Abstract: Chinese medical named entity recognition (CMNER) is a fundamental task in medical information extraction. It is crucial for building downstream applications, such as clinical knowledge graphs, and enabling intelligent clinical decision-making. However, existing mainstream approaches, including lexicon-enhanced, span-based, and grid-based tagging methods, struggle with the absence of natural boundaries, complex nested entity structures, and long-range contextual dependencies inherent in clinical texts. To address these challenges, we propose a novel dual-stream fusion with dynamic grid interaction (D2GI) model, which performs deep semantic mining by integrating complementary feature streams and adaptive grid interactions to accurately capture complex entity boundaries and inter-character relations. Specifically, our dual-stream fusion architecture leverages RoFormer to extract long-range dependencies and incorporates Word2Vec to provide stable prior semantics, thereby enhancing the representation of rare medical terms. Furthermore, to overcome the limitations of static refinement, the dynamic grid interaction module employs a gated attention mechanism to adaptively fuse local and global contexts, facilitating accurate recognition of nested entities. Multiple experiments on three public datasets demonstrate that D2GI is superior to state-of-the-art baselines, achieving F1-score improvements of 1.48 percentage points (PPs) on CMeEE-V2, 1.61 PPs on DiaKG, and 3.08 PPs on CCKS2020.

基于双流融合与动态网格交互的中文医学命名实体识别

[1]Albahli S, 2025. An advanced natural language processing framework for Arabic named entity recognition: a novel approach to handling morphological richness and nested entities. Appl Sci, 15(6):3073.

[2]Chang DJ, Chen MS, Liu CZ, et al., 2021. DiaKG: an annotated diabetes dataset for medical knowledge graph construction. Proc 6^th China Conf on Knowledge Graph and Semantic Computing: Knowledge Graph Empowers New Infrastructure Construction, p.308-314.

[3]Chen YX, Dan ZP, Gao Z, et al., 2024. Chinese medical continual named entity recognition based on continual learning and knowledge distillation. Proc IEEE Int Conf on High Performance Computing and Communications, p.885-893.

[4]Dai ZJ, Wang XT, Ni P, et al., 2019. Named entity recognition using BERT BiLSTM CRF for Chinese electronic health records. Proc 12^th Int Congress on Image and Signal Processing BioMedical Engineering and Informatics, p.1-5.

[5]Devlin J, Chang MW, Lee K, et al., 2019. BERT: pre-training of deep bidirectional Transformers for language understanding. Proc Conf of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers), p.4171-4186.

[6]Gu YJ, Qu XY, Wang ZF, et al., 2022. Delving deep into regularity: a simple but effective method for Chinese named entity recognition. Findings of the Association for Computational Linguistics, p.1863-1873.

[7]Hou SX, Qian YR, Chen JY, et al., 2025. HiNER: hierarchical feature fusion for Chinese named entity recognition. Neurocomputing, 611:128667.

[8]Ji XY, Chen LN, Gao H, et al., 2025. A high-precision generality method for Chinese nested named entity recognition. Proc 18^th Int Conf on Wireless Artificial Intelligent Computing Systems and Applications, p.290-301.

[9]Jia HT, Huang J, Zhao K, et al., 2024. Demonstration-based and attention-enhanced grid-tagging network for mention recognition. Electronics, 13(2):261.

[10]Jin ZG, He XY, Wu XD, et al., 2022. A hybrid Transformer approach for Chinese NER with features augmentation. Expert Syst Appl, 209:118385.

[11]Li JY, Fei H, Liu J, et al., 2022. Unified named entity recognition as word–word relation classification. Proc 36^th AAAI Conf on Artificial Intelligence, p.10965-10973.

[12]Li XN, Yan H, Qiu XP, et al., 2020. FLAT: Chinese NER using flat-lattice Transformer. Proc 58^th Annual Meeting of the Association for Computational Linguistics, p.6836-6842.

[13]Liu P, Guo YM, Wang FL, et al., 2022. Chinese named entity recognition: the state of the art. Neurocomputing, 473:37-53.

[14]Liu RB, Wei J, Jia CY, et al., 2021. Modulating language models with emotions. Findings of the Association for Computational Linguistics, p.4332-4339.

[15]Liu T, Gao J, Ni WJ, et al., 2023. A multi-granularity word fusion method for Chinese NER. Appl Sci, 13(5):2789.

[16]Ma C, Huang WK, 2022. Named entity recognition and event extraction in Chinese electronic medical records. Proc 6^th China Conf on Knowledge Graph and Semantic Computing, p.133-138.

