Novice to data scientist: a pediatric anesthesiologist uses RStudio to help kids access surgical ca
summary
Self-taught analysis of thousands of pediatric records shows a 5% day‑of‑surgery cancellation rate (about ~$1 per second lost) and uses R Studio/tidyverse, ggplot, caret for ML, geocoding and tidycensus geospatial data. Models predict no‑shows and NPO violations well (patient illness is harder to predict); the strongest predictor is prior cancellation. Maps reveal hotspots including Northern Kentucky and link cancellations to socioeconomic disadvantage and race. Colorful instructions, text reminders and pre‑op calls cut OR downtime ~17% (about an hour a day).
AI-written summary · about
chapters (12)
- 0:00Introduction
- 1:51Surgery cancellations impact
- 3:37Root causes in children
- 5:48Excel data challenges
- 7:16R and data exploration
- 9:28Meeting Emily’s mom and discovering machine learning
- 10:35Predictive models
- 11:43Patient-level insights
- 13:09Geospatial disparities
- 15:40Geospatial ML rates
- 16:46RStudio impact
- 18:31Q&A
related talks
24mUsing R Bridge the Gap in Electronic Medical Record Reporting
19mData science education as a public health intervention in E. Baltimore
38mThe Real World of Data Analytics
23mGrowth Hacking - Product Analytics at Scale using R and RStudio
07mBig Data and Cardiac Rehabilitation
20mData-Driven Approach in Treating a Medical Condition
01h 51mMorning Session: Genomics, Medicine, Epidemiology and Data Interpretation
54mR Not Only In Production