Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by

Same author

RankVar: machine learning-based variant ranking and reinterpretation for rare genetic diseases.

Genome medicine·2026

Same author

Improving Retrieval-Augmented Generation without Taxonomy-based Error Categorization.

Proceedings of the conference. Association for Computational Linguistics. Meeting·2026

Same author

Evaluating unsupervised and rule-based phenotyping methods versus administrative code counts for systemic sclerosis identification.

Seminars in arthritis and rheumatism·2026

Same author

Hard to Halt: Automation Bias in Agent-Driven Sequencing Prior Authorization Workflows.

medRxiv : the preprint server for health sciences·2026

Same author

Unsupervised characterization of 100,272 EHR patients identifies high-risk groups and comorbidities linked to premature aging.

NPJ digital medicine·2026

Same author

TimeX: Phenotype Onset Extraction from Clinical Narratives.

npj health systems·2026

Search research articles

Related Experiment Video

Updated: Jul 5, 2025

Augmenting Large Language Models via Vector Embeddings to Improve Domain-Specific Responsiveness

Augmenting Large Language Models via Vector Embeddings to Improve Domain-Specific Responsiveness

Published on: December 6, 2024

Enhancing phenotype recognition in clinical notes using large language models: PhenoBCBERT and PhenoGPT.

Jingye Yang^1,2, Cong Liu³, Wendy Deng¹

¹Raymond G. Perelman Center for Cellular and Molecular Therapeutics, Children's Hospital of Philadelphia, Philadelphia, PA 19104, USA.

Patterns (New York, N.Y.)

|January 24, 2024

Summary

New AI models, PhenoBCBERT and PhenoGPT, improve recognition of genetic disease phenotypes in clinical notes by expanding Human Phenotype Ontology (HPO) terms, aiding disease research.

Keywords:

BERT GPT Human Phenotype Ontology clinical notes electronic health records named entity recognition transformer

More Related Videos

A Metadata Extraction Approach for Clinical Case Reports to Enable Advanced Understanding of Biomedical Concepts

A Metadata Extraction Approach for Clinical Case Reports to Enable Advanced Understanding of Biomedical Concepts

Published on: September 20, 2018

Predicting Treatment Response to Image-Guided Therapies Using Machine Learning: An Example for Trans-Arterial Treatment of Hepatocellular Carcinoma

Predicting Treatment Response to Image-Guided Therapies Using Machine Learning: An Example for Trans-Arterial Treatment of Hepatocellular Carcinoma

Published on: October 10, 2018

Related Experiment Videos

Last Updated: Jul 5, 2025

Augmenting Large Language Models via Vector Embeddings to Improve Domain-Specific Responsiveness

Augmenting Large Language Models via Vector Embeddings to Improve Domain-Specific Responsiveness

Published on: December 6, 2024

A Metadata Extraction Approach for Clinical Case Reports to Enable Advanced Understanding of Biomedical Concepts

A Metadata Extraction Approach for Clinical Case Reports to Enable Advanced Understanding of Biomedical Concepts

Published on: September 20, 2018

Predicting Treatment Response to Image-Guided Therapies Using Machine Learning: An Example for Trans-Arterial Treatment of Hepatocellular Carcinoma

Predicting Treatment Response to Image-Guided Therapies Using Machine Learning: An Example for Trans-Arterial Treatment of Hepatocellular Carcinoma

Published on: October 10, 2018

Area of Science:

Computational Biology
Medical Informatics
Genetics

Background:

Human Phenotype Ontology (HPO) provides a standardized vocabulary for phenotypes in genetic diseases.
Existing phenotype recognition tools often have limitations in capturing the full spectrum of phenotypic abnormalities.
Traditional heuristic or rule-based approaches struggle with novel or complex phenotype descriptions.

Purpose of the Study:

To develop advanced models for automated phenotype recognition in clinical notes.
To expand the vocabulary of Human Phenotype Ontology (HPO) terms using large language models.
To improve the detection of both known and previously uncharacterized phenotype concepts.

Main Methods:

Development of two novel models: PhenoBCBERT (BERT-based) and PhenoGPT (GPT-based).
Leveraging large language models to automate the detection and expansion of HPO terms from clinical text.
Comparative analysis against existing tools like PhenoTagger.
Evaluation through case studies on biomedical literature and assessment of model architectures and accuracy.

Main Results:

PhenoBCBERT and PhenoGPT identified a wider range of phenotype concepts compared to existing tools.
The models successfully detected phenotypes not present in the current HPO vocabulary.
Demonstrated strong performance in case studies involving biomedical literature.
Comparative analysis highlighted the strengths and weaknesses of BERT- and GPT-based architectures for this task.

Conclusions:

The developed models significantly enhance automated phenotype detection from clinical texts.
These advancements improve the accuracy and comprehensiveness of phenotype recognition in genetic diseases.
The models facilitate more robust downstream analyses for human disease research and understanding.