⛳ Software for Clinical Natural Language Processing

This is a summary of openly available implementations (software deliverables) created in the Accumulate project at CLiPS.

List of repositories

Text Normalization (deliverable 2.2)

• clinspell: Clinical spelling correction with word and character n-gram embeddings
  

Concept Extraction and Disambiguation (deliverable 2.3)

• yarn: Disambiguating biomedical and clinical concepts with word embeddings
  

• conch: Extracting medical concepts and terminology from patient records, and linking them to UMLS CUI
  

Evaluation of extracted terminology (task 2.5)

• RDOC challenge: Psychiatric symptom severity identification. (Joint work with UZA)

Event Semantics (deliverables 3.1-3.3)

• clicr: Machine reading and question answering on clinical case reports

• memory-networks: Memory networks for medical machine reading

• relex: Relation extraction between problem, treatment and test mentions in clinical notes with background knowledge integration.
  

Patient Representation Learning (precursor to task 5.2)

Representation Learning: Unsupervised representation learning. which is capable of learning patient representations for end tasks like mortality prediction, and diagnosis and procedural identification (Joint work with UZA).

Explaining clinical neural models for engendering trust during valorization

Interpret RNNs: Rule-based explanations of LSTM and GRU classifiers for sepsis classification.

Contributors

Walter Daelemans, Pieter Fivez, Simon Šuster, Stéphan Tulkens, Madhumita Sushil

Licensing

See individual repositories.