Welcome to the BioGears conference! We are very excited to support our user and application development communities as well as the broader medical modeling and simulation community.
IMPORTANT INFORMATION. THIS CONFERENCE HAS CHANGED TO A VIRTUAL WEBEX CONFERENCE. SCHEDULE WILL BE THE SAME AS DISPLAYED ON THIS SITE.
Meeting password: BioGears2ndConf! Join by phoneTap to call in from a mobile device (attendees only)1-877-668-4493 Call-in toll-free number (US/Canada)1-650-479-3208 Call-in toll number (US/Canada)
This talk will detail how to use BioGears as a tool to create simulated Electronic Health Record data. Problems and possible future directions will be discussed.
Abstract:
Reliable machine learning (ML) algorithms require a training dataset free from noise and mislabeling. Electronic Health Records (EHR), the largest source of time series data on patient healthcare, often suffer from these common pitfalls, limiting their utility in ML development. The 4TDS project is partnering with BioGears to build an ML algorithm that predicts the onset of septic shock from a patient’s vital signs. By supplementing existing EHR datasets with BioGears patient simulations we are addressing important data quality issues and opening up the scope of AI in healthcare.
