دورية أكاديمية
An End-to-End Natural Language Processing System for Automatically Extracting Radiation Therapy Events From Clinical Texts.
العنوان: | An End-to-End Natural Language Processing System for Automatically Extracting Radiation Therapy Events From Clinical Texts. |
---|---|
المؤلفون: | Bitterman DS; Computational Health Informatics Program, Boston Children's Hospital, Harvard Medical School, Boston, Massachusetts; Department of Radiation Oncology, Brigham and Women's Hospital/Dana-Farber Cancer Institute, Harvard Medical School, Boston, Massachusetts; Artificial Intelligence in Medicine (AIM) Program, Mass General Brigham, Harvard Medical School, Boston, Massachusetts. Electronic address: Danielle_Bitterman@dfci.harvard.edu., Goldner E; Computational Health Informatics Program, Boston Children's Hospital, Harvard Medical School, Boston, Massachusetts., Finan S; Computational Health Informatics Program, Boston Children's Hospital, Harvard Medical School, Boston, Massachusetts., Harris D; Computational Health Informatics Program, Boston Children's Hospital, Harvard Medical School, Boston, Massachusetts., Durbin EB; College of Medicine, University of Kentucky, Lexington, Kentucky; Kentucky Cancer Registry, Lexington, Kentucky., Hochheiser H; Department of Biomedical Informatics, University of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania., Warner JL; Population Sciences Program, Legorreta Cancer Center, Brown University, Providence, Rhode Island; Lifespan Cancer Institute, Providence, Rhode Island., Mak RH; Department of Radiation Oncology, Brigham and Women's Hospital/Dana-Farber Cancer Institute, Harvard Medical School, Boston, Massachusetts; Artificial Intelligence in Medicine (AIM) Program, Mass General Brigham, Harvard Medical School, Boston, Massachusetts., Miller T; Computational Health Informatics Program, Boston Children's Hospital, Harvard Medical School, Boston, Massachusetts., Savova GK; Computational Health Informatics Program, Boston Children's Hospital, Harvard Medical School, Boston, Massachusetts. |
المصدر: | International journal of radiation oncology, biology, physics [Int J Radiat Oncol Biol Phys] 2023 Sep 01; Vol. 117 (1), pp. 262-273. Date of Electronic Publication: 2023 Mar 27. |
نوع المنشور: | Journal Article; Research Support, N.I.H., Extramural |
اللغة: | English |
بيانات الدورية: | Publisher: Elsevier, Inc Country of Publication: United States NLM ID: 7603616 Publication Model: Print-Electronic Cited Medium: Internet ISSN: 1879-355X (Electronic) Linking ISSN: 03603016 NLM ISO Abbreviation: Int J Radiat Oncol Biol Phys Subsets: MEDLINE |
أسماء مطبوعة: | Publication: New York, NY : Elsevier, Inc Original Publication: Elmsford, N. Y., Pergamon Press. |
مواضيع طبية MeSH: | Natural Language Processing* , Neoplasms*/radiotherapy, Humans ; Electronic Health Records |
مستخلص: | Purpose: Real-world evidence for radiation therapy (RT) is limited because it is often documented only in the clinical narrative. We developed a natural language processing system for automated extraction of detailed RT events from text to support clinical phenotyping. Methods and Materials: A multi-institutional data set of 96 clinician notes, 129 North American Association of Central Cancer Registries cancer abstracts, and 270 RT prescriptions from HemOnc.org was used and divided into train, development, and test sets. Documents were annotated for RT events and associated properties: