Temporal information extraction from clinical text

Tourille, Julien; Ferret, Olivier; Tannier, Xavier; Névéol, Aurélie

doi:10.18653/v1/e17-2117

Cited by 10 publications

(12 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Of the 105 reviewed publications, 70 dealt with shared-task datasets. Except for [31] and [32], which additionally used another dataset, the other 68 publications only used shared task-related datasets. Hence, most publications were related to shared task datasets.…”

Section: Global Quantitative Resultsmentioning

confidence: 99%

“…Publications in languages other than English include [39][40][41][42] in Chinese, [43,44] in Korean, [45] in Dutch, [46] in Italian, [47] in Swedish, and [48] in Spanish. Additionally, [31] dealt with English and French, extracting temporal relations from both the THYME corpus and the MERLOT corpus [49], which are from medical texts in French. One can conclude that there is room for research in languages other than English.…”

Section: Global Quantitative Resultsmentioning

confidence: 99%

“…Best strategy SE Results I2b2 2012 dataset [72] SVM clf Fm 0.6954 [12] 2 S V M c l f s Fm 0.695 [11] 2 SVM clfs Fm 0.6932 [73] CRF clf Fm 0.693 [127] 2 SVM clfs Fm 0.6849 I2b2/UTHealth 2014 dataset [59] label-powerset strategy + SVM clfs Fm 0.9268 [68] 21 RIPPER clfs + voting Fm 0.9185 [36] BI-LSTM Fm 0.9081 [62] OneRule clfs Fm 0.857 Clinical TempEval 2015 dataset [12] SVM clf √ Fm 0.807 [128] CRF clf √ Fm 0.791 Clinical TempEval 2016 dataset [103] B E R T + MTL Fm 0.86 (NTC) [31] SVM clf √ Fm 0.87 [97] Structured perceptron + ILP √ Fm 0.846 [28] CRF clf √ Fm 0.844 [94] SVM clf √ Fm 0.835 [86] L R c l f s √ Fm 0.815 [106] R F c l f √ Fm 0.807 [98] C N N + MLP √ Fm 0.788 [85] CRF clf √ Fm 0.714 [84] CRF clf √ Fm 0.712 [91] CRF clf √ Fm 0.687 Clinical TempEval 2017 dataset [113] SVM clf √ Fm 0.519 UDA, 0.591 SDA [109] Structured perceptron + ILP √ Fm 0.49 UDA, 0.56 SDA [112] CRF clf √ Fm 0.45 UDA, 0.52 SDA [111] CRF used a label-powerset strategy to transform the multi-label classification into several binary classification tasks addressed with SVM classifiers. For the i2b2 2012 dataset, the authors of [72] used a single SVM classifier, while the authors of [11], [12], and [127] involved two SVM classifiers, one for each DCT.…”

Section: Authorsmentioning

confidence: 99%

See 2 more Smart Citations

Temporal Relation Extraction in Clinical Texts

et al. 2021

View full text Add to dashboard Cite

Unstructured data in electronic health records, represented by clinical texts, are a vast source of healthcare information because they describe a patient's journey, including clinical findings, procedures, and information about the continuity of care. The publication of several studies on temporal relation extraction from clinical texts during the last decade and the realization of multiple shared tasks highlight the importance of this research theme. Therefore, we propose a review of temporal relation extraction in clinical texts. We analyzed 105 articles and verified that relations between events and document creation time, a coarse temporality type, were addressed with traditional machine learning–based models with few recent initiatives to push the state-of-the-art with deep learning–based models. For temporal relations between entities (event and temporal expressions) in the document, factors such as dataset imbalance because of candidate pair generation and task complexity directly affect the system's performance. The state-of-the-art resides on attention-based models, with contextualized word representations being fine-tuned for temporal relation extraction. However, further experiments and advances in the research topic are required until real-time clinical domain applications are released. Furthermore, most of the publications mainly reside on the same dataset, hindering the need for new annotation projects that provide datasets for different medical specialties, clinical text types, and even languages.

show abstract

Section: Global Quantitative Resultsmentioning

confidence: 99%

Section: Global Quantitative Resultsmentioning

confidence: 99%

Section: Authorsmentioning

confidence: 99%

See 1 more Smart Citation

Temporal Relation Extraction in Clinical Texts

et al. 2021

View full text Add to dashboard Cite

show abstract

“…The first one will investigate whether training two models, one for EVENT-TIMEX3 relations and one for EVENT-EVENT relations, as done by Dligach et al (2017), is a better option than training one model for all types of containment relations as presented herein. The second extension consists in transposing the model we have defined in this work for English to French, as done by Tourille et al (2017) for a more traditional approach based on a feature engineering approach.…”

Section: Discussionmentioning

confidence: 99%

Neural Architecture for Temporal Relation Extraction: A Bi-LSTM Approach for Detecting Narrative Containers

Tourille¹,

Ferret²,

Névéol³

et al. 2017

Proceedings of the 55th Annual Meeting of the Association For Computational Linguistics (Volume 2: Short Papers)

Self Cite

View full text Add to dashboard Cite

show abstract

“…Ling and Weld [3] propose an extractor for temporal information with probabilistic inference. Tourille et al [4] attempt to extract numeral information from clinical documents. Davidov and Rappoport [5] extract numerical information like size and depth from the web and experiment on the question answering task.…”

Section: Related Workmentioning

confidence: 99%

Numeral Understanding in Financial Tweets for Fine-Grained Crowd-Based Forecasting

Chen

Huang

Shiue

et al. 2018

2018 IEEE/WIC/ACM International Conference on Web Intelligence (WI)

View full text Add to dashboard Cite

Numerals that contain much information in financial documents are crucial for financial decision making. They play different roles in financial analysis processes. This paper is aimed at understanding the meanings of numerals in financial tweets for fine-grained crowd-based forecasting. We propose a taxonomy that classifies the numerals in financial tweets into 7 categories, and further extend some of these categories into several subcategories. Neural network-based models with word and character-level encoders are proposed for 7-way classification and 17-way classification. We perform backtest to confirm the effectiveness of the numeric opinions made by the crowd. This work is the first attempt to understand numerals in financial social media data, and we provide the first comparison of fine-grained opinion of individual investors and analysts based on their forecast price. The numeral corpus used in our experiments, called FinNum 1.0 1 , is available for research purposes.

show abstract

Temporal information extraction from clinical text

Abstract: In this paper, we present a method for temporal relation extraction from clinical narratives in French and in English. We experiment on two comparable corpora, the MERLOT corpus for French and the THYME corpus for English, and show that a common approach can be used for both languages.

Cited by 10 publications

References 16 publications

Temporal Relation Extraction in Clinical Texts

Temporal Relation Extraction in Clinical Texts

Neural Architecture for Temporal Relation Extraction: A Bi-LSTM Approach for Detecting Narrative Containers

Numeral Understanding in Financial Tweets for Fine-Grained Crowd-Based Forecasting

Contact Info

Product

Resources

About