A Combined Manual Annotation and Deep-Learning Natural Language Processing Study on Accurate Entity Extraction in Hereditary Disease Related Biomedical Literature
Interdisciplinary Sciences: Computational Life Sciences · 2024
DOI: 10.1007/s12539-024-00605-2 ↗Kurz gefasst
Fachartikel in der Zeitschrift Interdisciplinary Sciences: Computational Life Sciences, 2024. Verfasst von Dao-Ling Huang und 12 weiteren Autor:innen. Beteiligte Einrichtung: BGI Research, Shenzhen, 518083, China. dlhuang1217@gmail.com. Hauptthemen laut MEDLINE: Natural Language Processing, Deep Learning. Die Arbeit wurde bisher 7-mal zitiert (Crossref) und nennt 49 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Innovative Research Group Project of the National Natural Science Foundation of China.
Abstract (Originalsprache)
We report a combined manual annotation and deep-learning natural language processing study to make accurate entity extraction in hereditary disease related biomedical literature. A total of 400 full articles were manually annotated based on published guidelines by experienced genetic interpreters at Beijing Genomics Institute (BGI). The performance of our manual annotations was assessed by comparing our re-annotated results with those publicly available. The overall Jaccard index was calculated to be 0.866 for the four entity types—gene, variant, disease and species. Both a BERT-based large name entity recognition (NER) model and a DistilBERT-based simplified NER model were trained, validated and tested, respectively. Due to the limited manually annotated corpus, Such NER models were fine-tuned with two phases. The F1-scores of BERT-based NER for gene, variant, disease and species are 97.28%, 93.52%, 92.54% and 95.76%, respectively, while those of DistilBERT-based NER are 95.14%, 86.26%, 91.37% and 89.92%, respectively. Most importantly, the entity type of variant has been extracted by a large language model for the first time and a comparable F1-score with the state-of-the-art variant extraction model tmVar has been achieved. Graphical Abstract
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- BGI Research, Shenzhen, 518083, China. dlhuang1217@gmail.com.
- Clinical Laboratory of BGI Health, BGI-Shenzhen, Shenzhen, 518083, China. dlhuang1217@gmail.com.
- BGI-Wuhan Clinical Laboratories, BGI-Shenzhen, Wuhan, 430074, China.
- Clinical Laboratory of BGI Health, BGI-Shenzhen, Shenzhen, 518083, China.
Förderung
- Innovative Research Group Project of the National Natural Science Foundation of China
Zitieren
APA
Huang, D. L., Zeng, Q., Xiong, Y., Liu, S., Pang, C., Xia, M., Fang, T., Ma, Y., Qiang, C., Zhang, Y., Zhang, Y., Li, H., & Yuan, Y. (2024). A Combined Manual Annotation and Deep-Learning Natural Language Processing Study on Accurate Entity Extraction in Hereditary Disease Related Biomedical Literature. Interdisciplinary Sciences: Computational Life Sciences. https://doi.org/10.1007/s12539-024-00605-2
Vancouver
Huang DL, Zeng Q, Xiong Y, Liu S, Pang C, Xia M, et al. A Combined Manual Annotation and Deep-Learning Natural Language Processing Study on Accurate Entity Extraction in Hereditary Disease Related Biomedical Literature. Interdisciplinary Sciences: Computational Life Sciences. 2024. doi:10.1007/s12539-024-00605-2
BibTeX
@article{Huang2024,
author = {Huang, Dao-Ling and Zeng, Quanlei and Xiong, Yun and Liu, Shuixia and Pang, Chaoqun and Xia, Menglei and Fang, Ting and Ma, Yanli and Qiang, Cuicui and Zhang, Yi and Zhang, Yu and Li, Hong and Yuan, Yuying},
title = {A Combined Manual Annotation and Deep-Learning Natural Language Processing Study on Accurate Entity Extraction in Hereditary Disease Related Biomedical Literature},
journal = {Interdisciplinary Sciences: Computational Life Sciences},
year = {2024},
doi = {10.1007/s12539-024-00605-2}
}
EUVIMED