Fei Wang
Research Subjects
Research Interests
Fei Wang has a total of 19 co-authors. Here are the top co-authors by number of shared publications:
- Handong Ma
Synyi Research, Shanghai, China
4 shared publications - Ken Chen
Synyi Research, Shanghai, China
4 shared publications - Shaodian Zhang
Synyi Research, Shanghai, China
4 shared publications - Dong Wen
Center for Medical Informatics, Peking University, Beijing, China
2 shared publications - Gen Gu
Synyi Research, Shanghai, China
2 shared publications - Jianbo Lei
Center for Medical Informatics, Advanced Institute of Clinical Medicine, Peking University, Beijing, China
2 shared publications
Latest Publications
- Open Access
Table-based Fact Verification with Salience-aware Learning
Tables provide valuable knowledge that can be used to verify textual statements. While a number of works have considered table-based fact verification, direct alignments of tabular data with tokens in textual statements are rarely available. Moreover, training a generalized fact verification model requires abundant labeled training data. In this paper, we propose a novel system to address these problems. Inspired by counterfactual causality, our system identifies token-level salience in the statement with probing-based salience estimation. Salience estimation allows enhanced learning of fact verification from two perspectives. From one perspective, our system conducts masked salient token prediction to enhance the model for alignment and reasoning between the table and the statement. From the other perspective, our system applies salience-aware data augmentation to generate a more diverse set of training instances by replacing non-salient terms. Experimental results on TabFact show the effective improvement by the proposed salience-aware learning techniques, leading to the new SOTA performance on the benchmark. Our code is publicly available at https://github.com/luka-group/Salience-aware-Learning .
Open AccessTable-based Fact Verification with Salience-aware Learning
Tables provide valuable knowledge that can be used to verify textual statements. While a number of works have considered table-based fact verification, direct alignments of tabular data with tokens in textual statements are rarely available. Moreover, training a generalized fact verification model requires abundant labeled training data. In this paper, we propose a novel system to address these problems. Inspired by counterfactual causality, our system identifies token-level salience in the statement with probing-based salience estimation. Salience estimation allows enhanced learning of fact verification from two perspectives. From one perspective, our system conducts masked salient token prediction to enhance the model for alignment and reasoning between the table and the statement. From the other perspective, our system applies salience-aware data augmentation to generate a more diverse set of training instances by replacing non-salient terms. Experimental results on TabFact show the effective improvement by the proposed salience-aware learning techniques, leading to the new SOTA performance on the benchmark. Our code is publicly available at https://github.com/luka-group/Salience-aware-Learning .
- Open Access
Biomedical Evidence Generation Engine
With the rapid development of precision medicine, a large amount of health data (such as electronic health records, gene sequencing, medical images, etc.) has been produced. It encourages more and more interest in data-driven insight discovery from these data. It is a reasonable way to verify the derived insights in biomedical literature. However, manual verification is inefficient and not scalable. Therefore, an intelligent technique is necessary to solve this problem. In this paper, we propose a task of biomedical evidence generation, which is very novel and different from existing NLP tasks. Furthermore, we developed a biomedical evidence generation engine for this task with the pipeline of three components which are a literature retrieval module, a skeleton information identification module, and a text summarization module.
Open AccessBiomedical Evidence Generation Engine
With the rapid development of precision medicine, a large amount of health data (such as electronic health records, gene sequencing, medical images, etc.) has been produced. It encourages more and more interest in data-driven insight discovery from these data. It is a reasonable way to verify the derived insights in biomedical literature. However, manual verification is inefficient and not scalable. Therefore, an intelligent technique is necessary to solve this problem. In this paper, we propose a task of biomedical evidence generation, which is very novel and different from existing NLP tasks. Furthermore, we developed a biomedical evidence generation engine for this task with the pipeline of three components which are a literature retrieval module, a skeleton information identification module, and a text summarization module.
