Peer-Reviewed Publication
Annu Rev Biomed Data Sci2025;8(1):563-590.August 1, 2025Journal Article
Embedding Methods for Electronic Health Record Research.
Justin Kauffman1,2, Riccardo Miotto3, Eyal Klang1,2, Anthony Costa4, Beau Norgeot5, Marinka Zitnik6, Shameer Khader7,8, Fei Wang9, Girish N Nadkarni1,2, Benjamin S Glicksberg1,2,10
1Hasso Plattner Institute for Digital Health, Icahn School of Medicine at Mount Sinai, New York, NY, USA; email: benjamin.glicksberg@mssm.edu.
2Windreich Department of Artificial Intelligence and Human Health, Icahn School of Medicine at Mount Sinai, New York, NY, USA.
3Tempus AI, Inc., Chicago, Illinois, USA.
4NVIDIA, Santa Clara, California, USA.
5Qualified Health, Palo Alto, California, USA.
6Department of Biomedical Informatics, Harvard Medical School, Boston, Massachusetts, USA.
7School of Public Health, Faculty of Medicine, Imperial College of London, London, UK.
8Current affiliation: Precision Medicine and Computational Biology, Sanofi, Cambridge, Massachusetts, USA.
9Department of Population Health Sciences, Weill Cornell Medical College, New York, NY, USA.
10Mindich Child Health and Development Institute, Icahn School of Medicine at Mount Sinai, New York, NY, USA.
Abstract
This review aims to elucidate the role and impact of embedding techniques in the analysis and utilization of electronic health record data for research. By integrating multidimensional, incongruent, and often unstructured medical data for machine learning models, embeddings provide a powerful tool for enhancing data utility, especially under certain conditions and for asking certain questions. We…
Create a free account to keep reading
Free members get 10 full research views every month across publications, clinical trials, FDA clearances, adverse events, and NIH grants. No credit card required.
Want unlimited research access? See Pro plans