We're excited to join the FSTCS 2026 Annual Scientific Meeting this week! Stop by the PRISMAp table July 9–11 to explore our Digital Twin ICU project, try our VR demo firsthand, and see the future of healthcare innovation in action.
Learn more: https://t.co/cQUnfvXDKq
Delirium affects ~1 in 3 ICU patients, yet is detected only after symptoms appear. The LLM-based tool, DeLLiriuM, helps predict delirium risk within 24 hours of ICU admission to promote earlier intervention.
Read more: https://t.co/Ih6L8BEv13
Our new publication reviews how AI can support the prevention, detection, and management of acute kidney injury, while highlighting the need for stronger external validation before widespread bedside implementation.
Read more: https://t.co/ihw4MEJYYZ
Social determinants of health, like economic instability and social vulnerability, are strongly linked to worse ICU outcomes and higher mortality, highlighting the need to address health inequities in critical care.
Read more: https://t.co/oFqFwDEnL8
Introducing our Faculty & Staff Spotlight: Dr. Zhenhong Hu, PhD, MS, a multidisciplinary Research Assistant Professor whose career bridges high-level industrial innovation with high-impact clinical applications in healthcare research.
Our externally validated deep learning model uses patient data to predict worsening kidney damage up to 2 days in advance, allowing earlier interventions that may help prevent serious long-term kidney damage in hospitalized patients.
Read here: DOI: 10.34067/KID.0000000998
We’re proud to unveil a new chapter at PRISMAp.
Our new logo reflects our goal of designing, developing, and implementing Intelligent healthcare systems to transform medicine and patient care. As this field evolves, so do we.
Thank you to everyone a part of our journey.
🍁This Fall, uncover what the human eye can’t see.🍁
Our Biomedical Image Analysis module shows how AI reveals hidden insights in radiology, pathology & beyond.
Starts Oct 1 — enroll now: https://t.co/kSBJr8PAiE
#AIPassport#BiomedicalAI#UF_IC3
Tissue sample data can be imperfect, requiring hours of annotation. At April's Journal Club, Dr. Ruining Deng presented “CASC-AI: Consensus-Aware Self-Corrective Learning for Cell Segmentation with Noisy Labels,” an LLM to streamline annontation. See you May 19 for our next one!
Want to work on AI that could one day save lives? PRISMAp is offering a year-long, AI in medicine paid research internship. Interns will work on AI/ML bootcamps, ethics workshops, and research with opportunities for publication. Apply here by April 15: https://t.co/7dwit7P09R
Have you heard of a medical version of ChatGPT? In our latest Journal Club, Dr. Wei Shao’s team unpacked the paper, "HealthGPT: A Medical Large Vision-Language Model for Unifying Comprehension and Generation." Thank you everyone who came and see you at our next one, April 21!
Traditionally, healthcare vision and language tasks are handled by separate models, limiting efficiency. HealthGPT, a Medical Large Vision-Language Model (Med-LVLM), bridges this by unifying comprehension and generation. Join our next journal club with Dr. Wei Shao to learn more!
Intelligent Hospital Digital Twin is our move toward a health metaverse: an AI- virtual ICU! Dr. @AzraBihorac met with @uflorida to discuss how Digital Twin technology will shape the future of medicine. Take a look here! https://t.co/bkZwXGyF3Q
The IC3's February was busy!
- AICC Workshop: Hands-on AI in healthcare training
- AI Collab Hub: Potential research partnerships
- Journal Club: Divya Vellanki discussed CRAFT-MD
- @UFMedResearch's COM Celebration of Research: Digital Twin of our AI-powered Virtual ICU!
At yesterday's Journal Club, Divya Vellanki explored "An Evaluation Framework for the Clinical Use of Large Language Models," introducing CRAFT-MD, which simulates doctor-patient interactions to assess LLMs. Join us for the next JC, March 24th to discuss AI's impact on medicine!
Join us for Journal Club on Feb 24, 2025, from 2-3 p.m. at Malachowsky Hall! Divya Vellanki will present on "An Evaluation Framework for Clinical Use of LLMs in Patient Interaction Tasks." Learn about CRAFT-MD, a framework testing LLMs like GPT-4 in real-world clinical settings.