Data Science in Healthcare: Beyond the Hype
summary
Slides catalog high-profile claims that later fail: IBM Watson touted as better at cancer diagnosis then found to give unsafe recommendations after a roughly$62 million deployment, and the Stanford pneumonia x‑ray result collapses on a flawed dataset. Practical work focuses on decision theory, confusion matrices and net‑benefit heatmaps, simple models (logistic regression, gradient boosting) and an operational 6‑month mortality dashboard that changes workflow and increases palliative consults. Other results include a patient‑feed intervention that reduced missed antibiotic doses by 1/3* and small wearable sensors for continuous vitals.
AI-written summary · about
chapters (10)
- 0:00Data science in healthcare
- 1:52Healthcare AI reality
- 4:05Electronic health records challenges
- 8:32Messy medical data
- 8:52Airbag vs jetpack in models
- 12:15Model impact with decision theory
- 16:21Clinician collaboration
- 17:05Palliative care prediction
- 20:40Future directions
- 24:20Q&A
related talks
08mRevolutionizing Healthcare with Big Data and ML
09mHow AI is changing the healthcare system
59mKeynote Panel AI in Healthcare
23mEngineering AI-Powered Predictive Healthcare Systems
21mTransforming Healthcare Through Data-Driven Intelligence
22mLeveraging AI/ML for Transformative Software Solutions in Healthcare
15mTransforming Healthcare with BI & AI
17mAI-Ready Healthcare Data: Revolutionizing Fraud Detection