- Open Access
- 16 July 2019
- Published by JMIR Publications Inc. in JMIR Medical Informatics
- Vol. 7 (3) , e13331
Improving the Efficacy of the Data Entry Process for Clinical Research With a Natural Language Processing–Driven Medical Information Extraction System: Quantitative Field Research
Journal of Medical Internet Research - International Scientific Journal for Medical Research, Information and Communication on the Internet #Preprint #PeerReviewMe: Warning: This is a unreviewed preprint. Readers are warned that the document has not been peer-reviewed by expert/patient reviewers or an academic editor, may contain misleading claims, and is likely to undergo changes before final publication, if accepted, or may have been rejected/withdrawn. Readers with interest and expertise are encouraged to sign up as peer-reviewer, if the paper is within an open peer-review period. Please cite this preprint only for review purposes or for grant applications and CVs (if you are the author). Background: The growing interest in observational trials using patient historical data from electric medical records (EMRs) poses challenges to both efficiency and quality of clinical data collection and management. However, even using electronic data capture (EDC) system and electronic case report form (eCRF), a manual data entry process followed by chart review is still inevitable and becomes an efficiency bottleneck. Objective: To facilitate the data entry process, we developed a natural language processing (NLP)-driven medical information extraction system (NLP-MIES) based on the i2b2 reference standard. We aimed to evaluate whether our NLP-MIES based eCRF application could improve both the accuracy and speed of the data entry process. Methods: We conducted a randomized and controlled field experiment and recruited a total of 24 eligible participants, 12 for manual group and 12 for NLP-MIES support group. We simulated the real-world eCRF filling process using our system and compared the performance of data entry on two research topics, pediatric congenital heart disease and pneumonia. Results: For congenital heart disease eCRF, NLP-MIES support was associated with an increasing of 15% (95% CI 4% to 120%, P=0.0269) in geometric mean accuracy and a reduction of 33% (95% CI 22% to 42%, P <0.001) in geometric mean of the time consumed. For pneumonia eCRF, NLP-MIES support was associated with an increasing of 18% (95% CI 6% to 32%, P =0.0075) in geometric mean accuracy and a reduction of 31% (95% CI 19% to 41%, P =0.001) in geometric mean of the time consumed. Conclusions: Our system could improve both the accuracy and speed of the data entry process.
- 16 July 2019
- Published by JMIR Publications Inc. in JMIR Medical Informatics
- Vol. 7 (3) , e13331
Open AccessImproving the Efficacy of the Data Entry Process for Clinical Research With a Natural Language Processing–Driven Medical Information Extraction System: Quantitative Field Research
Journal of Medical Internet Research - International Scientific Journal for Medical Research, Information and Communication on the Internet #Preprint #PeerReviewMe: Warning: This is a unreviewed preprint. Readers are warned that the document has not been peer-reviewed by expert/patient reviewers or an academic editor, may contain misleading claims, and is likely to undergo changes before final publication, if accepted, or may have been rejected/withdrawn. Readers with interest and expertise are encouraged to sign up as peer-reviewer, if the paper is within an open peer-review period. Please cite this preprint only for review purposes or for grant applications and CVs (if you are the author). Background: The growing interest in observational trials using patient historical data from electric medical records (EMRs) poses challenges to both efficiency and quality of clinical data collection and management. However, even using electronic data capture (EDC) system and electronic case report form (eCRF), a manual data entry process followed by chart review is still inevitable and becomes an efficiency bottleneck. Objective: To facilitate the data entry process, we developed a natural language processing (NLP)-driven medical information extraction system (NLP-MIES) based on the i2b2 reference standard. We aimed to evaluate whether our NLP-MIES based eCRF application could improve both the accuracy and speed of the data entry process. Methods: We conducted a randomized and controlled field experiment and recruited a total of 24 eligible participants, 12 for manual group and 12 for NLP-MIES support group. We simulated the real-world eCRF filling process using our system and compared the performance of data entry on two research topics, pediatric congenital heart disease and pneumonia. Results: For congenital heart disease eCRF, NLP-MIES support was associated with an increasing of 15% (95% CI 4% to 120%, P=0.0269) in geometric mean accuracy and a reduction of 33% (95% CI 22% to 42%, P <0.001) in geometric mean of the time consumed. For pneumonia eCRF, NLP-MIES support was associated with an increasing of 18% (95% CI 6% to 32%, P =0.0075) in geometric mean accuracy and a reduction of 31% (95% CI 19% to 41%, P =0.001) in geometric mean of the time consumed. Conclusions: Our system could improve both the accuracy and speed of the data entry process.
