Kliniske publikationer, der anvender vores sprogmodeller
Identification of hematuria with a natural language processing model and validation of hematuria diagnosecodes. Hansen RS, Lynggaard RB, Laursen MS, Lykke FM, Vinholt PJ. Thromb Res. 2024 Oct 12;244:109182. doi: 10.1016/j.thromres.2024.109182. PMID: 39426095
Validity of Major Osteoporotic Fracture Diagnoses in the Danish National Patient Registry. Clausen A , Möller S , Skjødt MK , Lynggaard RB , Vinholt PJ , Lindberg-Larsen M , Søndergaard J , Abrahamsen B , Rubin KH. Clin Epidemiol . 2024 Apr 13:16:257-266. doi: 10.2147/CLEP.S444447.
Cumulative rib fracture risk after stereotactic body radiotherapy in patients with localized non-small cell lung cancer. Kristian Kirkelund Bentsen, Carsten Brink, Tine Bjørn Nielsen, Rasmus Bank Lynggaard, Pernille Just Vinholt, Tine Schytte, Olfred Hansen, Stefan Starup Jeppesen. Radiother Oncol 2024 17:110481. doi: 10.1016/j.radonc.2024.110481.
Validation of Anorexia Nervosa and Bulimia Nervosa Diagnosis Coding in Danish Hospitals Assisted by a Natural Language Processing model. Egedal, J. M., Støving, R. K., Lynggaard, R. B., Laursen, M. S., Vinholt, P. J., & Hansen, R. S. (2024). Journal of Psychiatric Research.
Tekniske publikationer
Mining Electronic Health Records: Turning Unstructured Text into Research Data Using Natural Language Processing – Ph.d. afhandling af Martin Sundahl Laursen
Unlocking the Potential of Electronic Health Records With Danish Clinical Language Models for Text Mining – Ph.d.afhandling af Jannik Skyttegaard Pedersen
Doctors identify hemorrhage better during chart review when assisted by artificial intelligence. Martin S. Laursen , Jannik S. Pedersen, Rasmus S. Hansen, Thiusius R. Savarimuthu and Pernille J. Vinholt. Clinical Applied Informatics (accepted)
Automatic Annotation of Training Data for Deep Learning Based De-identification of Narrative Clinical Text. Martin S. Laursen , Jannik S. Pedersen, Pernille J. Vinholt and Thiusius Rajeeth Savarimuthu. WNLPe-Health 2022
Danish Clinical Named Entity Recognition and Relation Extraction. Martin S. Laursen, Jannik S. Pedersen, Rasmus S H.ansen, Thiusius R. Savarimuthu, Pernille J. Vinholt. Proceedings of the 24th Nordic Conference on Computational Linguistics (NoDaLiDa), 2023.
MeDa-BERT: A medical Danish pretrained transformer model. Jannik S. Pedersen, Martin S. Laursen, Pernille J. Vinholt, Thiusius R. Savarimuthu. Proceedings of the 24th Nordic Conference on Computational Linguistics (NoDaLiDa), 2023.
Investigating anatomical bias in clinical machine learning algorithms. Jannik S. Pedersen, Martin S. Laursen, Pernille J. Vinholt, Anne Alnor, Thiusius R. Savarimuthu. Findings of the Association for Computational Linguistics: EACL 2023, 1368-1380.
Benchmark for Evaluation of Danish Clinical Word Embeddings. Martin S Laursen, Jannik S. Pedersen, Pernille J. Vinholt, Rasmus S. Hansen, Thiusius R. Savarimuthu. Northern European Journal of Language Technology 2023/3/1.
Domain over size: Clinical ELECTRA surpasses general BERT for bleeding site classification in the free text of electronic health records. Jannik S. Pedersen, Martin S. Laursen, Cristina S. Ruiz, Thiusius R. Savarimuthu, Rasmus S. Hansen, Pernille J. Vinholt. IEEE-EMBS International Conference on Biomedical and Health Informatics, 2022.
Factors Facilitating the Acceptance of Diagnostic Robots in Healthcare: A Survey. Martin S. Laursen, Jannik S. Pedersen, Søren A. Just, Thiusius R. Savarimuthu, Brian Blomholt, Jakob KH Andersen, and Pernille J. Vinholt. In 2022 IEEE 10th International Conference on Healthcare Informatics (ICHI), pp. 442-448. IEEE, 2022.
Deep learning detects and visualizes bleeding events in electronic health records. Jannik S. Pedersen, Martin S. Laursen, Thiusius R. Savarimuthu, Rasmus S. Hansen, Anne Alnor, Bjerre KV, Kjær IM, Gils C, Thorsen AF, Andersen ES, Nielsen CB, Andersen LC, Just SA, Vinholt PJ. Res Pract Thromb Haemost. 2021 May 5;5(4):e12505. doi: 10.1002/rth2.12505.