[17]Méndez M, Merayo MG, Núñez M, 2023. Long-term traffic flow forecasting using a hybrid CNN-BiLSTM model. Eng Appl Artif Intell, 121:106041.

[18]Shen YL, Song KT, Tan X, et al., 2023. DiffusionNER: boundary diffusion for named entity recognition. Proc 61^st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), p.3875-3890.

[19]Su JL, Murtadha A, Pan SF, et al., 2022. Global pointer: novel efficient span-based approach for named entity recognition.

[20]Su JL, Ahmed M, Lu Y, et al., 2024. RoFormer: enhanced Transformer with rotary position embedding. Neurocomputing, 568:127063.

[21]Wang J, Shou L, Chen K, et al., 2020. Pyramid: a layered model for nested named entity recognition. Proc 58^th Annual Meeting of the Association for Computational Linguistics, p.5918-5928.

[22]Wang Y, Wei Q, Yu H, et al., 2025. Cross-interaction of Chinese characters structures and boundary features for improving clinical named entity recognition. IEEE J Biomed Health Inform, 29(11):8508-8521.

[23]Weigang L, Brom PC, 2025. LLM-BT-Terms: back-translation as a framework for terminology standardization and dynamic semantic embedding.

[24]Wu S, Song XN, Feng ZH, 2021. MECT: multi-metadata embedding based cross-Transformer for Chinese named entity recognition. Proc 59^th Annual Meeting of the Association for Computational Linguistics and the 11^th Int Joint Conf on Natural Language Processing (Volume 1: Long Papers), p.1529-1539.

[25]Xiao YL, Ji ZC, Li JQ, et al., 2024. MVT: Chinese NER using multi-view Transformer. IEEE/ACM Trans Audio Speech Lang Process, 32:3656-3668.

[26]Xu L, Li S, Wang YC, et al., 2021. Named entity recognition of BERT-BiLSTM-CRF combined with self-attention. Proc 18^th Int Conf on Web Information Systems and Applications, p.556-564.

[27]Xu X, Ma K, 2025. Named entity recognition for Chinese electronic medical records by integrating knowledge graph and ClinicalBERT. Front Artif Intell, 8:1634774.

[28]Yan H, Sun Y, Li XN, et al., 2023. An embarrassingly easy but strong baseline for nested named entity recognition. Proc 61^st Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers), p.1442-1452.

[29]Yan Y, Kang YF, Huang WB, et al., 2025. Chinese medical named entity recognition utilizing entity association and gate context awareness. PLoS ONE, 20(2):e0319056.

[30]Yang LY, Xing SL, Mao GJ, 2025. MISS: multiple information span scoring for Chinese named entity recognition. Comput Speech Lang, 92:101783.

[31]Yang XM, Gao ZH, Li YM, et al., 2018. Bidirectional LSTM-CRF for biomedical named entity recognition. Proc 14^th Int Conf on Natural Computation, Fuzzy Systems and Knowledge Discovery, p.239-242.

[32]Ye YX, Teufel S, 2021. End-to-end argument mining as biaffine dependency parsing. Proc 16^th Conf of the European Chapter of the Association for Computational Linguistics (Main Volume), p.669-678.

[33]Yu JT, Bohnet B, Poesio M, 2020. Named entity recognition as dependency parsing. Proc 58^th Annual Meeting of the Association for Computational Linguistics, p.6470-6476.

[34]Zhang GS, Gao ML, Li QL, et al., 2024. Multi-modal generative DeepFake detection via visual-language pretraining with gate fusion for cognitive computation. Cogn Comput, 16(6):2953-2966.

[35]Zhang HY, Lyu L, Chang WF, et al., 2025. A Chinese medical named entity recognition method considering length diversity of entities. Eng Appl Artif Intell, 150:110649.

[36]Zhang L, Xia PF, Ma XX, et al., 2024. Enhanced Chinese named entity recognition with multi-granularity BERT adapter and efficient global pointer. Compl Intell Syst, 10(3):4473-4491.

[37]Zhang NY, Chen MS, Bi Z, et al., 2022. CBLUE: a Chinese biomedical language understanding evaluation benchmark. Proc 60^th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), p.7888-7915.

[38]Zhu JY, Ni P, Li YM, et al., 2019. An Word2Vec based on Chinese medical knowledge. Proc IEEE Int Conf on Big Data, p.6263-6265.

Similar articles