[Objective] : To improve the accuracy and professionalism of question-answering (QA) model in traditional Chinese medicine (TCM) lung cancer by integrating large language models with structured knowledge graphs using the knowledge graph (KG) to text-enhanced retrieval-augmented generation (KG2TRAG) method. [Methods] : The TCM lung cancer model (TCMLCM) was constructed by fine-tuning ChatGLM2-6B on the specialized datasets Tianchi TCM, HuangDi, and ShenNong-TCM-Dataset, as well as a TCM lung cancer KG. The KG2TRAG method was applied to enhance the knowledge retrieval, which can convert KG triples into natural language text via ChatGPT-aided linearization, leveraging large language models (LLMs) for context-aware reasoning. For a comprehensive comparison, MedicalGPT, HuatuoGPT, and BenTsao were selected as the baseline models. Performance was evaluated using bilingual evaluation understudy (BLEU), recall-oriented understudy for gisting evaluation (ROUGE), accuracy, and the domain-specific TCM-LCEval metrics, with validation from TCM oncology experts assessing answer accuracy, professionalism, and usability. [Results] : The TCMLCM model achieved the optimal performance across all metrics, including a BLEU score of 32.15%, ROUGE-L of 59.08%, and an accuracy rate of 79.68%. Notably, in the TCM-LCEval assessment specific to the field of TCM, its performance was 3% − 12% higher than that of the baseline model. Expert evaluations highlighted superior performance in accuracy and professionalism. [Conclusion] TCMLCM can provide an innovative solution for TCM lung cancer QA, demonstrating the feasibility of integrating structured KGs with LLMs. This work advances intelligent TCM healthcare tools and lays a foundation for future AI-driven applications in traditional medicine.

Objective To compare the preparation efficiency of mouse pox and mouse hepatitis antibodies between two substrains of BALB/c and Big-BALB/c (B-BALB/c) mice, and to provide a theoretical basis and reference for the selection of laboratory animals in the preparation of monoclonal antibodies inducedin vivo through hybridoma.Methods Individuals weighing more than 5% of the weight of normal animals at 4 weeks of age (the criterion for late selection is more than 10%) were selected from a population of conventionally bred BALB/c mice and bred individually, and a subline of B-BALB/c mice was prepared after 10 generations of selection. A total of 40 BALB/c mice and 40 B-BALB/c mice aged 10 to 11 weeks, half male and half female, were selected and inoculated with the mousepox monoclonal antibody hybridoma cell line G23 or the murine hepatitis monoclonal antibody hybridoma cell line Y15 pre-treated with liquid paraffin, respectively. Mice ascites containing monoclonal antibodies were obtained by in vivo induction. The antibody titer was tested by indirect ELISA. The mice were grouped based on the sub-strains, gender and inoculation type of hybridoma to analyze the ascites production, antibody titer and antibody production, and to evaluate the antibody preparation efficiency of the two BALB/c mouse sub-strains.ResultsAfter 10 generations of breeding, the body weight of 10-week-old male and female B-BALB/c mice increased by 22.3% and 12.8%, respectively, compared with BALB/c mice of the same age. Compared with BALB/c mice, B-BALB/c mice had better tolerance and adaptation to secondary ascites collection. Compared with BALB/c mice, the ascites production and antibody titer during the preparation of antibodies in B-BALB/c mice were significantly increased, especially in the hybridoma cell line G23 vaccination group (both P＜0.000 1) . After inoculation with the hybridoma cell lines G23 or Y15, the average antibody production of B-BALB/c mice (14.99×10⁴ U and 33.22×10⁴ U) was higher than that of BALB/c mice (5.33×10⁴ U and 19.31×10⁴ U) (both P＜0.01). After inoculation with hybridoma cell line G23, the average antibody production per unit body weight of B-BALB/c mice (0.55×10⁴ U/g) was higher than that of BALB/c mice (0.23×10⁴ U/g) (P＜0.000 1). And the antibody production per unit body weight of female B-BALB/c or BALB/c mice was higher than that of male B-BALB/c or BALB/c mice (bothP＜0.01).Conclusion B-BALB/c mice can be used as an alternative to BALB/c mice in the in vivo induction of monoclonal antibody preparation, which can achieve the purpose of reducing the number of experimental animals used, lowering the labor cost, and improving the efficiency of antibody preparation.