EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

SNOMED CT entity linking challenge

Rory Davidson iD, Will Hardman, Guy Amit, Yonatan Bilu, Vincenzo Della Mea, Aleksandr Galaida, Irena Girshovitz iD, Mikhail Kulyabin iD, Mihai Horia Popescu, Kevin Roitero, Gleb Sokolov, Chen Yanover

Journal of the American Medical Informatics Association · 01.09.2025 · 32 (9) · 1397-1406

DOI: 10.1093/jamia/ocaf104 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of the American Medical Informatics Association, online erschienen am 14.07.2025. Verfasst von Rory Davidson und 11 weiteren Autor:innen. Beteiligte Einrichtung: SNOMED International, London W2 6BD, United Kingdom. Hauptthemen laut MEDLINE: Systematized Nomenclature of Medicine, Electronic Health Records. Die Arbeit wurde bisher 14-mal zitiert (Crossref) und nennt 13 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: SNOMED International, Entity Linking Challenge.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Objective

This paper presents the results from a competition challenging participants to develop entity linking models using a subset of annotated MIMIC-IV-Note data and the SNOMED CT Terminology. Materials and Methods As a basis for this work, a large set of 74 808 annotations was curated across 272 discharge notes spanning 6624 unique clinical concepts. Submissions were evaluated using the mean Intersection-over-Union metric, evaluated at the character level with the 3 best performing solutions awarded a cash prize.

Results

The winning solutions employed contrasting approaches: a dictionary-based method, an encoder-based method, and a decoder-based method.

Discussion

Our analysis reveals that concept frequency in training data significantly impacts model performance, with rare concepts proving particularly challenging. High concept entropy and annotation ambiguity were also associated with decreased performance.

Conclusion

Findings from this work suggest that future projects should focus on improving entity linking for rare concepts and developing methods to better leverage contextual information when training examples are scarce.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Systematized Nomenclature of MedicineElectronic Health RecordsHumans

Stichwörter der Autor:innen

SNOMED CTartificial intelligenceentity linkingterminology

Autor:innen und Institutionen

  • SNOMED International, London W2 6BD, United Kingdom.
  • Veratai Ltf, Woking GU22 7QW, United Kingdom.
  • KI Research Institute, Kfar Malal, Israel.
  • Department of Mathematics, Computer Science and Physics, University of Udine, Udine 33100, Italy.
  • Grid Dynamics, Yerevan 0023, Armenia.
  • VisioMed.AI, Moscow 125212, Russia.

ORCID-Profile: Rory Davidson, Irena Girshovitz, Mikhail Kulyabin

Förderung

  • SNOMED International
  • Entity Linking Challenge

Zitieren

APA

Davidson, R., Hardman, W., Amit, G., Bilu, Y., Della Mea, V., Galaida, A., Girshovitz, I., Kulyabin, M., Horia Popescu, M., Roitero, K., Sokolov, G., & Yanover, C. (2025). SNOMED CT entity linking challenge. Journal of the American Medical Informatics Association, 32(9), 1397-1406. https://doi.org/10.1093/jamia/ocaf104

Vancouver

Davidson R, Hardman W, Amit G, Bilu Y, Della Mea V, Galaida A, et al. SNOMED CT entity linking challenge. Journal of the American Medical Informatics Association. 2025;32(9):1397-1406. doi:10.1093/jamia/ocaf104

BibTeX
@article{Davidson2025,
  author = {Davidson, Rory and Hardman, Will and Amit, Guy and Bilu, Yonatan and Della Mea, Vincenzo and Galaida, Aleksandr and Girshovitz, Irena and Kulyabin, Mikhail and Horia Popescu, Mihai and Roitero, Kevin and Sokolov, Gleb and Yanover, Chen},
  title = {SNOMED CT entity linking challenge},
  journal = {Journal of the American Medical Informatics Association},
  year = {2025},
  volume = {32},
  number = {9},
  pages = {1397--1406},
  doi = {10.1093/jamia/ocaf104}
}
Kostenlos starten →