PatientPheX: A Full-Text Biomedical Corpus for Patient-Specific Phenotype Information Extraction
- 1Dalian University of Technology
Abstract
Patient-specific phenotype information extraction identifiers and normalizes phenotype entities and links them with individual patients, supporting rare disease research and genotype-phenotype analysis. However, existing datasets rarely combine full-text biomedical content with detailed annotations of patients, phenotypes, related entities, and their associations. In this work, we present PatientPheX, a corpus for patient-specific phenotype information extraction comprising 200 full-text articles, primarily case reports and case series. The corpus provides mention annotations and normalization for seven biomedical entity types, together with explicit associations between patients and entities across sentences, paragraphs, and sections. Negated findings and composite phenotype mentions are also annotated. Corpus construction combines automated pre-annotation, large language model (LLM)-assisted verification, and manual review and adjudication. Benchmarks cover entity recognition and normalization and patient-phenotype association extraction using gold-standard and predicted entities. The overall document-level entity F1 was 86.95%, while phenotype F1 was 79.33%. The best relation extraction macro-F1 decreased from 84.14% with gold-standard entities to 54.98% with predicted entities, highlighting the combined challenges of phenotype extraction and patient attribution. PatientPheX provides a resource for developing and evaluating patient-centered information extraction methods and supports the construction of structured patient-level knowledge resources for phenotype-driven biomedical research.
Related
- Fucoidan and alginate from brown seaweeds: extraction, structural diversity, biocompatibility, biodegradability, and biomedical applications
- DBCLS BioHackathon 2026 report: Open, Reproducible Extraction of Data-Entity References from the Biomedical Literature – with Entry-Level Resolution via Identifiers.org / TogoID
- Information processing-like phenotype of human brain organoids
- Activation of a tissue-specific stress response in the aqueous outflow pathway of the eye defines the glaucoma disease phenotype
- Shared genetic architecture between adhesive capsulitis and type 2 diabetes: a phenotype-specific cross-trait analysis
BibTeX
@misc{zhu2026patientphex,
title = {{PatientPheX: A Full-Text Biomedical Corpus for Patient-Specific Phenotype Information Extraction}},
author = {X. Zhu and L. Luo and Z. Wang and C. Huang and Y. Li and Y. Zhong and T. Lv and J. Li and C. Liu and F. Tian and Y. Sun and J. Wang and H. Lin},
howpublished = {bioRxiv},
year = {2026},
month = oct,
doi = {10.64898/2026.10.07.757319},
url = {https://www.biorxiv.org/content/10.64898/2026.10.07.757319v1},
note = {CC BY},
}
RIS
TY - UNPB
TI - PatientPheX: A Full-Text Biomedical Corpus for Patient-Specific Phenotype Information Extraction
AU - X. Zhu
AU - L. Luo
AU - Z. Wang
AU - C. Huang
AU - Y. Li
AU - Y. Zhong
AU - T. Lv
AU - J. Li
AU - C. Liu
AU - F. Tian
AU - Y. Sun
AU - J. Wang
AU - H. Lin
PY - 2026
DA - 2026/10/07
PB - bioRxiv
DO - 10.64898/2026.10.07.757319
UR - https://www.biorxiv.org/content/10.64898/2026.10.07.757319v1
AB - Patient-specific phenotype information extraction identifiers and normalizes phenotype entities and links them with individual patients, supporting rare disease research and genotype-phenotype analysis. However, existing datasets rarely combine full-text biomedical content with detailed annotations of patients, phenotypes, related entities, and their associations. In this work, we present PatientPheX, a corpus for patient-specific phenotype information extraction comprising 200 full-text articles, primarily case reports and case series. The corpus provides mention annotations and normalization for seven biomedical entity types, together with explicit associations between patients and entities across sentences, paragraphs, and sections. Negated findings and composite phenotype mentions are also annotated. Corpus construction combines automated pre-annotation, large language model (LLM)-assisted verification, and manual review and adjudication. Benchmarks cover entity recognition and normalization and patient-phenotype association extraction using gold-standard and predicted entities. The overall document-level entity F1 was 86.95%, while phenotype F1 was 79.33%. The best relation extraction macro-F1 decreased from 84.14% with gold-standard entities to 54.98% with predicted entities, highlighting the combined challenges of phenotype extraction and patient attribution. PatientPheX provides a resource for developing and evaluating patient-centered information extraction methods and supports the construction of structured patient-level knowledge resources for phenotype-driven biomedical research.
ER -