CD-REST: a system for extracting chemical-induced disease relation in literature

Xu, Jun; Wu, Yonghui; Zhang, Yaoyun; Wang, Jingqi; Lee, Hee Jin; Xu, Hua

doi:10.1093/database/baw036

Cited by 73 publications

(89 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To make the data clean and standardized for further analysis, we followed Banda’s work [6] to normalize FAERS data by removing duplicate records and mapping the drug name to RxNorm [7]. For this study, we focused on the FAERS reports from 01/01/2004 to 12/31/2015.…”

Section: Methodsmentioning

confidence: 99%

Post-marketing Drug Safety Evaluation Using Data Mining Based on FAERS

Duan

Zhang

et al. 2017

Data Mining and Big Data

View full text Add to dashboard Cite

Healthcare is going through a big data revolution. The amount of data generated by healthcare is expected to increase significantly in the coming years. Therefore, efficient and effective data processing methods are required to transform data into information. In addition, applying statistical analysis can transform the information into useful knowledge. We developed a data mining method that can uncover new knowledge in this enormous field for clinical decision making while generating scientific methods and hypotheses. The proposed pipeline can be generally applied to a variety of data mining tasks in medical informatics. For this study, we applied the proposed pipeline for post-marketing surveillance on drug safety using FAERS, the data warehouse created by FDA. We used 14 kinds of neurology drugs to illustrate our methods. Our result indicated that this approach can successfully reveal insight for further drug safety evaluation.

show abstract

Section: Methodsmentioning

confidence: 99%

Post-marketing Drug Safety Evaluation Using Data Mining Based on FAERS

Duan

Zhang

et al. 2017

Data Mining and Big Data

View full text Add to dashboard Cite

show abstract

“…Chemical-induced cancer relation extraction (CID). Xu et al 30 proposed the model that classifies both sentence-level and document-level candidate drug-disease pairs by SVM, reaching a F-score of 58.53%. Table 4 shows the performance of different methods on the CID task.…”

Section: Results Analysismentioning

confidence: 99%

A novel deep learning method for extracting unspecific biomedical relation

Bai

Wang

et al. 2018

Concurrency and Computation

View full text Add to dashboard Cite

Biomedical relation extraction is an important research subject in Natural language processing (NLP). Deep learning technology has shown greater value in improving accuracy of relation extraction results recently. Existing methods mostly focus on extracting (1) specific relation from short texts (eg, drug-drug interaction and protein-protein interaction) and (2) unspecific relation from full text corpora. However, extracting unspecific relation from short text, which is more and more important in practical use, is rarely studied. In this paper, a new model called MAT-LSTM is proposed to extract unspecific relation from short text in biomedical literatures. Experiments on two Biocreative benchmark datasets and one BioNLP benchmark datasets were made to measure the validity of the proposed model MAT-LSTM, and better performance is achieved. The MAT-LSTM model is also applied practically in extracting unspecific relation contained in the PubMed literatures. The results extracted from PubMed by using the proposed model were verified by experts mostly, indicating the practical value of the MAT-LSTM model. KEYWORDS biomedical relation, deep learning, natural language processing, unspecific relation INTRODUCTIONWith the increasing popularity of precision medicine, the extraction of semantic relation between entities from biomedical literatures has attracted widespread attention in the areas of information extraction and natural language processing. 1 Discovering relations between different biomedical factors (eg, genome, metabolome, and transcriptome) is the foundation to better serve precision medicine. 2 Existing methods to extract specific types of biomedical relations (such as drug-drug interactions 3 and gene-disease interaction 4 ) are mature, whereas a method to extract unspecific types of relation is needed to discover different types relations. In this paper, a model is proposed to extract such unspecific relations.Existing biomedical relation extraction methods, including co-occurrence-based methods, rule-based methods, and machine learning methods, can be classified into two categories: (1) extracting unspecific relations in the literature from the full text and (2) extracting one specific type of relation in the literatures from short text (eg, gene-disease interaction and drug-disease interaction), instead of the full text.(1) Extracting unspecific relations in the literature from the full text.The co-occurrence-based approach is the simplest and most direct one. This method shows two features: first, in the same sentence, the closer the distance between two entities, the greater the correlation; second, the more times the two entities often appear in the same sentence, the greater Concurrency Computat Pract Exper. 2020;32:e5005. wileyonlinelibrary.com/journal/cpe

show abstract

“…Various machine learning-based methods including supervised machine learning methods (30, 31), pattern clustering (32) and topic modeling (33) were used before deep learning models became dominant among the recent advances. Besides conventional DNN models (34, 35), dependency (15, 36) and character level (16) information have been used to enhance the models with improvement over their baselines.…”

Section: Related Workmentioning

confidence: 99%

Extracting chemical–protein relations using attention-based neural networks

et al. 2018

View full text Add to dashboard Cite

Relation extraction is an important task in the field of natural language processing. In this paper, we describe our approach for the BioCreative VI Task 5: text mining chemical–protein interactions. We investigate multiple deep neural network (DNN) models, including convolutional neural networks, recurrent neural networks (RNNs) and attention-based (ATT-) RNNs (ATT-RNNs) to extract chemical–protein relations. Our experimental results indicate that ATT-RNN models outperform the same models without using attention and the ATT-gated recurrent unit (ATT-GRU) achieves the best performing micro average F1 score of 0.527 on the test set among the tested DNNs. In addition, the result of word-level attention weights also shows that attention mechanism is effective on selecting the most important trigger words when trained with semantic relation labels without the need of semantic parsing and feature engineering. The source code of this work is available at https://github.com/ohnlp/att-chemprot.

show abstract

CD-REST: a system for extracting chemical-induced disease relation in literature

Cited by 73 publications

References 27 publications

Post-marketing Drug Safety Evaluation Using Data Mining Based on FAERS

Post-marketing Drug Safety Evaluation Using Data Mining Based on FAERS

A novel deep learning method for extracting unspecific biomedical relation

Extracting chemical–protein relations using attention-based neural networks

Contact Info

Product

Resources

About