Ziyi Wei
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Research Interests
Ziyi Wei has a total of 14 co-authors. Here are the top co-authors by number of shared publications:
- Honglian Huang
School of Life Sciences and Technology, Tongji University Department of Bioinformatics Shanghai 200092
2 shared publications - Hui Zong
Joint Laboratory of Artificial Intelligence for Critical Care Medicine, Department of Critical Care Medicine and Institutes for Systems Genetics, Frontiers Science Center for Disease-related Molecular Network, West China Hospital, Sichuan University, Chengdu, China
2 shared publications - James James Cardwell Crabbe
Wolfson College, Oxford University, Oxford, UK
2 shared publications - Meng Fang
Department of Laboratory Medicine, Shanghai Eastern Hepatobiliary Surgery Hospital, Shanghai, China
2 shared publications - Rebecca Wu
University of California, Berkeley, Berkeley, US
2 shared publications - Shiyang Cheng
Research Center for Translational Medicine, Shanghai East Hospital, School of Life Sciences and Technology, Tongji University, 1239 Siping Road, Shanghai, CN
2 shared publications
Latest Publications
- Open Access
- 20 September 2023
- Published by JMIR Publications Inc. in Journal of Medical Internet Research
- Vol. 25, e48115
Large-Scale Biomedical Relation Extraction Across Diverse Relation Types: Model Development and Usability Study on COVID-19
Background: Biomedical relation extraction (RE) is of great importance for researchers to conduct systematic biomedical studies. It not only helps knowledge mining, such as knowledge graphs and novel knowledge discovery, but also promotes translational applications, such as clinical diagnosis, decision-making, and precision medicine. However, the relations between biomedical entities are complex and diverse, and comprehensive biomedical RE is not yet well established. Objective: We aimed to investigate and improve large-scale RE with diverse relation types and conduct usability studies with application scenarios to optimize biomedical text mining. Methods: Data sets containing 125 relation types with different entity semantic levels were constructed to evaluate the impact of entity semantic information on RE, and performance analysis was conducted on different model architectures and domain models. This study also proposed a continued pretraining strategy and integrated models with scripts into a tool. Furthermore, this study applied RE to the COVID-19 corpus with article topics and application scenarios of clinical interest to assess and demonstrate its biological interpretability and usability. Results: The performance analysis revealed that RE achieves the best performance when the detailed semantic type is provided. For a single model, PubMedBERT with continued pretraining performed the best, with an F1-score of 0.8998. Usability studies on COVID-19 demonstrated the interpretability and usability of RE, and a relation graph database was constructed, which was used to reveal existing and novel drug paths with edge explanations. The models (including pretrained and fine-tuned models), integrated tool (Docker), and generated data (including the COVID-19 relation graph database and drug paths) have been made publicly available to the biomedical text mining community and clinical researchers. Conclusions: This study provided a comprehensive analysis of RE with diverse relation types. Optimized RE models and tools for diverse relation types were developed, which can be widely used in biomedical text mining. Our usability studies provided a proof-of-concept demonstration of how large-scale RE can be leveraged to facilitate novel research.