dose, fraction frequency, fraction number, date, treatment site, and boost. Named entity recognition models for properties were developed by fine-tuning BioClinicalBERT and RoBERTa transformer models. A multiclass RoBERTa-based relation extraction model was developed to link each dose mention with each property in the same event. Models were combined with symbolic rules to create a hybrid end-to-end pipeline for comprehensive RT event extraction. Results: Named entity recognition models were evaluated on the held-out test set with F1 results of 0.96, 0.88, 0.94, 0.88, 0.67, and 0.94 for dose, fraction frequency, fraction number, date, treatment site, and boost, respectively. The relation model achieved an average F1 of 0.86 when the input was gold-labeled entities. The end-to-end system F1 result was 0.81. The end-to-end system performed best on North American Association of Central Cancer Registries abstracts (average F1 0.90), which are mostly copy-paste content from clinician notes. Conclusions: We developed methods and a hybrid end-to-end system for RT event extraction, which is the first natural language processing system for this task. This system provides proof-of-concept for real-world RT data collection for research and is promising for the potential of natural language processing methods to support clinical care. (Copyright © 2023 Elsevier Inc. All rights reserved.) |
References: | J Clin Oncol. 2010 Dec 10;28(35):5160-5. (PMID: 20956628) Med Phys. 2018 Oct;45(10):e811-e819. (PMID: 30229946) JCO Clin Cancer Inform. 2022 Jul;6:e2200006. (PMID: 35917480) J Am Med Inform Assoc. 2010 Sep-Oct;17(5):563-7. (PMID: 20819865) J Oncol Pract. 2015 May;11(3):e336-50. (PMID: 25736385) J Am Med Inform Assoc. 2010 Sep-Oct;17(5):507-13. (PMID: 20819853) Cancer. 2005 Sep 15;104(6):1129-37. (PMID: 16080176) Int J Radiat Oncol Biol Phys. 2021 Jul 1;110(3):641-655. (PMID: 33545300) Nat Rev Clin Oncol. 2020 Dec;17(12):771-781. (PMID: 32843739) JAMA Netw Open. 2021 Jul 1;4(7):e2115334. (PMID: 34279650) J Oncol Pract. 2011 Jul;7(4):e15-9. (PMID: 22043196) Proc Conf. 2013 Jun;2013:14-19. (PMID: 29082384) JCO Clin Cancer Inform. 2021 Apr;5:379-393. (PMID: 33822653) J Am Med Inform Assoc. 2020 Jan 1;27(1):13-21. (PMID: 31135882) AMIA Annu Symp Proc. 2018 Dec 05;2018:1524-1533. (PMID: 30815198) Int J Radiat Oncol Biol Phys. 2018 Mar 15;100(4):1057-1066. (PMID: 29485047) Pract Radiat Oncol. 2019 Nov;9(6):395-401. (PMID: 31445187) Proc Conf. 2019 Jun;2019:22-27. (PMID: 34027520) J Biomed Inform. 2019 Dec;100:103301. (PMID: 31589927) Cancer Res. 2019 Nov 1;79(21):5463-5470. (PMID: 31395609) JAMA Oncol. 2016 Jun 1;2(6):797-804. (PMID: 27124593) Lancet Digit Health. 2020 Sep;2(9):e447-e449. (PMID: 33328110) |
معلومات مُعتمدة: | UH3 CA243120 United States CA NCI NIH HHS; R01 LM010090 United States LM NLM NIH HHS; P30 CA177558 United States CA NCI NIH HHS; UG3 CA243120 United States CA NCI NIH HHS; HHSN261201800013I United States CA NCI NIH HHS; R01 GM114355 United States GM NIGMS NIH HHS; U24 CA265879 United States CA NCI NIH HHS; U24 CA248010 United States CA NCI NIH HHS; R01 LM013486 United States LM NLM NIH HHS |
تواريخ الأحداث: | Date Created: 20230329 Date Completed: 20230929 Latest Revision: 20240426 |
رمز التحديث: | 20240426 |
مُعرف محوري في PubMed: | PMC10522797 |
DOI: | 10.1016/j.ijrobp.2023.03.055 |
PMID: | 36990288 |
قاعدة البيانات: | MEDLINE |
تدمد: | 1879-355X |
---|---|
DOI: | 10.1016/j.ijrobp.2023.03.055 |