Vital Announces Acceptance of Peer-Reviewed Manuscript: AI-Generated Radiology Summaries
Vital is pleased to announce the acceptance of our peer-reviewed scientific manuscript, "Patient-Readable Radiology Report Summaries Generated via Large Language Model: Safety & Quality," in the Journal of Patient Experience (JPX).
The study focused on evaluating the safety and quality of AI-generated summaries of complex clinical radiology reports. These summaries were produced using large language models (LLMs) that translate specialized medical terminology into simpler, more understandable language. The goal is to improve patients' comprehension of their medical findings and enhance patient autonomy.
Key Findings: Eight board-certified physicians reviewed and rated the safety and quality of 1,982 LLM-generated radiology report summaries from multiple imaging modalities, including CT, MRI, ultrasound, and x-ray. Employing rigorous pre-defined safety criteria:
Safety: The LLM-generated reports demonstrated high safety, with 99.2% (1,967 out of 1,982) of the summaries determined to be safe by the physician reviewers
Quality: The quality of the summaries received high marks, with 92% rated as "Very Good" or "Good”
The successful generation of high-quality, easy-to-understand, and safe summaries demonstrates the potential of AI tools in breaking down the barriers posed by complex medical jargon. This advancement is pivotal for enhancing patient understanding and fostering informed decision-making.
“By making radiology reports more accessible, we aim to improve the overall patient experience, ensuring that patients have full understanding of their health conditions and treatment options,” said Nicholas Sterling, MD, PHD and Director of Clinical Innovation and Research at Vital. “This study marks a significant step forward in using AI to enhance patient autonomy and informed decision-making.”
Study Reference: Sterling, NW, Brann F, Frisch, SO, Schrager JD. Patient-Readable Radiology Report Summaries Generated via Large Language Model: Safety & Quality. J Patient Experience (accepted, in-press). DOI: 10.1177/23743735241259477
About Vital
Vital is redefining patient experience with software that gives more control, clarity, and predictability to emergency department visits and hospital stays. Using advanced artificial intelligence (AI), Vital transforms complex health record data into easy-to-use, personalized interfaces that inform and engage over one million patients per year — no downloads or passwords required. With Vital, patients can better understand progress toward discharge, request service and comfort items, set goals, view labs, share health status with family, book follow-up care, and much more. Hospitals across the U.S. use Vital to improve patient satisfaction, drive growth and patient loyalty, achieve better clinical outcomes, and reduce workload for care teams.
Founded by Mint.com creator Aaron Patzer and Emergency Physician Dr. Justin Schrager, Vital is a HIPAA-compliant cloud-based experience layer that integrates with any electronic health record system. For more information, please visit vital.io or follow us on LinkedIn or Twitter.
For media inquiries, interviews, or additional information, please contact:
Kanchhi Solanki | Director of Marketing
kanchhi.solanki@vital.io