Building a Medical Device with R
summary
A precision-dosing medical device built in R implements pharmacology models with Bayesian updating to tailor antibiotic, oncology and transplant doses and supports over 100 pharmacology models by generating model packages from spec files using the open source pkpdsim engine. Verification uses testthat plus an in-house json2test tool that records and compares API payloads. The cloud API (OpenCPU, exploring plumber or AWS Lambdas) handles about a few hundred API calls an hour, integrates with Epic and Cerner, and is used by pharmacists in clinical workflow.
AI-written summary · about
chapters (6)
- 0:00Building medical devices with R
- 3:10Software as medical device
- 7:47Regulatory requirements and validation
- 11:43Architecture and API development
- 15:59Lessons and compliance
- 17:28Q&A on clinical integration
related talks
54mR Not Only In Production
57mGood Practices for Applied Machine Learning
05mTranslating clinical guidance to actionable insights with R
20mUsing R, Python, and Cloud Infrastructure to Battle Aquatic Invasive Species
01h 30mBayBE: A Bayesian Back End for Experimental Planning in the Low-To-No-Data Regime
41mUsing Data Riches: A Tale of two projects
39mSimulating a Million Patients: Realistic Health Data Generation
13mData Nerding in Public Health