Evaluating language model embeddings for Parkinson’s disease cohort harmonization using a novel manually curated variable mapping schema

Abstract Data Harmonization is an important yet time-consuming process. With the recent popularity of applications using Language Models (LMs) due to their high capabilities in text understanding, we investigated whether LMs could facilitate data harmonization for clinical use cases. To evaluate thi...

Full description

Saved in:
Bibliographic Details
Main Authors: Yasamin Salimi, Tim Adams, Mehmet Can Ay, Helena Balabin, Marc Jacobs, Martin Hofmann-Apitius
Format: Article
Language:English
Published: Nature Portfolio 2025-06-01
Series:Scientific Reports
Subjects:
Online Access:https://doi.org/10.1038/s41598-025-06447-2
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850113080538693632
author Yasamin Salimi
Tim Adams
Mehmet Can Ay
Helena Balabin
Marc Jacobs
Martin Hofmann-Apitius
author_facet Yasamin Salimi
Tim Adams
Mehmet Can Ay
Helena Balabin
Marc Jacobs
Martin Hofmann-Apitius
author_sort Yasamin Salimi
collection DOAJ
description Abstract Data Harmonization is an important yet time-consuming process. With the recent popularity of applications using Language Models (LMs) due to their high capabilities in text understanding, we investigated whether LMs could facilitate data harmonization for clinical use cases. To evaluate this, we created PASSIONATE, a novel Parkinson’s disease (PD) variable mapping schema as a ground truth source for pairwise cohort harmonization using LLMs. Additionally, we extended our investigation using an existing Alzheimer’s disease (AD) CDM. We computed text embeddings based on two language models to perform automated cohort harmonization for both AD and PD. We additionally compared the results to a baseline method using fuzzy string matching to determine the degree to which the semantic capabilities of language models can be utilized for automated cohort harmonization. We found that mappings based on text embeddings performed significantly better than those generated by fuzzy string matching, reaching an average accuracy of over 80% for almost all tested PD cohorts. When extended to a further neighborhood of possible matches, the accuracy could be improved to up to 96%. Our results suggest that language models can be used for automated harmonization with a high accuracy that can potentially be improved in the future by applying domain-trained models.
format Article
id doaj-art-13a27ffcc79b4a0ebd6bc7f7374c0a23
institution OA Journals
issn 2045-2322
language English
publishDate 2025-06-01
publisher Nature Portfolio
record_format Article
series Scientific Reports
spelling doaj-art-13a27ffcc79b4a0ebd6bc7f7374c0a232025-08-20T02:37:14ZengNature PortfolioScientific Reports2045-23222025-06-0115111310.1038/s41598-025-06447-2Evaluating language model embeddings for Parkinson’s disease cohort harmonization using a novel manually curated variable mapping schemaYasamin Salimi0Tim Adams1Mehmet Can Ay2Helena Balabin3Marc Jacobs4Martin Hofmann-Apitius5Department of Bioinformatics, Fraunhofer Institute for Algorithms and Scientific Computing (SCAI)Department of Bioinformatics, Fraunhofer Institute for Algorithms and Scientific Computing (SCAI)Department of Bioinformatics, Fraunhofer Institute for Algorithms and Scientific Computing (SCAI)Laboratory for Cognitive Neurology, Department of Neurosciences, KU LeuvenDepartment of Bioinformatics, Fraunhofer Institute for Algorithms and Scientific Computing (SCAI)Department of Bioinformatics, Fraunhofer Institute for Algorithms and Scientific Computing (SCAI)Abstract Data Harmonization is an important yet time-consuming process. With the recent popularity of applications using Language Models (LMs) due to their high capabilities in text understanding, we investigated whether LMs could facilitate data harmonization for clinical use cases. To evaluate this, we created PASSIONATE, a novel Parkinson’s disease (PD) variable mapping schema as a ground truth source for pairwise cohort harmonization using LLMs. Additionally, we extended our investigation using an existing Alzheimer’s disease (AD) CDM. We computed text embeddings based on two language models to perform automated cohort harmonization for both AD and PD. We additionally compared the results to a baseline method using fuzzy string matching to determine the degree to which the semantic capabilities of language models can be utilized for automated cohort harmonization. We found that mappings based on text embeddings performed significantly better than those generated by fuzzy string matching, reaching an average accuracy of over 80% for almost all tested PD cohorts. When extended to a further neighborhood of possible matches, the accuracy could be improved to up to 96%. Our results suggest that language models can be used for automated harmonization with a high accuracy that can potentially be improved in the future by applying domain-trained models.https://doi.org/10.1038/s41598-025-06447-2Alzheimer’s diseaseAutomatic data harmonizationParkinson’s diseaseLarge language modelsData stewardship
spellingShingle Yasamin Salimi
Tim Adams
Mehmet Can Ay
Helena Balabin
Marc Jacobs
Martin Hofmann-Apitius
Evaluating language model embeddings for Parkinson’s disease cohort harmonization using a novel manually curated variable mapping schema
Scientific Reports
Alzheimer’s disease
Automatic data harmonization
Parkinson’s disease
Large language models
Data stewardship
title Evaluating language model embeddings for Parkinson’s disease cohort harmonization using a novel manually curated variable mapping schema
title_full Evaluating language model embeddings for Parkinson’s disease cohort harmonization using a novel manually curated variable mapping schema
title_fullStr Evaluating language model embeddings for Parkinson’s disease cohort harmonization using a novel manually curated variable mapping schema
title_full_unstemmed Evaluating language model embeddings for Parkinson’s disease cohort harmonization using a novel manually curated variable mapping schema
title_short Evaluating language model embeddings for Parkinson’s disease cohort harmonization using a novel manually curated variable mapping schema
title_sort evaluating language model embeddings for parkinson s disease cohort harmonization using a novel manually curated variable mapping schema
topic Alzheimer’s disease
Automatic data harmonization
Parkinson’s disease
Large language models
Data stewardship
url https://doi.org/10.1038/s41598-025-06447-2
work_keys_str_mv AT yasaminsalimi evaluatinglanguagemodelembeddingsforparkinsonsdiseasecohortharmonizationusinganovelmanuallycuratedvariablemappingschema
AT timadams evaluatinglanguagemodelembeddingsforparkinsonsdiseasecohortharmonizationusinganovelmanuallycuratedvariablemappingschema
AT mehmetcanay evaluatinglanguagemodelembeddingsforparkinsonsdiseasecohortharmonizationusinganovelmanuallycuratedvariablemappingschema
AT helenabalabin evaluatinglanguagemodelembeddingsforparkinsonsdiseasecohortharmonizationusinganovelmanuallycuratedvariablemappingschema
AT marcjacobs evaluatinglanguagemodelembeddingsforparkinsonsdiseasecohortharmonizationusinganovelmanuallycuratedvariablemappingschema
AT martinhofmannapitius evaluatinglanguagemodelembeddingsforparkinsonsdiseasecohortharmonizationusinganovelmanuallycuratedvariablemappingschema