- Open Access
- 23 May 2019
- Published by JMIR Publications Inc. in JMIR Medical Informatics
- Vol. 7 (2) , e12704-224
Development of a Consumer Health Vocabulary by Mining Health Forum Texts Based on Word Embedding: Semiautomatic Approach
Journal of Medical Internet Research - International Scientific Journal for Medical Research, Information and Communication on the Internet #Preprint #PeerReviewMe: Warning: This is a unreviewed preprint. Readers are warned that the document has not been peer-reviewed by expert/patient reviewers or an academic editor, may contain misleading claims, and is likely to undergo changes before final publication, if accepted, or may have been rejected/withdrawn. Readers with interest and expertise are encouraged to sign up as peer-reviewer, if the paper is within an open peer-review period. Please cite this preprint only for review purposes or for grant applications and CVs (if you are the author). Background: Vocabulary gap between consumers and professionals in the medical domain hinders information seeking and communication. Consumer health vocabularies (CHVs) have been developed to aid such informatics applications. This purpose is best served if the vocabulary evolves with consumers’ language. Objective: Our objective is to develop a method for identifying and adding new terms to CHV, so it can keep up with the constantly evolving medical knowledge and language use. Methods: A consumer health terms finding framework based on distributed word vector space model is proposed. We first learn word vectors from large-scale text corpus and then adopt a supervised method with existing CHVs for learning vector representation of words, which can provide additional supervised fine tuning after unsupervised word embedding learning. With a fine-tuned word vector space, we identify pairs of professional terms and their consumer variants by their semantic distance in the vector space. A subsequent manual review of the extracted and labeled pairs of entities was conducted to validate the results generated by the proposed approach. Results: The results are evaluated using mean reciprocal rank (MRR). After manual evaluation, results show that it is feasible to identify alternative medical concepts by using professional or consumer concepts as queries in the word vector space without fine tuning, and the results are more promising in final fine-tuned word vector space. The MRR values indicate that on average, a professional or consumer concept is about 14th closest to its counterpart in word vector space without fine tuning, and the MMR in final fine-tuned word vector space is 8. Furthermore, the results demonstrate that our method is able to collect abbreviations and common typos frequently used by consumers. Conclusions: By integrating a large amount of text information and existing CHVs, our method outperformed several baseline ranking methods and is effective for generating a list of candidate terms for human review during CHV development.
- 23 May 2019
- Published by JMIR Publications Inc. in JMIR Medical Informatics
- Vol. 7 (2) , e12704-224
Open AccessDevelopment of a Consumer Health Vocabulary by Mining Health Forum Texts Based on Word Embedding: Semiautomatic Approach
Journal of Medical Internet Research - International Scientific Journal for Medical Research, Information and Communication on the Internet #Preprint #PeerReviewMe: Warning: This is a unreviewed preprint. Readers are warned that the document has not been peer-reviewed by expert/patient reviewers or an academic editor, may contain misleading claims, and is likely to undergo changes before final publication, if accepted, or may have been rejected/withdrawn. Readers with interest and expertise are encouraged to sign up as peer-reviewer, if the paper is within an open peer-review period. Please cite this preprint only for review purposes or for grant applications and CVs (if you are the author). Background: Vocabulary gap between consumers and professionals in the medical domain hinders information seeking and communication. Consumer health vocabularies (CHVs) have been developed to aid such informatics applications. This purpose is best served if the vocabulary evolves with consumers’ language. Objective: Our objective is to develop a method for identifying and adding new terms to CHV, so it can keep up with the constantly evolving medical knowledge and language use. Methods: A consumer health terms finding framework based on distributed word vector space model is proposed. We first learn word vectors from large-scale text corpus and then adopt a supervised method with existing CHVs for learning vector representation of words, which can provide additional supervised fine tuning after unsupervised word embedding learning. With a fine-tuned word vector space, we identify pairs of professional terms and their consumer variants by their semantic distance in the vector space. A subsequent manual review of the extracted and labeled pairs of entities was conducted to validate the results generated by the proposed approach. Results: The results are evaluated using mean reciprocal rank (MRR). After manual evaluation, results show that it is feasible to identify alternative medical concepts by using professional or consumer concepts as queries in the word vector space without fine tuning, and the results are more promising in final fine-tuned word vector space. The MRR values indicate that on average, a professional or consumer concept is about 14th closest to its counterpart in word vector space without fine tuning, and the MMR in final fine-tuned word vector space is 8. Furthermore, the results demonstrate that our method is able to collect abbreviations and common typos frequently used by consumers. Conclusions: By integrating a large amount of text information and existing CHVs, our method outperformed several baseline ranking methods and is effective for generating a list of candidate terms for human review during CHV development.
