PhD, chief data science officer at Mass General Brigham, and leader of Mass General Brigham AI, the healthcare system's AI business. "This unprecedented growth leaves clinicians struggling to determine the effectiveness of these innovations in safely delivering value to healthcare providers and our patients.

"The Healthcare AI Challenge is a collective response to the complexities involved in advancing the responsible development and use of AI in healthcare. This new approach strives to put clinicians in the driver's seat, allowing them to evaluate the utility of different AI technologies and, ultimately, determine which solutions have the greatest promise to advance patient care."

The first in the series of events will focus on radiology, not least because of AI's already demonstrated impact in the field and its broad potential for the future.

Addressing the Epidemic of Loneliness

Loneliness and social isolation are profound concerns in the U.S. — so much so that, in a 2023 report, the U.S. Surgeon General identified it as an epidemic. While public health experts and others called attention to this epidemic even before the COVID-19 pandemic, the pandemic greatly exacerbated the problem. At the same time, even as the worst days of COVID-19 recede into the past, the issues themselves have not gone away.

The impact of the epidemic has been especially profound in some of the most vulnerable populations in the U.S. Researchers have shown, for example, that loneliness and isolation are particularly common in those suffering from serious mental illnesses, including psychotic disorders, affecting approximately 80% of people with these conditions.

All of this has broad public health implications. Not least, a growing body of literature has linked loneliness and isolation to poor cardiometabolic health and premature death. Researchers are working to address these negative health outcomes. However, little is known about the brain mechanisms underlying them and how they might vary across different groups of people.

At the Athinoula A. Martinos Center for Biomedical Imaging at Mass General, researchers Daphne J. Holt, MD, PhD, and Roger Tootell, PhD, are working to uncover these mechanisms.

Using functional MRI, Drs. Holt and Tootell and colleagues, including Faye McKenna, PhD, Louis Vinke, PhD, and Mona Avanaki, MD, have identified a potential neural marker of loneliness: a specific pattern of neural responses to stimuli associated with loneliness. This could prove an important first step in addressing the negative outcomes tied to the epidemic.

"Understanding how the brain responds when a person's social needs are not being met gives us an objective way to measure and monitor this type of deficit," says Dr. Holt, "which will help us understand the effects of loneliness on the body and how to reverse those effects."

More recently, the Martinos-based team has been testing whether the neural marker is tied to some of the known biological effects of loneliness (for example, cardiometabolic and inflammatory dysregulation), and whether these effects contribute to the poor health outcomes and earlier mortality associated with psychotic disorders and loneliness across populations.

Reducing Healthcare's Contributions to Climate Change

Human-attributable climate change can profoundly affect human health. According to the Centers for Disease Control and Prevention (CDC), it can lead to higher rates of respiratory and cardiovascular disease, greater prevalence of food- and water-borne illnesses (and changes in the geographical distribution of those illnesses), and more.

Researchers have long pursued deeper understandings of disease prevalence stemming from climate change — and have achieved significant insights that will help mitigate its impact. However, tackling the effects of climate change is only one piece of the puzzle. The healthcare industry must also work to reduce its own contributions to such change, which can also play a role in adverse, climate-related health outcomes. Healthcare is responsible for 9% of U.S. greenhouse gas emissions; radiology is a major source of these emissions, largely because of its use of energy-intensive imaging technologies such as MRI.

Mass General Radiology researchers are now doing just that: identifying their field's contributions to climate change — and those of the healthcare industry as a whole — and finding ways to reduce the contributions.

In a paper published in the Journal of the American College of Radiology, Michael E. Zalis, MD, a cardiovascular and interventional radiologist and director of the Mass General Brigham Radiology Center for Sustainability, and Jonathan E. Slutzman, MD, a physician in the Department of Emergency Medicine and director of the Center for the Environment and Health, outlined ways in which the field of radiology can limit its climate impact.

Among the recommendations: Drs. Zalis and Slutzman suggest that radiology should set sustainability goals, particularly through the use of emerging modalities such as low-field MRI and advanced X-ray technologies, which require less power and reduce emissions. Operational changes, including powering down equipment during idle times, can further lower energy use.

Learn more about Radiology research at Massachusetts General Hospital

Learn more about Mass General Brigham AI