Elamur Research⌕
☰
home / papers / 2026 / FHexchange: Resources for Family Health History Extraction and Normalization From Consumer Dialog Sources

FHexchange: Resources for Family Health History Extraction and Normalization From Consumer Dialog Sources

Paper · 2026 · Michelle Nguyen, Nidhi Soley, Ayah Zirikly et al.

Summary

Family health history (FHx) helps estimate a person's disease risk, but it is usually recorded as free text. Existing FHx corpora mostly come from clinician-written notes, are often proprietary or restricted, and rarely include normalized concepts. This gap matters more now that ambient AI scribes and chatbots collect histories through conversation. The authors present FHexchange, a public benchmark of patient-reported FHx dialogs for entity extraction, relation extraction and entity linking. It has two parts. FHexchange-KIT contains real user transcripts from the KIT chatbot. FHexchange-MTS contains synthetic patient–doctor conversations taken from MTS-Dialog. Annotators labeled family members, clinical observations, modifiers such as ages, living status and CauseofDeath, and relations, and they mapped observations to UMLS CUIs. Baseline tests used proprietary models (including GPT-5-mini) and the open-weight Llama 3.3. The models found family members reliably but missed age and cause-of-death modifiers, and CUI linking remained weak. The comparison also suggests that the way FHx is collected shapes what patients reveal. FHexchange therefore offers a testbed for conversational clinical AI.

Key points

  • Two subsets: FHexchange-KIT has 104 real chatbot–user dialogs from a crowdsourced study. FHexchange-MTS has 154 FHx-related dialogs selected from the 1,701 synthetic conversations in MTS-Dialog. Both are in English.
  • Access: The base entity and relation annotations are freely available at github.com/tirilab/FHexchange. The UMLS CUI links, produced with sciSpacy's entity linker, require a UMLS license.
  • Baseline errors: Models often rewrote lay wording as formal terms, for example turning "fast heart beat" into tachycardia. Under strict token matching these count as errors, so scores are underestimated. GPT-5-mini was cautious when assigning CUIs, which cut both wrong and correct links.
  • Prompting: Few-shot prompting improved Llama 3.3's FHx extraction.
  • Disclosure: Two of the five most frequent conditions in FHexchange-KIT were mental-health related. This hints that patients may share stigmatized information more readily with a chatbot.

Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.

 Institutions

🎓Johns Hopkins UniversityBaltimore · United States240 works · 96,487 citations🎓New York UniversityNew York · United States201 works · 100,180 citations🎓George Washington UniversityWashington · United States51 works · 8,148 citations

 Fields, topics and concepts

📚 Shelves 6The field of science this work belongs to (OpenAlex fields).Computer Science · 16,667 worksArtificial Intelligence · 16,194 worksBiochemistry, Genetics and Molecular Biology · 234 worksMolecular Biology · 231 worksHealth Professions · 69 worksHealth Information Management · 50 works

🧭 Topics 1Research subjects and languages the work is about.Dialogue systems · 624 works

Cite

Cite as Michelle Nguyen, Nidhi Soley, Ayah Zirikly, João Sedoc and Casey Taylor. 2026. FHexchange: Resources for Family Health History Extraction and Normalization From Consumer Dialog Sources. BioNLP 2026.

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.18653/v1/2026.bionlp-1.82

@inproceedings{nguyen2026fhexchange,
  title = {{FHexchange: Resources for Family Health History Extraction and Normalization From Consumer Dialog Sources}},
  author = {Michelle Nguyen, Nidhi Soley, Ayah Zirikly, João Sedoc and Casey Taylor},
  booktitle = {BioNLP 2026},
  year = {2026},
  doi = {10.18653/v1/2026.bionlp-1.82},
  url = {https://doi.org/10.18653/v1/2026.bionlp-1.82}
}