- Open Access
- 18 January 2019
- Published by JMIR Publications Inc. in JMIR Preprints
Improving the Efficacy of the Data Entry Process for Clinical Research With a Natural Language Processing–Driven Medical Information Extraction System: Quantitative Field Research (Preprint)
BACKGROUND The growing interest in observational trials using patient data from electronic medical records poses challenges to both efficiency and quality of clinical data collection and management. Even with the help of electronic data capture systems and electronic case report forms (eCRFs), the manual data entry process followed by chart review is still time consuming. OBJECTIVE To facilitate the data entry process, we developed a natural language processing–driven medical information extraction system (NLP-MIES) based on the i2b2 reference standard. We aimed to evaluate whether the NLP-MIES–based eCRF application could improve the accuracy and efficiency of the data entry process. METHODS We conducted a randomized and controlled field experiment, and 24 eligible participants were recruited (12 for the manual group and 12 for NLP-MIES–supported group). We simulated the real-world eCRF completion process using our system and compared the performance of data entry on two research topics, pediatric congenital heart disease and pneumonia. RESULTS For the congenital heart disease condition, the NLP-MIES–supported group increased accuracy by 15% (95% CI 4%-120%, P=.03) and reduced elapsed time by 33% (95% CI 22%-42%, P<.001) compared with the manual group. For the pneumonia condition, the NLP-MIES–supported group increased accuracy by 18% (95% CI 6%-32%, P=.008) and reduced elapsed time by 31% (95% CI 19%-41%, P<.001). CONCLUSIONS Our system could improve both the accuracy and efficiency of the data entry process.
- 18 January 2019
- Published by JMIR Publications Inc. in JMIR Preprints
Open AccessImproving the Efficacy of the Data Entry Process for Clinical Research With a Natural Language Processing–Driven Medical Information Extraction System: Quantitative Field Research (Preprint)
BACKGROUND The growing interest in observational trials using patient data from electronic medical records poses challenges to both efficiency and quality of clinical data collection and management. Even with the help of electronic data capture systems and electronic case report forms (eCRFs), the manual data entry process followed by chart review is still time consuming. OBJECTIVE To facilitate the data entry process, we developed a natural language processing–driven medical information extraction system (NLP-MIES) based on the i2b2 reference standard. We aimed to evaluate whether the NLP-MIES–based eCRF application could improve the accuracy and efficiency of the data entry process. METHODS We conducted a randomized and controlled field experiment, and 24 eligible participants were recruited (12 for the manual group and 12 for NLP-MIES–supported group). We simulated the real-world eCRF completion process using our system and compared the performance of data entry on two research topics, pediatric congenital heart disease and pneumonia. RESULTS For the congenital heart disease condition, the NLP-MIES–supported group increased accuracy by 15% (95% CI 4%-120%, P=.03) and reduced elapsed time by 33% (95% CI 22%-42%, P<.001) compared with the manual group. For the pneumonia condition, the NLP-MIES–supported group increased accuracy by 18% (95% CI 6%-32%, P=.008) and reduced elapsed time by 31% (95% CI 19%-41%, P<.001). CONCLUSIONS Our system could improve both the accuracy and efficiency of the data entry process.