- 20 September 2023
- Published by JMIR Publications Inc. in Journal of Medical Internet Research
- Vol. 25, e48115
Open AccessLarge-Scale Biomedical Relation Extraction Across Diverse Relation Types: Model Development and Usability Study on COVID-19
Background: Biomedical relation extraction (RE) is of great importance for researchers to conduct systematic biomedical studies. It not only helps knowledge mining, such as knowledge graphs and novel knowledge discovery, but also promotes translational applications, such as clinical diagnosis, decision-making, and precision medicine. However, the relations between biomedical entities are complex and diverse, and comprehensive biomedical RE is not yet well established. Objective: We aimed to investigate and improve large-scale RE with diverse relation types and conduct usability studies with application scenarios to optimize biomedical text mining. Methods: Data sets containing 125 relation types with different entity semantic levels were constructed to evaluate the impact of entity semantic information on RE, and performance analysis was conducted on different model architectures and domain models. This study also proposed a continued pretraining strategy and integrated models with scripts into a tool. Furthermore, this study applied RE to the COVID-19 corpus with article topics and application scenarios of clinical interest to assess and demonstrate its biological interpretability and usability. Results: The performance analysis revealed that RE achieves the best performance when the detailed semantic type is provided. For a single model, PubMedBERT with continued pretraining performed the best, with an F1-score of 0.8998. Usability studies on COVID-19 demonstrated the interpretability and usability of RE, and a relation graph database was constructed, which was used to reveal existing and novel drug paths with edge explanations. The models (including pretrained and fine-tuned models), integrated tool (Docker), and generated data (including the COVID-19 relation graph database and drug paths) have been made publicly available to the biomedical text mining community and clinical researchers. Conclusions: This study provided a comprehensive analysis of RE with diverse relation types. Optimized RE models and tools for diverse relation types were developed, which can be widely used in biomedical text mining. Our usability studies provided a proof-of-concept demonstration of how large-scale RE can be leveraged to facilitate novel research.
- Open Access
- 11 April 2023
- Published by JMIR Publications Inc. in JMIR Preprints
Large-Scale Biomedical Relation Extraction Across Diverse Relation Types: Model Development and Usability Study on COVID-19 (Preprint)
BACKGROUND Biomedical relation extraction (RE) is of great importance for researchers to conduct systematic biomedical studies. It not only helps knowledge mining, such as knowledge graphs and novel knowledge discovery, but also promotes translational applications, such as clinical diagnosis, decision-making, and precision medicine. However, the relations between biomedical entities are complex and diverse, and comprehensive biomedical RE is not yet well established. OBJECTIVE We aimed to investigate and improve large-scale RE with diverse relation types and conduct usability studies with application scenarios to optimize biomedical text mining. METHODS Data sets containing 125 relation types with different entity semantic levels were constructed to evaluate the impact of entity semantic information on RE, and performance analysis was conducted on different model architectures and domain models. This study also proposed a continued pretraining strategy and integrated models with scripts into a tool. Furthermore, this study applied RE to the COVID-19 corpus with article topics and application scenarios of clinical interest to assess and demonstrate its biological interpretability and usability. RESULTS The performance analysis revealed that RE achieves the best performance when the detailed semantic type is provided. For a single model, PubMedBERT with continued pretraining performed the best, with an F1-score of 0.8998. Usability studies on COVID-19 demonstrated the interpretability and usability of RE, and a relation graph database was constructed, which was used to reveal existing and novel drug paths with edge explanations. The models (including pretrained and fine-tuned models), integrated tool (Docker), and generated data (including the COVID-19 relation graph database and drug paths) have been made publicly available to the biomedical text mining community and clinical researchers. CONCLUSIONS This study provided a comprehensive analysis of RE with diverse relation types. Optimized RE models and tools for diverse relation types were developed, which can be widely used in biomedical text mining. Our usability studies provided a proof-of-concept demonstration of how large-scale RE can be leveraged to facilitate novel research.