References

44 references, 3 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Muhammad Amith, Kirk Roberts, Cui Tao, Licong Cui, and Hua Xu. 2019. Ontology of Consumer Health Vocabulary: providing a formal and interoperable semantic resource for linking lay language and medical terminology. Proceedings. IEEE International Conference on Bioinformatics and Biomedicine, 2019:117… find
  2. Mahmoud Azab, Stephane Dadian, Vivi Nastase, Larry An, and Rada Mihalcea. 2019. Towards Extracting Medical Family History from Natural Language Interactions: A New Dataset and Baselines. In Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th Internation… find
  3. Asma Ben Abacha, Wen-wai Yim, Yadan Fan, and Thomas Lin. 2023. An Empirical Study of Clinical Note Generation from Doctor-Patient Encounters. In Proceedings of the 17th Conference of the European Chapter of the Association for Computational Linguistics, pages 2291–2302, Dubrovnik, Croatia. Associati… find
  4. Robert Bill, Serguei Pakhomov, Elizabeth S. Chen, Tamara J. Winden, Elizabeth W. Carter, and Genevieve B. Melton. 2014. Automated Extraction of Family History Information from Clinical Notes. AMIA Annual Symposium Proceedings, 2014:1709– 1717. find
  5. Pål H. Brekke, Taraka Rama, Ildikó Pilán, Øystein Nytrø, and Lilja Øvrelid. 2021. Synthetic data for annotation and extraction of family history information from clinical text. Journal of Biomedical Semantics, 12(1):11. find
  6. Victor M. Castro, Vivian S. Gainer, Danielle M. Crookes, Shawn N. Murphy, and Justin Manjourides. 2025. Comparing patient-reported symptoms and structured clinician documentation in electronic health records. Journal of the American Medical Informatics Association: JAMIA, 32(9):1454–1461. find
  7. Informatics, Imaging and Systems Biology, pages 237–242. find
  8. Elizabeth S Chen, Elizabeth W Carter, Tamara J Winden, Indra Neil Sarkar, Yan Wang, and Genevieve B Melton. 2015. Multi-source development of an integrated model for family health history. Journal of the American Medical Informatics Association : JAMIA, 22(e1):e67–e80. find
  9. Neal Lewis, Daniel Gruhl, Hui Yang, San Francisco, San Jose, and San Francisco. Extracting Family History Diagnoses From Clinical Texts. find
  10. Samuel Colvin, Eric Jolibois, Hasan Ramezani, Adrian Garcia Badaracco, Terrence Dorsey, David Montague, Serge Matveenko, Marcelo Trylesinski, Sydney Runkle, David Hewitt, Alex Hall, and Victorien Plot. 2026. Pydantic Validation. find
  11. Faiha Fareez, Tishya Parikh, Christopher Wavell, Saba Shahab, Meghan Chevalier, Scott Good, Isabella De Blasi, Rafik Rhouma, Christopher McMahon, Jean-Paul Lam, Thomas Lo, and Christopher W. Smith. 2022. A dataset of simulated patient-physician medical interviews with a focus on respiratory cases. S… find
  12. W. Gregory Feero, Mary Beth Bigley, and Kristin M. Brinner. 2008. New Standards and Enhanced Utility for Family Health History Information in the Electronic Health Record: An Update from the American Health Information Community’s Family Health History Multi-Stakeholder Workgroup. Journal of the Ame… find
  13. Yanjun Gao, Dmitriy Dligach, Timothy Miller, John Caskey, Brihat Sharma, Matthew M. Churpek, and Majid Afshar. 2023. DR.BENCH: Diagnostic Reasoning Benchmark for Clinical Natural Language Processing. Journal of biomedical informatics, 138:104286. find
