Biomedical Information Extraction

Python tools primarily intended for bioinformatics and computational molecular biology purposes, but also a convenient way to obtain data, including documents/abstracts from PubMed (see Chapter 9 of the documentation).

A framework for biomedical coreference resolution.

A system for building predictive medical natural language processing models. Built on the spaCy framework.

A version of the spaCy framework for scientific and biomedical documents.

R utilities for accessing NCBI resources, including PubMed.

a Python package and model (for use with spaCy) for doing NER with medication-related concepts.

A system for processing the text in electronic medical records. Widely used and open source.

A system for processing documents describing cancer presentations. Based on cTAKES (see above).

A framework for running text mining tools on the newest set(s) of documents from PubMed.

an IE infrastructure for electronic health records (EHR). Built on the CogStack project.

a framework for IE from tables in the literature.

A PubMed and PubMed Central-trained version of the BERT language model.

A PubMed and PubMed Central-trained version of the BERT language model.

A BERT model trained on >1M papers from the Semantic Scholar database.

A BERT model pre-trained on PubMed text and MIMIC-III notes.

A GPT-2 model pre-trained on 15 million PubMed abstracts, along with fine-tuned versions for several biomedical tasks.

Word embeddings derived from biomedical text (>27 million PubMed titles and abstracts), including subword embedding model based on MeSH.

A corpus of scholarly manuscripts concerning COVID-19. Articles are primarily from PubMed Central and preprint servers, though the set also includes metadata on papers without full-text availability.

paper - 120 full text articles annotated for PPI and genetic interactions. Used in the BioCreative V BioC task.

paper - 77 sentences from research articles about the bacterium Bacillus subtilis, annotated for protein–gene interactions (so, fairly close to PPI annotations). Additional information is here.

paper - 77 sentences from research articles about the bacterium Bacillus subtilis, annotated for protein–gene interactions (so, fairly close to PPI annotations). Additional information is here.

The MIMIC Chest X-Ray database. Contains more than 377,000 radiographic images and accompanying free-text radiology reports. As with MIMIC-III, requires acceptance of a data use agreement.

Deidentified health data from ~60,000 intensive care unit admissions. Requires completion of an online training course (CITI training) and acceptance of a data use agreement prior to use.

An update to MIMIC-III's multimodal patient data, now covering more recent years of admissions, plus a new data structure, emergency department records, and links to MIMIC-CXR images.

a database of observations from more than 200 thousand intensive care unit admissions, with consistent structure. Requires registration, training course completion, and data use agreement.

An ontology of human diseases. Has cross-links to MeSH, ICD, NCI Thesaurus, SNOMED, and OMIM. Public domain. Available on GitHub and on the OBO Foundry.

A data model of biological entities. Provided as a YAML file.

a standard for observational healthcare data.

A language model available through the Flair framework and embedding method. Trained over a 5% sample of PubMed abstracts until 2015, or > 1.2 million abstracts in total.