Other Diseases/Conditions

Imaging of the endocrine system is essential for diagnosing and managing disorders affecting hormone-producing glands, including the thyroid, adrenal glands, and pancreas. Techniques such as ultrasound, MRI, and PET scans provide detailed views of glandular structure and function, helping to detect tumors, cysts, and other abnormalities. Ultrasound offers real-time imaging for evaluating the thyroid and parathyroid glands, while MRI provides high-resolution images of the pituitary and adrenal glands, allowing for precise assessment of lesions and tumors. PET imaging, particularly with radiolabeled glucose, helps in identifying metabolic activity in endocrine tumors, aiding in staging and treatment planning.

Diabetes is a critical area of study due to its widespread prevalence and significant impact on global health. Beta cell loss occurs in the endocrine pancreas resulting in loss of insulin secretion and subsequent onset of diabetes. Current clinical measurements of beta cell function may underrepresent total beta cell mass, as function varies in response to glucose demands. PET and MR imaging play crucial roles in diabetes research and management by providing detailed insights into metabolic and structural changes of the pancreas. Together, these imaging techniques enhance diabetes diagnosis, monitoring, and targeted therapy development.

We are using PET and MR imaging to investigate mechanisms of receptor/enzyme pharmacology and overall structural changes in diabetes. Current projects include using PET to image dopamine receptors in the pancreas both in type 1 and type 2 diabetes as a surrogate for beta cell mass and vesicular monoamine transporters as a surrogate measure of insulin vesicle capacity, both important targets for tracking diabetes progression and treatment. We have developed and optimized the use of two PET radioligands, [18F]FP-(+)-DTBZ and [11C](+)-PHNO, targeting beta-cell VMAT2 and dopamine (D2/D3) receptors, respectively.

Investigator: Jason Bini

The prevalence of obesity in the United States population is over 30%, predisposing a large portion of the population to metabolic diseases. Cortisol, a steroid hormone, is responsible for stimulating gluconeogenesis in the liver, promoting differentiation and maturation of adipocytes, and regulation of acute and long-term stress (e.g., inflammation) via neurons and glial cells. Understanding the role of cortisol is of critical importance in metabolic diseases, including obesity. Cortisol is activated from cortisone by the intracellular enzyme 11β-hydroxysteroid dehydrogenase type 1 (11β-HSD1). Reducing intracellular cortisol exposure may prevent obesity progression, cognitive decline and Alzheimer’s disease. 11β-HSD1 inhibitors have been proposed as treatments for obesity and Alzheimer’s disease; however, further study with PET/CT imaging of this enzyme may reveal appropriate populations for these treatments.

We are currently investigating these mechanisms in both preclinical models of obesity and clinical studies in individuals with obesity using PET/CT imaging and [¹⁸F]AS2471907, an 11β-HSD1 antagonist radioligand.

Investigator: Jason Bini

Imaging allows for detailed visualization of immune cell activity and the inflammatory processes within the body. PET imaging, using specific radiotracers, can highlight areas of high immune activity and track the spread of immune responses, while MRI provides high-resolution images of the lymphatic system and affected tissues. These imaging methods enable precise assessment of immune-related conditions, guiding treatment decisions and enhancing our understanding of immune system dynamics.

Advanced imaging techniques like PET and MRI provide detailed insights into the extent and activity of inflammation within the body. PET imaging identifies areas of metabolic activity linked to inflammation by using radiotracers (e.g., ¹⁸F-FDG) that highlight inflamed tissues. MRI offers high-resolution images that reveal structural changes and tissue damage caused by inflammation. These imaging methods allow for precise localization, assessment of severity, and monitoring of inflammatory processes, thereby guiding effective treatment strategies and improving clinical outcomes.

Neuroinflammation, primarily driven by activated microglia, is involved in neurodegenerative diseases and can cause behavioral changes. Activated microglia express translocator protein (TSPO), visualized using the radiotracer [11C]PBR28 with PET scans. We studied tracer uptake changes in healthy subjects before and after administering lipopolysaccharide (LPS), finding that LPS increases brain microglial activation, inflammatory cytokines, and sickness symptoms. In PTSD patients, severe anhedonia was linked to a suppressed neuroimmune response. These results indicate that PTSD patients have a reduced microglia-mediated response to immune challenges, corresponding to symptom severity. This PET imaging can help test new drugs to reduce acute neuroinflammation.

Investigator: Kelly Cosgrove

We are developing non-invasive quantitative techniques to monitor the immune response using PET imaging. Due to the presence of a hyper- or hypo-active immune response in many types of disease (including cancers, autoimmune conditions, or infectious diseases), a tool to monitor specific immune subtypes is greatly needed. We use radiolabeled nanoparticle imaging agents that specifically target circulating monocytes and tissue-resident macrophages. These same nanoparticles can be modified to target additional immune subtypes as well, including B’s, T’s, and neutrophils. We have applied these imaging techniques to a range of applications including cancer, multiple sclerosis, wound healing, atherosclerosis, and fungal infection. We have also expanded the use of these particles in theranostic applications for targeted radiotherapy in solid tumors. These techniques not only advance basic science knowledge of the course of inflammation in a host of clinically relevant applications, but also promise to assist in drug development and personalized medicine.

Investigator: Moses Wilks

Lung ultrasound is essential for the diagnosis and management of respiratory conditions such as pneumonia and pulmonary edema, especially in patients with breathing issues. It is portable, affordable, and radiation-free. However, proper use requires skill for both acquisition and interpretation. Artificial intelligence can help with both of these essential skills. Automated acquisition guidance with autocapture and interpretation for non-experts could broaden access to quality lung ultrasound in regions without skilled personnel, improving care and diagnostics in underserved areas.

We have several multicenter academic-industry partnerships with both industry and BARDA funding focusing on development and validation of artificial intelligence algorithms for the detection and quantification of B-lines, consolidations, effusions, and pleural based abnormalities on lung ultrasound. These are pertinent to both fluid overload states, such as heart failure detection and management, but also respiratory viral and bacterial illness such as COVID-19 and bacterial pneumonia.

In addition, we are spearheading a multicenter academic-industry partnership, with funding from the Bill and Melinda Gates Foundation, for the development and validation of deep-learning-based algorithms to guide novices in the acquisition of high-quality lung ultrasound clips, which is aided by auto-capture and annotation.

Investigator: Cristiana Baloescu

Skin and soft tissue infections (SSTIs), including abscesses, are among the most common reasons for emergency department visits, accounting for over 3 million cases annually in the U.S. Accurate diagnosis is critical, as missed or delayed treatment can lead to systemic infection, hospitalization, and unnecessary procedures. Despite the widespread use of point-of-care ultrasound (POCUS), diagnosis remains highly operator-dependent, contributing to variability in care and poor outcomes.
We are starting an interdisciplinary project that aims to develop an AI-enhanced POCUS tool to assist clinicians in accurately identifying abscesses at the bedside. By supporting more consistent and timely diagnosis, the tool has the potential to improve patient triage, reduce overtreatment, and enhance antibiotic stewardship in frontline care settings.

Investigator: Cristiana Baloescu