- 11 April 2023
- Published by JMIR Publications Inc. in JMIR Preprints
Open AccessLarge-Scale Biomedical Relation Extraction Across Diverse Relation Types: Model Development and Usability Study on COVID-19 (Preprint)
BACKGROUND Biomedical relation extraction (RE) is of great importance for researchers to conduct systematic biomedical studies. It not only helps knowledge mining, such as knowledge graphs and novel knowledge discovery, but also promotes translational applications, such as clinical diagnosis, decision-making, and precision medicine. However, the relations between biomedical entities are complex and diverse, and comprehensive biomedical RE is not yet well established. OBJECTIVE We aimed to investigate and improve large-scale RE with diverse relation types and conduct usability studies with application scenarios to optimize biomedical text mining. METHODS Data sets containing 125 relation types with different entity semantic levels were constructed to evaluate the impact of entity semantic information on RE, and performance analysis was conducted on different model architectures and domain models. This study also proposed a continued pretraining strategy and integrated models with scripts into a tool. Furthermore, this study applied RE to the COVID-19 corpus with article topics and application scenarios of clinical interest to assess and demonstrate its biological interpretability and usability. RESULTS The performance analysis revealed that RE achieves the best performance when the detailed semantic type is provided. For a single model, PubMedBERT with continued pretraining performed the best, with an F1-score of 0.8998. Usability studies on COVID-19 demonstrated the interpretability and usability of RE, and a relation graph database was constructed, which was used to reveal existing and novel drug paths with edge explanations. The models (including pretrained and fine-tuned models), integrated tool (Docker), and generated data (including the COVID-19 relation graph database and drug paths) have been made publicly available to the biomedical text mining community and clinical researchers. CONCLUSIONS This study provided a comprehensive analysis of RE with diverse relation types. Optimized RE models and tools for diverse relation types were developed, which can be widely used in biomedical text mining. Our usability studies provided a proof-of-concept demonstration of how large-scale RE can be leveraged to facilitate novel research.
- Open Access
- 1 September 2020
- Published by IOP Publishing in Journal of Physics: Conference Series
- Vol. 1634 (1) , 012034
Recognition of Common Insect in Field Based on Deep Learning
Traditional insect taxonomy methods have high technical requirements, and the ability of artificial identification of insects is insufficient. In order to solve the problem, this paper proposes a method of common insects recognition in field based on transfer learning. A total of 9 kinds of insects, such as mythimna separata, rice borer, rice plant hopper, mole cricket, mantis, locust, grass fly, ladybug, and ditch beetle are collected for classification and identification, those include the main insect pests and some natural enemies of the main food crops in the field, such as wheat, rice, corn, etc. Then we use the digital image processing technology and the confrontation generation network to expend the insect dataset, and build a model based on transfer learning to transfer the knowledge learned by VGG16, VGG19, InceptionV3, and InceptionV4 on the ImageNet dataset to the insect classification and recognition. Experimental results show that the transfer learning training model has better classification performance and higher convergence speed, and data expansion can help extend sample and avoid overfitting. The highest recognition accuracy is up to 97.39% among models, which adopt the VGG19 convolutional neural network to pretrain the model for transfer learning. This method has the high recognition accuracy, less time consumption, simple and convenient, robustness in particular for the translation and rotation, which provides a reference for the identification and classification method of field insects.
- 1 September 2020
- Published by IOP Publishing in Journal of Physics: Conference Series
- Vol. 1634 (1) , 012034
Open AccessRecognition of Common Insect in Field Based on Deep Learning
Traditional insect taxonomy methods have high technical requirements, and the ability of artificial identification of insects is insufficient. In order to solve the problem, this paper proposes a method of common insects recognition in field based on transfer learning. A total of 9 kinds of insects, such as mythimna separata, rice borer, rice plant hopper, mole cricket, mantis, locust, grass fly, ladybug, and ditch beetle are collected for classification and identification, those include the main insect pests and some natural enemies of the main food crops in the field, such as wheat, rice, corn, etc. Then we use the digital image processing technology and the confrontation generation network to expend the insect dataset, and build a model based on transfer learning to transfer the knowledge learned by VGG16, VGG19, InceptionV3, and InceptionV4 on the ImageNet dataset to the insect classification and recognition. Experimental results show that the transfer learning training model has better classification performance and higher convergence speed, and data expansion can help extend sample and avoid overfitting. The highest recognition accuracy is up to 97.39% among models, which adopt the VGG19 convolutional neural network to pretrain the model for transfer learning. This method has the high recognition accuracy, less time consumption, simple and convenient, robustness in particular for the translation and rotation, which provides a reference for the identification and classification method of field insects.