Harry Marshall, MD, PhD

Welcome

I'm an abdominal radiologist at Vanderbilt University Medical Center. I run clinical imaging trials to improve how we practice radiology, whether it's testing AI tools, refining MRI protocols, or rethinking how we teach diagnostic interpretation.

I've built this site to share my research and teaching material. Whether you're here to learn, collaborate, or just browse, I hope you find something that sparks your interest.

• View CV

Research

My research focuses on improving how medical images are acquired, analyzed, and interpreted. I develop and evaluate image analysis methods that could help radiologists interpret images, automate time-consuming tasks, and support learning. Reader studies and clinical trials are central to testing whether new tools and imaging approaches improve diagnostic performance and how we deliver care.

This work spans my PhD research on PET/MRI image correction, computational segmentation and lesion analysis, radiologist performance and education, and a clinical trial comparing abbreviated and full liver MRI protocols. Explore these projects and the software we are building to support reader studies below. You can also view my full list of publications on Google Scholar or visit my ORCID profile.

• Explore Research

Teaching

Teaching is one of the most rewarding parts of my work. I regularly teach medical students, radiology residents, and fellows, and have created a series of lectures spanning a wide range of topics in abdominopelvic imaging, from diffuse liver disease and prostate MRI to dual-energy CT and colorectal cancer staging.

This section features online presentations. Whether you're reviewing foundational material or diving into more advanced topics, I hope you find these resources useful.

• Explore teaching

Contact

Feel free to reach out using the form below or connect with me on LinkedIn, GitHub, Bluesky, or YouTube.

Please do not send patient information or other confidential health information through this form. For matters involving patient care, use your institution’s approved communication channels.

• LinkedIn
• GitHub
• Bluesky
• YouTube

Elements

i = 0;

while (!deck.isInOrder()) {
    print 'Iteration ' + i;
    deck.shuffle();
    i++;
}

print 'It took ' + i + ' iterations to sort the deck.';