Clinical Concept Extraction on i2b2 2010 (test)
90.25Exact F1BERTlarge
Evaluation results
| Method | Links | |
|---|---|---|
| BERTlargePre-training Corpus=MIMIC2019.02 | 90.25 | |
| BERTbasePre-training Corpus=MIMIC2019.02 | 89.55 | |
| Baseline_BlueBERTBackbone=BlueBERT2026.03 | 89.2 | |
| Prior SOTA (Zhu et al., 2018)2019.02 | 88.6 | |
| ELMoPre-training Corpus=MIMIC2019.02 | 87.8 | |
| Alsentzer 2019b2026.03 | 87.8 | |
| Lee 20202026.03 | 86.7 | |
| BERTlargePre-training Corpus=General2019.02 | 85.48 | |
| i2b2/VA Best2026.03 | 85.2 | |
| GloVePre-training Corpus=MIMIC2019.02 | 85.07 | |
| BioBERTPre-training Corpus=General2019.02 | 84.76 | |
| BERTbasePre-training Corpus=General2019.02 | 84.33 | |
| word2vecPre-training Corpus=MIMIC2019.02 | 84.32 | |
| fastTextPre-training Corpus=MIMIC2019.02 | 84.19 | |
| GloVePre-training Corpus=General2019.02 | 84.08 | |
| ELMoPre-training Corpus=General2019.02 | 83.83 | |
| fastTextPre-training Corpus=General2019.02 | 83.46 | |
| word2vecPre-training Corpus=General2019.02 | 80.38 |