Scholar
Emily Alsentzer
Google Scholar ID: wKcw9Y8AAAAJ
Assistant Professor, Stanford University
Citations & Impact
Citations
4,584
H-index
18
i10-index
20
Publications
20
Co-authors
10
list available
Publications
Resume (English only)
Academic Achievements
- Received Best Oral Presentation Award at ISMB 2023 for work on few-shot diagnosis of rare disease patients; won Best Paper Award at CHIL 2023 for 'Do we still need clinical language models?'. Also published several papers in medRxiv, including assessments of racial and gender bias in GPT-4 for medical applications and zero-shot interpretable phenotyping of postpartum hemorrhage using large language models.
Research Experience
- Was a postdoctoral fellow at Brigham and Women’s Hospital and Harvard Medical School, working on deploying ML models within the Mass General Brigham healthcare system. During her PhD, she created ClinicalBERT, a language model trained on EHRs with millions of downloads on HuggingFace, and developed SHEPHERD, a GNN approach for diagnosing patients with rare genetic diseases in the Undiagnosed Disease Network.
Education
- PhD in Medical Engineering & Medical Physics (HST) from MIT & HMS in 2022, co-advised by Zak Kohane and Pete Szolovits; MS in Biomedical Informatics from Stanford University in 2017; BS in Computer Science from Stanford University in 2016.
Background
- Research interests include deployable machine learning, few-shot learning, LLMs & foundation models, graph neural networks, summarization, and rare disease diagnosis. She is an Assistant Professor of Biomedical Data Science and, by courtesy, of Computer Science at Stanford University. Her core research goal is to augment clinical decision making and broaden access to high quality healthcare through the use of ML and NLP.
Miscellany
- Her lab is recruiting students and postdocs to advance trustworthy, deployable AI methods for healthcare.
