MACHINE LEARNING & AI FOR HEALTHCARE: A HIMSS EVENT
Boston, MA, June 13-14, 2019
Simer Sodhi is the Director of Data Management & Analytics at Westchester Medical center University hospital, a regional trauma center providing health services to residents of Hudson Valley in New York, northern New Jersey, and southern Connecticut. Simer joined Westchester Medical Center in 2016, bringing over 15 years of Healthcare and Technology experience with her. Simer leads the Analytics solution delivery side of WMC and manages all business intelligence requirements. She is responsible for implementation and development of Data Analytics, Predictive Analytics and Population Health projects across WMC Health Network.
Before joining WMC, Simer was with Blue Cross Blue Shield of Massachusetts. She was instrumental in expanding the organization’s Business Intelligence and Analytics strategies. She spear headed implementation of visualization platform to accommodate organization’s growing demand for quick self-service analytics. Prior to BCBS, Simer held managerial roles at BearingPoint, Accenture and other consulting firms. Her consulting experience gave her immense allowed her to experience many fortune 100 environments and best practices that companies followed. Having worked with a diverse set of industries and variety of business intelligence and statistical tools, she brings in depth knowledge on Data Analytics, Data Mining, Visualizations, HealthCare Analytics, Machine Learning and Artificial Intelligence.
Many hospitals employ ambulatory care coordinators to coordinate the care of recently discharged patients and prioritize attention to those at high risk of 30-day readmission. But a lack of timely information and software to accurately predict readmission risk means ambulatory care managers often lack the tools to identify and prioritize high-risk patients. Valuable care management resources are spent manually pulling data from patient records to prioritize workloads and determine interventions.
A regional health system in New York remedied this by applying machine learning to multiple data sources and creating a risk model that identifies high- and low-risk patients more effectively. The payoff, as attendees will learn, has been improved organizational efficiency by enabling care managers to spend more time with those patients most in need of their care.
• Evaluate your organization’s ability to use machine learning tools to build a model for predicting readmission risk.
• Describe how the risk scores from such a model can be made actionable in the workflow of care managers.
• Demonstrate how a discharge platform that includes discharge lists and risk scores can be combined with EHR data to help care managers determine priorities and select interventions.