Dr Prakash Data Science Manager, ZS Associates at Cypher 2017
summary
Case study shows patient centricity work using integrated claims, EHR and pharmacy data to predict therapy dropout and disease events; pipeline segments patients into early-drop, mid-drop and compliant cohorts via longitudinal clustering, uses random forest to surface drivers (payer switches, claim rejections, copay utilization, age, pharmacy channel) and builds embeddings fed to a bi-directional LSTM to predict cancer onset within three months — models identify actionable drivers and enable prescriptive interventions though exact accuracy numbers are not reported.
AI-written summary · about
chapters (6)
- 0:00Start
- 3:32Customer Centricity
- 6:57High Deductible Plans
- 8:35Patient Centricity
- 20:26Patient Journey
- 24:00Challenges with the Patient Journey
related talks
33mData Science in Healthcare: Beyond the Hype
10mTransforming Population Health with Predictive BI
59mKeynote Panel AI in Healthcare
21mTransforming Healthcare Through Data-Driven Intelligence
22mBuilding Predictive Analytics Data Warehouses in Healthcare
23mEngineering AI-Powered Predictive Healthcare Systems
17mAI-Ready Healthcare Data: Revolutionizing Fraud Detection
03mRichard Seiersen from Kaiser Permanente interviewed at Strata Rx 2013