Workflow Integration in AI in Healthcare
summary
GE builds Edison AI to embed models into imaging workflows and claims concrete gains: a smart scheduling model that predicts no-shows per patient (site-fine-tuned, examples like 88% likelihood), an Intelligent Protocol Manager that suggests and learns CT/MR protocols, AIRx that auto-defines MRI acquisition planes to cut variability, and AIR Recon DL that uses deep learning to reconstruct images faster and cleaner. Mobile X‑ray runs quality checks and pneumothorax detection on-device. Platform routes DICOM studies via Open AI Orchestrator to third‑party AI, prioritizes worklists, and returns results; note $40B invested and 350 healthcare AI companies in 2019.
AI-written summary · about
chapters (18)
- 0:00Introduction
- 1:23Building AI in Healthcare
- 6:37Where is AI being spent
- 8:22Clinical workflows are complex
- 10:16How can we make AI invisible
- 12:30Noshows
- 15:17Intelligent Protocoling
- 17:06AirX
- 19:19Ultrasound
- 20:48MRI
- 22:52Reporting
- 25:43Interpretation
- 26:42Funding
- 27:27Impact of 350 companies
- 28:49The Edison ecosystem
- 29:26Open AI Orchestrator
- 31:13AI Orchestrator
- 33:01Summary
related talks
09mHow AI is changing the healthcare system
47mThe Future of Healthcare and AI
23mAI in the Life Science Sector
03mAI solutions for healthcare
35mHow may AI help you? - healthcare
01h 00mKeynote Panel: AI in Healthcare
47mAI for making decisions in clinical medicine
28mThe Future of Healthcare – AI’s Promise and the Challenges Ahead