
Kexin Huang
@kexinhuang5
PhD Student @Stanford CS with @jure; AI + Biomedicine
ID: 1025489425495678977
http://www.kexinhuang.com 03-08-2018 21:11:46
500 Tweet
2,2K Takipçi
583 Takip Edilen

How to extract a clinical variable out of the entire patient’s history notes of > 200K words, using an expensive BERT-style model? Happy to share SnipBERT done Flatiron Health in NeurIPS ML4H! - up to >20% prediction gain - clues for model interpretation Details ⬇️
