Study on structured method of Chinese MRI report of nasopharyngeal carcinoma

BMC Medical Informatics and Decision Making

Jing-Dong Yan¹  Xin Huang²  Hui Chen³ 

[1] Nanfang Hospital, Southern Medical University, 510515, Guangzhou, Guangdong, China;School of Biomedical Engineering, Southern Medical University, 510515, Guangzhou, Guangdong, China;Nanfang Hospital, Southern Medical University, 510515, Guangzhou, Guangdong, China;Shuguang Hospital, Shanghai University of Traditional Chinese Medicine, 201203, Shanghai, China;

关键词: Structured medical text;    Named entity recognition;    Knowledge network;   

DOI  :  10.1186/s12911-021-01547-1

来源: Springer

【 摘 要 】

BackgroundImage text is an important text data in the medical field at it can assist clinicians in making a diagnosis. However, due to the diversity of languages, most descriptions in the image text are unstructured data. The same medical phenomenon may also be described in various ways, such that it remains challenging to conduct text structure analysis. The aim of this research is to develop a feasible approach that can automatically convert nasopharyngeal cancer reports into structured text and build a knowledge network.MethodsIn this work, we compare commonly used named entity recognition (NER) models, choose the optimal model as our triplet extraction model, and present a Chinese structuring algorithm. Finally, we visualize the results of the algorithm in the form of a knowledge network of nasopharyngeal cancer.ResultsIn NER, both accuracy and recall of the BERT-CRF model reached 99%. The structured extraction rate is 84.74%, and the accuracy is 89.39%. The architecture based on recurrent neural network does not rely on medical dictionaries or word segmentation tools and can realize triplet recognition.ConclusionsThe BERT-CRF model has high performance in NER, and the triplet can reflect the content of the image report. This work can provide technical support for the construction of a nasopharyngeal cancer database.

【 授权许可】

CC BY