Transforming Population Health with Predictive BI
summary
Predictive stack ingests EHRs, claims and SDOH into 8.5 million patient records and 360° profiles, running ETL pipelines, secure API gateways and HL7 FHI compliance into models (random forest, gradient boosting) over 3,000 variables; results claim 87–92% prediction accuracy for six‑month hospitalization, 25% fewer hospitalizations, 20% lower readmissions, 30% fewer ER visits, 20% cost reduction and a 42% drop in diabetic admissions saving $3.2M while GIS catchment modeling improves facility placement.
AI-written summary · about
chapters (11)
- 0:00Introduction to Predictive Analytics in Healthcare
- 0:42The Population Health Challenge
- 1:45Creating a 360-Degree Patient View
- 2:48Risk Stratification Models
- 3:54Impact of Predictive Analytics
- 4:46Geospatial Healthcare Analytics
- 5:46Machine Learning Applications in Healthcare
- 6:39Case Study: Chronic Disease Management
- 7:28Addressing Healthcare Disparities
- 8:13Implementation Roadmap for Predictive Analytics
- 9:08Key Takeaways and Conclusion
related talks
22mBuilding Predictive Analytics Data Warehouses in Healthcare
52mBig Data - Top Ingredient to Improve Healthcare
15mTransforming Healthcare with BI & AI
33mData Science in Healthcare: Beyond the Hype
54mCrossing the Last Mile with Predictable Hospital Readmissions
08mRevolutionizing Healthcare with Big Data and ML
10mCloud-Native Data Engineering in Healthcare
27mDr Prakash Data Science Manager, ZS Associates at Cypher 2017