FHIR Fusion: Analytics on Combined Clinical and Imaging Data
summary
Open source dicomcast runs as a Linux container that reads a DICOM service change feed, matches patient IDs to an Azure API for FHIR server, and upserts ImagingStudy bundles so FHIR and DICOM join for queries. Demo extracts DICOM SR findings (radiation dose, observations), loads cohorts into Power BI (121 patients, 18 studies) and Azure ML, returns a model pneumothorax score ~80% and identifies cohort drug levothyroxine 50 mcg; quick deploy works in under three minutes.
AI-written summary · about
chapters (20)
- 0:00Introduction
- 2:02What is FHIR
- 2:42Use cases
- 4:17Imaging Study
- 5:33Linking Imaging Study to FHIR
- 9:20Examples
- 10:10Structured Reports
- 11:36Data in DICOM
- 13:13Data Extraction
- 19:14FHIR Services
- 20:11DICOMcast
- 22:59Demo
- 23:50Power BI
- 27:06Azure Machine Learning
- 32:49Whats Next
- 34:21Getting Started
- 35:56DICOM Server
- 37:11ML and Analytical Tools
- 37:44Other Microsoft Talks
- 38:38Questions
related talks
25mLesion tracking DICOM SR to FHIR mapping
34mML and analytics addressing nationwide COVID-19 impact and recovery
37mClinical Quality Improvement with FHIR
10mCloud-Native Data Engineering in Healthcare
10mTransforming Population Health with Predictive BI
55mSocial Listening for Richer Data, Unstructured Analytics
45mAzure DW and Power BI, reporting at the speed of light!
49mBuilding an NFL Scouting Dashboard with Azure Data Services