Hybrid Semantic-Lexical Retrieval for Source-to-CDE Mapping in the NCI caDSR
- 1National Cancer Institute, NCI Shady Grove, 9609 Medical Center Drive, Rockville, MD 20850, USA
Abstract
Accurate mapping of local source data elements to standardized common data elements (CDEs) is critical for biomedical data integration but is challenged by short, abbreviated, and heterogeneous source descriptions. We constructed a benchmark of approximately 69,000 expert-linked source-to-CDE mappings from the National Cancer Institute\textquotesingle s caDSR and evaluated multiple bi-encoder models with two-stage fine-tuning. Here we developed a hybrid retrieval system combining semantic bi-encoder retrieval, CDE Match-inspired keyword retrieval, cross-encoder scoring, and supervised reranking. This system achieved Recall@5 ranging from 0.802 to 0.972 across internal and five distribution-shifted holdouts, outperforming the other evaluated methods on five of six datasets and approaching ceiling performance on the sixth dataset (GDC). These results demonstrate the effectiveness and robustness of integrating semantic and lexical evidence for source-to-CDE mapping, supporting improved interoperability and reuse of biomedical metadata across diverse clinical and research contexts.
Related
- Frequency-specific oscillations track parallel processing of linguistic structural and lexical-semantic information from perception to decision
- The Semantic Shift in Autobiographical Memory and Aging: Do Retrieval Goals Contribute to Age-Related Memory Differences?
- Mapping Gene Expression to an Interpretable Semantic Space
- Mapping Artificial Intelligence Research in Aphasia: An AI-Assisted Scoping Review and Semantic Network Analysis
- Optical genome mapping identifies source-associated structural variant differences across early-passage human iPSCs
BibTeX
@misc{tanis2026hybrid,
title = {{Hybrid Semantic-Lexical Retrieval for Source-to-CDE Mapping in the NCI caDSR}},
author = {J. H. Tanis and D. B. Warzel and R. P. Khanna and Q. Chen and C. Yan and R. Velleleth and M. T. Nash and D. Manathara and B. B. Biggers and J. B. Soares and D. Meerzaman},
howpublished = {bioRxiv},
year = {2026},
month = oct,
doi = {10.64898/2026.10.14.758967},
url = {https://www.biorxiv.org/content/10.64898/2026.10.14.758967v1},
note = {CC BY},
}
RIS
TY - UNPB
TI - Hybrid Semantic-Lexical Retrieval for Source-to-CDE Mapping in the NCI caDSR
AU - J. H. Tanis
AU - D. B. Warzel
AU - R. P. Khanna
AU - Q. Chen
AU - C. Yan
AU - R. Velleleth
AU - M. T. Nash
AU - D. Manathara
AU - B. B. Biggers
AU - J. B. Soares
AU - D. Meerzaman
PY - 2026
DA - 2026/10/14
PB - bioRxiv
DO - 10.64898/2026.10.14.758967
UR - https://www.biorxiv.org/content/10.64898/2026.10.14.758967v1
AB - Accurate mapping of local source data elements to standardized common data elements (CDEs) is critical for biomedical data integration but is challenged by short, abbreviated, and heterogeneous source descriptions. We constructed a benchmark of approximately 69,000 expert-linked source-to-CDE mappings from the National Cancer Institute\textquotesingle s caDSR and evaluated multiple bi-encoder models with two-stage fine-tuning. Here we developed a hybrid retrieval system combining semantic bi-encoder retrieval, CDE Match-inspired keyword retrieval, cross-encoder scoring, and supervised reranking. This system achieved Recall@5 ranging from 0.802 to 0.972 across internal and five distribution-shifted holdouts, outperforming the other evaluated methods on five of six datasets and approaching ceiling performance on the sixth dataset (GDC). These results demonstrate the effectiveness and robustness of integrating semantic and lexical evidence for source-to-CDE mapping, supporting improved interoperability and reuse of biomedical metadata across diverse clinical and research contexts.
ER -