  14. Yanjun Gao, Ruizhe Li, Emma Croxford, John Caskey, Brian W Patterson, Matthew Churpek, Timothy Miller, Dmitriy Dligach, and Majid Afshar. 2025. Leveraging Medical Knowledge Graphs Into Large Language Models for Diagnosis Prediction: Design and Application Study. JMIR AI, 4:e58670. find
  15. Sergey Goryachev, Hyeoneui Kim, and Qing Zeng-Treitler. 2008. Identification and Extraction of Family History Information from Clinical Reports. AMIA Annual Symposium Proceedings, 2008:247– 251. find
  16. Gale M. Lucas, Jonathan Gratch, Aisha King, and Louis-Philippe Morency. 2014. It’s only a computer: Virtual humans increase willingness to disclose. Computers in Human Behavior, 37:94–100. find
  17. Microsoft. a. gpt-4o. Microsoft. b. gpt-5-mini. Microsoft. c. Llama-3-3-70B-Instruct. find
  18. Sungrim Moon, Liwei Wang, Xuan Chen, Nan Wang, Sheila M. Manemann, Nicholas B. Larson, Suzette J. Bielinski, and Hongfang Liu. 2022. Bridging the Granularity Gap in Family History Information Extracted from Clinical Narratives. AMIA ... Annual Symposium proceedings. AMIA Symposium, 2022:795–804. find
  19. MTSamples. Medical Transcription Sample Reports & Examples \| MTSamples. find
  20. Tom Nadarzynski, Oliver Miles, Aimee Cowie, and Damien Ridge. 2019. Acceptability of artificial intelligence (AI)-led chatbot services in healthcare: A mixed-methods study. Digital Health, 5. find
  21. Mark Neumann, Daniel King, Iz Beltagy, and Waleed Ammar. 2019. ScispaCy: Fast and Robust Models for Biomedical Natural Language Processing. Proceedings of the 18th BioNLP Workshop and Shared Task, pages 319–327. Conference Name: Proceedings of the 18th BioNLP Workshop and Shared Task.
  22. Michelle Hoang Nguyen, João Sedoc, and Casey Overby Taylor. 2024. Usability, Engagement, and Report Usefulness of Chatbot-Based Family Health History Data Collection: Mixed Methods Analysis. Journal of Medical Internet Research, 26(1):e55164. Company: Journal of Medical Internet Research Distributor… find
  23. Michael Hindelang, Sebastian Sitaru, and Alexander Zink. 2024. Transforming Health Care Through Chatbots for Medical History-Taking and Future Directions: Comprehensive Systematic Review. JMIR Medical Informatics, 12(1):e56628. find
  24. Shirui Pan, Linhao Luo, Yufei Wang, Chen Chen, Jiapu Wang, and Xindong Wu. 2024. Unifying Large Language Models and Knowledge Graphs: A Roadmap. IEEE Transactions on Knowledge and Data Engineering, 36(7):3580–3599.
  25. J. R. Landis and G. G. Koch. 1977. The measurement of observer agreement for categorical data. Biometrics, 33(1):159–174. find
  26. Hyeryun Park, Jiye Son, Jeongwon Min, and Jinwook Choi. 2023. Selective UMLS knowledge infusion for biomedical question answering. Scientific Reports, 13(1):14214. find
  27. Neal Lewis, Daniel Gruhl, and Hui Yang. 2011. Dependency Parsing for Extracting Family History. In 2011 IEEE First International Conference on Healthcare Quynh Pham, Gary Graham, Carme Carrion, Plinio P. Morita, Emily Seto, Jennifer N. Stinson, and Joseph A. Cafazzo. 2019. A Library of Analytic Indi… find
  28. Taraka Rama, Pål Brekke, Øystein Nytrø, and Lilja Øvrelid. 2018. Iterative development of family history annotation guidelines using a synthetic corpus of clinical text. In Proceedings of the Ninth International Workshop on Health Text Mining and Information Analysis, pages 111–121, Brussels, Belgiu… find
  29. Majid Rastegar-Mojarad, Sijia Liu, Yanshan Wang, Naveed Afzal, Liwei Wang, Feichen Shen, Sunyang Fu, and Hongfang Liu. 2018. BioCreative/OHNLP Challenge 2018. In Proceedings of the 2018 ACM International Conference on Bioinformatics, Computational Biology, and Health Informatics, BCB ’18, page 575,… find
