Standardizing the Translation of Traditional Chinese Medicine Terminology: A Framework for Consistency

Abstract

Traditional Chinese Medicine (TCM) records contain many synonymous terms with different definitions, which do not allow it to be precisely translated and applied, across languages or systems. The absence of a standardized framework hinders cross-linguistic understanding and limits the integration of TCM into global healthcare. No standardized framework could exist to transcend linguistic barriers and integrate traditional Chinese practice into global medicine. The consistent framework for TCM terminology in translation is proposed in this research. The proposed Conversion of Synonymous Terms (CST) model normalizes TCM terminology is to improve the accuracy and consistency of translation. A Dynamic Backtracking Search Optimized Siamese Long Short-Term Memory (DBSO-Siamese LSTM) algorithm is used in text classification and for translation purposes of TCM terms. A large dataset was created from TCM medical records, literature, and expert-verified term dictionaries. The dataset experienced text preprocessing, including tokenization and stop-word removal, to improve model efficiency. To select the most critical terms in the TCM description, Term Frequency-Inverse Document Frequency (TF-IDF) is used. The DBSO algorithm improves the Siamese LSTM performance over TCM terminology classification by optimizing search parameters, and dealing with multimodal data in an efficient technique. Experimental results proved that the DBSO-Siamese LSTM model reached a 92.23% accuracy, 91.78% recall, and 92.15% F1-score. The CST model, being classification-based, improves translation consistency and synonym recognition, thus ensuring the global integration of TCM into modern healthcare systems.