AME Seminar

Speaker: Xiao Hu, Emory University

Talk Title: Unleashing the Power of AI for Precision Health: The Vital Role of Physiological and Nursing Data

Abstract: Artificial intelligence (AI) has tremendous potential to advance clinical practice and patient care by providing clinicians augmented abilities to derive diagnostic and prognostic insights from various types of data. Medical images, structured data, clinical notes in electronic health record systems are data modalities that have so far received much attention. In addition to these data modalities in spotlight, continuous physiological data including electrocardiography, blood pressure, intracranial pressure, electroencephalography, photoplethysmography signals are part of standard of care, hence ubiquitously available for patients in acute care, and least susceptible to practice variations. Rich and dynamic pathophysiological information is embedded in these signals and yet there are no experts like radiologists dedicated to interpreting these signals at scale. Therefore, there is a vast amount of untapped information in these signals. In this keynote, we will explore three overarching approaches to process physiological data: The single modality approach, where novel metrics are derived from a single signal, unveiling physiological insights that remain concealed in conventional patient monitors. The multi-signal approach, which analyzes multiple signal modalities to elucidate the intrinsic interplay among different organ systems, providing more precise signatures of acute illnesses. The multimodality approach, which integrates physiological data with other clinical information, enabling enhanced patient monitoring capabilities and more precise care delivery. Bedside nurses play a pivotal role in continuously managing, interpreting, documenting, and communicating physiological data. However, they often face alarm fatigue due to inferior built-in algorithms of patient monitors. By harnessing the power of AI tools to process physiological data, we can alleviate this burden, elevate the nursing profession, and ultimately improve patient care outcomes.

Biography: Xiao Hu is Asa Griggs Candler Chair Professor at the Nell Hodgson Woodruff School of Nursing, associated faculty at the Departments of Computer Sciences and Biomedical Informatics, and PhD program faculty at the joint Biomedical Engineering program of Georgia Tech and Emory University. He also serves as the Associate Director of the Center for Data Science. In his remarkable career, he has held faculty positions at esteemed institutions like UCLA, UCSF, and Duke University. Dr. Hu's pioneering research lies at the intersection of computational and health sciences, using advanced algorithms to transform healthcare data into actionable patient care insights. His significant contributions include over 160 peer-reviewed publications, multiple NIH research projects, and nine US patents.

Host: AME Department

More Info: https://ame.usc.edu/seminars/

Webcast: https://usc.zoom.us/j/98121141178?pwd=VGEyaXVWYnRaazFYWUVhbVAycGVWQT09

Location: Seaver Science Library (SSL) - 202

WebCast Link: https://usc.zoom.us/j/98121141178?pwd=VGEyaXVWYnRaazFYWUVhbVAycGVWQT09

Audiences: Everyone Is Invited

Contact: Tessa Yao

Event Link: https://ame.usc.edu/seminars/