  30. Maciej Rybinski, Xiang Dai, Sonit Singh, Sarvnaz Karimi, and Anthony Nguyen. 2021. Extracting Family History Information From Electronic Health Records: Natural Language Processing Analysis. JMIR Medical Informatics, 9(4):e24020. find
  31. Vishal Vivek Saley, Goonjan Saha, Rocktim Jyoti Das, Dinesh Raghu, and Mausam . 2024. MediTOD: An English Dialogue Dataset for Medical History Taking with Comprehensive Annotations. In Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing, pages 16843–16877, Miami, F…
  32. SFHERE Project. 2014. Sample sentences and annotations for family history. find
  33. SFHERE Team. 2017. Digital Library for Social and Family History-Extraction-Representation, and Evaluation Project. find
  34. Feichen Shen, Sijia Liu, Sunyang Fu, Yanshan Wang, Sam Henry, Ozlem Uzuner, and Hongfang Liu. 2021. Family History Extraction From Synthetic Clinical Narratives Using Natural Language Processing: Overview and Evaluation of a Challenge Data Set and Solutions for the 2019 National NLP Clinical Challen… find
  35. Han-Chin Shing, Chaitanya Shivade, Nima Pourdamghani, Feng Nan, Philip Resnik, Douglas Oard, and Parminder Bhatia. 2021. Towards Clinical Encounter Summarization: Learning to Compose Discharge Summaries from Prior Notes. arXiv preprint. ArXiv:2104.13498 [cs]. — [arXiv:2104.13498](https://arxiv.org/abs/2104.13498) find
  36. João Figueira Silva, João Rafael Almeida, and Sérgio Matos. 2020. Extraction of Family History Information From Clinical Notes: Deep Learning and Heuristics Approach. JMIR Medical Informatics, 8(12):e22898. find
  37. Luca Soldaini and Nazli Goharian. QuickUMLS: a fast, unsupervised approach for medical concept extraction. find
  38. Zhi Rui Tam, Cheng-Kuang Wu, Yi-Lin Tsai, Chieh-Yen Lin, Hung-yi Lee, and Yun-Nung Chen. 2024. Let Me Speak Freely? A Study on the Impact of Format Restrictions on Performance of Large Language Models. arXiv preprint. ArXiv:2408.02442 [cs]. — [arXiv:2408.02442](https://arxiv.org/abs/2408.02442) find
  39. The All of Us Research Program Investigators. 2019. The “All of Us” Research Program. The New England journal of medicine, 381(7):668–676. find
  40. U.S. National Library of Medicine. Unified Medical Language System (UMLS). find
  41. Liwei Wang, Huan He, Andrew Wen, Sungrim Moon, Sunyang Fu, Kevin J. Peterson, Xuguang Ai, Sijia Liu, Ramakanth Kavuluru, and Hongfang Liu. 2023. Acquisition of a Lexicon for Family History Information: Bidirectional Encoder Representations From Transformers–Assisted Sublanguage Analysis. JMIR Medica… find
  42. Robert S. Wildin, Donna J. Messersmith, and Elisa J. F. Houwink. 2021. Modernizing family health history: achievable strategies to reduce implementation gaps. Journal of Community Genetics, 12(3):493–496. find
  43. Kecheng Zhan, Weihua Peng, Ying Xiong, Huhao Fu, Qingcai Chen, Xiaolong Wang, and Buzhou Tang. 2021. Novel Graph-Based Model With Biaffine Attention for Family History Extraction From Clinical Text: Modeling Study. JMIR Medical Informatics, 9(4):e23587. find
  44. L. Zhou, Y. Lu, C.J. Vitale, P.L. Mar, F. Chang, N. Dhopeshwarkar, and R.A. Rocha. 2014. Representation of Information about Family Relatives as Structured Data in Electronic Health Records. Applied Clinical Informatics, 5(2):349–367. find

Updated

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/nguyen-2026-fhexchange-resources-family-health-history/