Objective: To develop QingNangTCM, a specialized large language model (LLM) tailored for expert-level traditional Chinese medicine (TCM) question-answering and clinical reasoning, addressing the scarcity of domain-specific corpora and specialized alignment. Methods: We constructed QnTCM_Dataset, a corpus of 100 000 entries, by integrating data from ShenNong_TCM_Dataset and SymMap v2.0, and synthesizing additional samples via retrieval-augmented generation (RAG) and persona-driven generation. The dataset comprehensively covers diagnostic inquiries, prescriptions, and herbal knowledge. Utilizing P-Tuning v2, we fine-tuned the GLM-4-9B-Chat backbone to develop QingNangTCM. A multi-dimensional evaluation framework, assessing accuracy, coverage, consistency, safety, professionalism, and fluency, was established using metrics such as bilingual evaluation understudy (BLEU), recall-oriented understudy for gisting evaluation (ROUGE), metric for evaluation of translation with explicit ordering (METEOR), and LLM-as-a-Judge with expert review. Qualitative analysis was conducted across four simulated clinical scenarios: symptom analysis, disease treatment, herb inquiry, and failure cases. Baseline models included GLM-4-9B-Chat, DeepSeek-V2, HuatuoGPT-II (7B), and GLM-4-9B-Chat (freeze-tuning). Results: QingNangTCM achieved the highest scores in BLEU-1/2/3/4 (0.425/0.298/0.137/0.064), ROUGE-1/2 (0.368/0.157), and METEOR (0.218), demonstrating a balanced and superior normalized performance profile of 0.900 across the dimensions of accuracy, coverage, and consistency. Although its ROUGE-L score (0.299) was lower than that of HuatuoGPT-II (7B) (0.351), it significantly outperformed domain-specific models in expert-validated win rates for professionalism (86%) and safety (73%). Qualitative analysis confirmed that the model strictly adheres to the “symptom-syndrome-pathogenesis-treatment” reasoning chain, though occasional misclassifications and hallucinations persisted when dealing with rare medicinal materials and uncommon syndromes. Conclusion Combining domain-specific corpus construction with parameter-efficient prompt tuning enhances the reasoning behavior and domain adaptation of LLMs for TCM-related tasks. This work provides a technical framework for the digital organization and intelligent utilization of TCM knowledge, with potential value for supporting diagnostic reasoning and medical education.

Objective:To summarize and analyze the clinical features of McCune-Albright syndrome (MAS) in 26 children, to improve the understanding of MAS diagnosis and treatment, and to achieve early clinical diagnosis of MAS.Methods:The clinical data of 26 children with MAS treated in Department of Pediatrics, Tongji Hospital Affiliated to Tongji Medical College of Huazhong University of Science and Technology from August 2011 to June 2021 were retrospectively analyzed.Their clinical characteristics were summarized and studied. t-test, Mann- Whitney U test, χ 2 test or Fisher′ s exact probability method was used for comparison between groups. Results:(1) Among the 26 MAS patients enrolled, there were 22 females and 4 males.The average onset age of female and male patients was (5.87±2.94) years old and (7.48±3.36) years old, respectively.(2) In female patients, there were 7 cases with the typical triad and 15 cases with the atypical triad.(3) Female patients had the first symptom of vaginal bleeding (8/22) and breast development (14/22). Among the 4 male children, 1 case had " fracture" and 3 cases " lagged behind their peers in height" at the first visit.(4) Compared with the breast development group, the vaginal bleeding group had an earlier onset age[(4.06±1.88) years old vs.(7.82±1.82) years old] ( t=5.023, P<0.001), earlier bone maturation[(1.26±0.07) vs.(1.09±0.13)] ( t=2.933, P<0.05), a greatly lowered predicted adult height[(-2.16±0.98) SDS vs.(-0.96±1.09) SDS]( t=1.352, P<0.05), a lower blood phosphorus level[(1.41±0.14) nmol/L vs.(1.67±0.24) nmol/L] ( t=1.941, P<0.05), and a significantly elevated alkaline phosphatase level[339(313, 656) U/L vs.243(205, 452) U/L] ( U=1.000, P<0.05). All patients (8 cases) in the vaginal bleeding group had fibrous dysplasia of bone.(5) Ten patients progressed to central precocious puberty (CPP). They showed an older average age of onset[(7.27±2.69)years old vs.(4.69±2.68)years old] ( U=44.000, P<0.05), significantly earlier bone maturation at diagnosis[(1.23±0.11) vs.(1.01±0.13)] ( t=1.834, P<0.05), and a lower predicted adult height[(152.00±4.62) cm vs.(162.10±6.91) cm] ( t=3.805, P<0.05), compared with those who did not progress to CPP.(6) Eleven children developed polyostotic fibrous dysplasia of bone, and most common type (8 cases) was polyostotic fibrous dysplasia of bone, primarily at lower limb bones and skulls.(7) Of the 26 children, 20 cases had scattered Café au lait pigments on the skin.The Café au lait pigmented skin lesions in 35% (7/20) cases crossed the midline. Conclusions:With complicated clinical manifestations, MAS is more common and occurs earlier in girls than boys.It is difficult to clinically diagnose MAS in boys due to the insidious onset and atypical symptoms.Female MAS patients with vaginal bleeding as the first symptom usually have an early age of onset, advanced bone age and lower predicted adult height, so they should be monitored and evaluated clinically.Vaginal bleeding is significantly associated with polyostotic fiber dysplasia of bone in MAS patients.Therefore, it is recommended that patients with vaginal bleeding should undergo a routine bone single-photon emission computed tomography scan.This helps understand the situation of occult fiber dysplasia of bone.Patients with later diagnosis and advanced bone age should be aware of the possibility of progression to CPP.