Loading Events

« All Events

  • This event has passed.

Health Data Science Brown Bag Talk – Using Machine Learning to Improve Diagnosis in the Intensive Care Unit – Dr. Jason Adams

April 30, 2021 @ 11:30 am - 1:00 pm


Abstract: Diagnosis and prognostication is challenging in complex, high stress, high stakes clinical environments such as the intensive care unit (ICU).  ICU patients are highly monitored by multiple physiologic sensors that generate high-sampling rate time series data streams (i.e., ‘waveforms) interpreted by clinician at the bedside, but interpretation remains subjective, requires specialized training, and these data are not captured routinely for remote review or research.  At UC Davis Health, we have developed a model use case to investigate the utility of automated processing of physiologic waveform data in the ICU, focusing on the use of ventilator waveform data and machine learning (ML) to detect acute respiratory distress syndrome (ARDS).  We have developed custom informatics to enable waveform acquisition, and both classical supervised ML and deep learning models to detect ARDS.  Dr. Adams will discuss this research and some of the challenges of applying data sciences methods to complex clinical problems.

Biography: Dr. Adams is a Pulmonary and Critical Care Medicine physician with extensive clinical experience caring for patients with critical illness and both acute and chronic pulmonary diseases. He has special interests in cardiovascular critical care and in the management of patients with respiratory failure requiring mechanical ventilation. Dr. Adams also serves as the Director of Digital Health Innovation for UCDH, the co-director of the UCDH Data Provisioning Core, and the co-chair of the UCDH Health Data Oversight Program, working to drive data driven insights to improve patient care and help establish UCDH as a top tier learning health system.

Dr. Adams runs a research lab focused on the use of artificial intelligence (AI) methods, physiologic sensors, and electronic health record data to improve the way we diagnose critical illnesses and predict clinical trajectories in the intensive care. His research lab is multidisciplinary with co-investigators and graduate students from Computer Science, Engineering, Biomedical Informatics, Pulmonary and Critical Care Medicine, Nursing, and Respiratory Therapy.




Health Data Science and Systems RLC