The University of Florida is participating in a large national program to create an efficient and mobile healthcare delivery system for rural areas, awarded by Advanced Research Projects Agency for Health.
As hundreds of hospitals in rural areas struggle to survive financially, a new program envisions a cost-effective way to ensure high-quality healthcare in these regions. The program, Platform Accelerating Rural Access to Distributed and Integrated Medical Care (PARADIGM), has announced teams across the U.S. that will develop a smart, mobile fleet to deliver hospital-level healthcare on the road.
AI Team for Mobile Lab
A University of Florida team and its partners will develop an artificial intelligence system to guide basic lab procedures. The research team must train computers to recognize and use complex medical terminology, images, and other inputs. The system’s intelligence grows over time based on a large language model (LLM) or multimodal model that goes beyond text.
Yonghui Wu, Ph.D., inspects the Hipergator supercomputer at the University of Florida.
“AI and especially large models have demonstrated high potential to transform medical discovery and improve healthcare. This project will deliver an AI system, called Multi-Tags, to improve healthcare for rural populations through an electric vehicle that integrates various medical devices and technologies to deliver hospital-level care where no hospital exists,” said Yonghui Wu, Ph.D., an Associate Professor in the UF College of Medicine’s Department of Health Outcomes and Biomedical Informatics (HOBI), and Director of Natural Language Processing at the UF Clinical and Translational Science Institute.
Wu will lead the development of the multimodal vision language models in collaboration with California-based research center SRI and Mayo Clinic Jacksonville.
Wu has deep experience with LLMs. He developed and disseminated GatorTron, the most widely used clinical LLM in the medical domain, with over 1.8 million downloads from the model’s repository (https://huggingface.co/UFNLP).
Boosting Care Efficiency
When completed, the PARADIGM program will deploy high-tech vehicles that may function similarly to the new UF Mobile Cancer Screening Connector. Instead of relying on medical doctors, the mobile care units will enable healthcare workers to perform clinical procedures directly.
This solution intends to reduce training requirements, increase accessibility in rural areas, and enhance healthcare efficiency, ultimately lowering costs while improving patient outcomes.
On these new vehicles, the AI Multi-Tags system will provide intelligent task guidance. It will provide real-time prompts that can turn a generalist into a just-in-time specialist.
If successful, the AI models developed in this collaborative effort will extend healthcare services beyond the walls of a hospital by bringing medical care to even the most remote areas and enabling rural patients to access the care they need in their own communities.
