Regulatory-Grade Clinical Data Abstraction on Databricks

Neha Pande, Senior Solution Architect, Healthcare & Life Sciences at Databricks and Veysel Kocaman, CTO at John Snow Labs, show what that architecture looks like in practice: Healthcare NLP for signal discovery, document ranking to reduce noise, medical terminology and provenance for traceability, and private Medical LLMs for reasoning. Chapters 00:00 Clinical Data Abstraction: The Challenge 04:04 Patient Journey Intelligence (PJI) Overview 09:38 Why Entity Extraction Alone Is Not Enough 14:55 Hybrid NLP + Medical LLM Architecture 27:46 Clinical Data Creation Workflow 31:23 Document Ranking and Evidence Selection 35:59 PJI + Databricks Live Demo 45:17 Benchmarks, Cost, and Key Takeaways Resources: Patient Journey Intelligence (PJI) overview: https://www.johnsnowlabs.com/patient-journey-intelligence/ Patient Journey Intelligence: https://www.johnsnowlabs.com/pji/ PJI architecture and documentation: https://www.johnsnowlabs.com/pji/getting_started/platform_architecture Clinical data integration and OMOP workflows: https://www.johnsnowlabs.com/pji/data_integration/overview The session also points viewers to PJI cookbooks, tutorials, and videos covering workflows such as FHIR-to-OMOP transformation, trial eligibility screening, and de-identification. #Databricks #JohnSnowLabs #PatientJourneyIntelligence #ClinicalNLP #MedicalAI #HealthcareAI #ClinicalAI