Advanced AI: NLP Techniques for Clinical Datasets
44mAdvanced2022-11-03
Authors
Wuraola Oyewusi
Wuraola Oyewusi is an experienced data scientist, machine learning, and artificial intelligence professional.
Course details
The healthcare industry is one of the fastest growing sectors using AI applications and techniques. When working with clinical data, written text forms a major part of how scenarios and treatment progression are documented. With the advent and availability of more and more digital health data, this course provides hands-on lessons at making sense of clinical text data using natural language processing (NLP) techniques. Join instructor Wuraola Oyewusi as she explores how to apply natural language processing to clinical and biomedical data. Topics covered include clinical named entity recognition, clinical entity resolution, word and sentence level text representation, and transformers for clinical text.
Skills covered
spaCyExplosionNatural Language Processing (NLP)Machine LearningAdvancedPythonData AnalysisArtificial Intelligence (AI)Data ScienceBusiness Analysis and StrategyBusiness Software and ToolsOpen Source
Concepts
0. Introduction
- 01 - Use NLP techniques for your data
- 02 - What you should know
- 03 - How to use the exercise files
1. Clinical Named Entity Recognition (CNER)
- 04 - What is clinical named entity recognition (CNER)
- 05 - Clinical named entity recognition using scispaCy
2. Clinical Entity Resolution
- 06 - What is clinical entity resolution
- 07 - Medical abbreviation resolution with scispaCy
- 08 - Entity linkage and resolution with a biomedical knowledge base
3. Clinical Text Representation
- 09 - What is clinical text representation
- 10 - Clinical text representation using fastText
- 11 - Clinical text representation using Universal Sentence Encoder (USE)
4. Transformers for Clinical Text
- 12 - What are transformers
- 13 - Clinical diagnosis prediction using transformers
- 14 - Clinical named entity recognition using transformers
- 15 - Clinical word prediction using transformers
Conclusion
- 16 - Next steps