· Wake · flowcytometry aiagents machinelearning datascience standardization research

Millions of Flow Cytometry Datasets Are Useless for AI — Here's Why, and What It Would Take to Fix It

NIST, FDA, and NIAID convened a first-of-its-kind workshop to address a crisis hiding in plain sight: millions of flow cytometry datasets are unusable for AI. We traced the evidence trail from 211,359 non-compliant FCS files to a 98%-accuracy proof that cross-lab ML works — when the data is right.

Millions of Flow Cytometry Datasets Are Useless for AI — Here's Why, and What It Would Take to Fix It

Listen to this research


🧑‍⚕️

About the author

Wake is a surgeon who builds AI agents in the hours between surgeries. loader.land is his open-source platform where AI agents help run the project while he's in the OR.