Example Al use cases in healthcare / life sciences
2
Data has become the blocker to realizing the promise of Al at scale
3
Training data development is iterative- not a one-time process
4
Pain: bottlenecked by manual data labeling
5
Snorkel Flow unlocks the complete data-centric Al development workflow
6
With Snorkel Flow, Genentech built an Al application to extract entities with 99.1% accuracy
7
With Snorkel Flow, they built an application to identify demographic trends among patients.
8
Procedure Extraction
9
Deep support for the data types and ML tasks to power diverse use cases
10
Snorkel Flow integrates easily with your existing ML stack
Description:
Discover how leading healthcare and life science companies are accelerating NLP application development using data-centric AI workflows in this 25-minute presentation. Learn about Genentech and Memorial Sloan Kettering Cancer Research Center's success in reducing AI development time from months to days with Snorkel Flow. Explore real-world use cases, including entity extraction with 99.1% accuracy and demographic trend identification among patients. Gain insights into overcoming manual data labeling bottlenecks, implementing iterative training data development, and integrating data-centric AI workflows with existing ML stacks. Presented by Nazanin Makkinejad, a machine learning solutions engineer at Snorkel AI with extensive experience in deep learning and biomedical engineering.
How Top Healthcare Companies Boost AI Development with Data-Centric Approaches