Towards Patient-Driven Phenotyping and Similarity for Precision Medicine (Project Write-up)
Authors/Creators
- 1. Computational Bioscience Program, University of Colorado Anschutz Medical Campus
- 2. Lister Hill National Center for Biomedical Communications, National Library of Medicine, National Institutes of Health
Description
Clinical phenotyping techniques provide clinicians and clinical researchers with important insight into the treatment of rare and complex diseases for a population of patients. Many, traditional methods require multiple iterations of refinement with a domain expert, lack generalizability and interoperability, and often have limited reproducibility. Compared to traditional phenotyping approaches, patient similarity-based techniques aim to derive personalized patient risk models. Although relatively new, these methods have already been shown to be more accurate, even when applied to sparse data or poorly characterized diseases/phenotypes. Here, we present preliminary results from an initial evaluation of a novel data-driven method for calculating patient similarity.
Files
2017_NLM_Final_paper.pdf
Additional details
Related works
- Cites
- Presentation: 10.5281/zenodo.5718875 (DOI)
- Has part
- Presentation: 10.5281/zenodo.5718867 (DOI)
- References
- Software: https://gist.github.com/callahantiff/597ed54a684da2e7ac832c5bb57c162a (URL)