- Open Access
- 6 November 2018
- Published by JMIR Publications Inc. in JMIR Preprints
Development of a Consumer Health Vocabulary by Mining Health Forum Texts Based on Word Embedding: Semiautomatic Approach (Preprint)
BACKGROUND The vocabulary gap between consumers and professionals in the medical domain hinders information seeking and communication. Consumer health vocabularies have been developed to aid such informatics applications. This purpose is best served if the vocabulary evolves with consumers’ language. OBJECTIVE Our objective is to develop a method for identifying and adding new terms to consumer health vocabularies, so that it can keep up with the constantly evolving medical knowledge and language use. METHODS In this paper, we propose a consumer health term–finding framework based on a distributed word vector space model. We first learned word vectors from a large-scale text corpus and then adopted a supervised method with existing consumer health vocabularies for learning vector representation of words, which can provide additional supervised fine tuning after unsupervised word embedding learning. With a fine-tuned word vector space, we identified pairs of professional terms and their consumer variants by their semantic distance in the vector space. A subsequent manual review of the extracted and labeled pairs of entities was conducted to validate the results generated by the proposed approach. The results were evaluated using mean reciprocal rank (MRR). RESULTS Manual evaluation showed that it is feasible to identify alternative medical concepts by using professional or consumer concepts as queries in the word vector space without fine tuning, but the results are more promising in the final fine-tuned word vector space. The MRR values indicated that on an average, a professional or consumer concept is about 14th closest to its counterpart in the word vector space without fine tuning, and the MRR in the final fine-tuned word vector space is 8. Furthermore, the results demonstrate that our method can collect abbreviations and common typos frequently used by consumers. CONCLUSIONS By integrating a large amount of text information and existing consumer health vocabularies, our method outperformed several baseline ranking methods and is effective for generating a list of candidate terms for human review during consumer health vocabulary development.
- 6 November 2018
- Published by JMIR Publications Inc. in JMIR Preprints
Open AccessDevelopment of a Consumer Health Vocabulary by Mining Health Forum Texts Based on Word Embedding: Semiautomatic Approach (Preprint)
BACKGROUND The vocabulary gap between consumers and professionals in the medical domain hinders information seeking and communication. Consumer health vocabularies have been developed to aid such informatics applications. This purpose is best served if the vocabulary evolves with consumers’ language. OBJECTIVE Our objective is to develop a method for identifying and adding new terms to consumer health vocabularies, so that it can keep up with the constantly evolving medical knowledge and language use. METHODS In this paper, we propose a consumer health term–finding framework based on a distributed word vector space model. We first learned word vectors from a large-scale text corpus and then adopted a supervised method with existing consumer health vocabularies for learning vector representation of words, which can provide additional supervised fine tuning after unsupervised word embedding learning. With a fine-tuned word vector space, we identified pairs of professional terms and their consumer variants by their semantic distance in the vector space. A subsequent manual review of the extracted and labeled pairs of entities was conducted to validate the results generated by the proposed approach. The results were evaluated using mean reciprocal rank (MRR). RESULTS Manual evaluation showed that it is feasible to identify alternative medical concepts by using professional or consumer concepts as queries in the word vector space without fine tuning, but the results are more promising in the final fine-tuned word vector space. The MRR values indicated that on an average, a professional or consumer concept is about 14th closest to its counterpart in the word vector space without fine tuning, and the MRR in the final fine-tuned word vector space is 8. Furthermore, the results demonstrate that our method can collect abbreviations and common typos frequently used by consumers. CONCLUSIONS By integrating a large amount of text information and existing consumer health vocabularies, our method outperformed several baseline ranking methods and is effective for generating a list of candidate terms for human review during consumer health vocabulary development.