Resources


Model Credentialed Access

EntityBERT: BERT-based Models Pretrained on MIMIC-III with or without Entity-centric Masking Strategy for the Clinical Domain

Chen Lin, Steven Bethard, Guergana Savova, et al.

Pretraining of models with a broad representation of biomedical terminology (PubMedBERT) on MIMIC-III corpus along with or without a novel entity-centric masking strategy.

Published: March 17, 2022. Version: 1.0.1

Model Credentialed Access

Clinical BERT Models Trained on Pseudo Re-identified MIMIC-III Notes

Eric Lehman, Sarthak Jain, Karl Pichotta, et al.

We explore recovering sensitive info from BERT trained over non-deidentified EHR. We make our models and data available to further facilitate research.

Published: April 28, 2021. Version: 1.0.0

Model Credentialed Access

Clinical-T5: Large Language Models Built Using MIMIC Clinical Text

Eric Lehman, Alistair Johnson

We train a T5-Base and T5-Large from scratch on MIMIC-III and MIMIC-IV. Additionally, we further pretrain T5-Base and SciFive on notes from MIMIC. We release these model weights on PhysioNet.

Published: Jan. 25, 2023. Version: 1.0.0

Database Credentialed Access

MIMIC-Ext-SynNotes: Large-Scale Synthetic Clinical Notes via Rephrasing MIMIC-III/IV with LLMs

Jinghui Liu, Sarvesh Soni, Anthony Nguyen

Large-scale synthetic clinical notes, produced by rephrasing all MIMIC-III and MIMIC-IV with open-weight LLMs to support clinical language model development.

natural language processing data augmentation mimic large language models synthetic data pretraining clinical notes

Published: Oct. 5, 2026. Version: 1.0