ObjectiveThe essence of syndrome manifestation recognition in traditional Chinese medicine (TCM) is to infer the body’s latent pathogenesis state from clinical observational information, rather than to perform simple label matching. However, previous studies have largely modeled this task as syndrome pattern classification within a fixed label space, which does not adequately reflect the cognition process of TCM syndrome differentiation centered on pathogenesis reasoning, and is also insufficient to capture the openness, semantic variability, and cross-disease reusability of syndrome manifestation expression. This study aimed to investigate whether introducing pathogenesis reasoning chain-of-thought (PR-CoT) supervision into large language models (LLMs) could improve the quality and cognitive consistency of syndrome manifestation recognition and support cross-disease transfer. MethodsSyndrome manifestation recognition was formulated as a conditional generation task under the framework of clinical observational information (X)→pathogenesis structure (Z)→syndrome pattern output (Y), where Z serves as an explicit intermediate structural variable linking the clinical evidence and syndrome judgment. Within this framework, a PR-CoT-supervised dataset for syndrome manifestation recognition was constructed based on medical case records of spleen-stomach disorders. After preprocessing, information extraction, manual proofreading, and data cleaning, the dataset comprised 4 800 training cases, 400 development cases, and 400 test cases. Each sample was annotated with a structured PR-CoT consisting of three progressive levels: clinical information summarization, comprehensive pathogenesis analysis, and syndrome pattern output. Supervised fine-tuning was conducted on open-source LLMs, with an end-to-end model serving as the baseline. Qwen3-32B was used as the primary experimental model, and Qwen3-14B as the scale comparison model. A progressive multidimensional evaluation framework was further established, comprising a structural parsing level, a semantic similarity level, and an expert blind review level. At the structural parsing level, syndrome pattern expressions were decomposed into structural elements and evaluated using Precision, Recall, F1 score, and Jaccard similarity. At the semantic similarity level, independent LLMs scored the theoretical proximity between predicted and reference syndrome patterns. At the expert blind review level, three TCM experts independently evaluated model outputs on two dimensions: syndrome differentiation consistency and terminology standardization of syndrome patterns. In addition, zero-shot cross-disease transfer evaluation was conducted on gynecological and heart-system disorder test sets. ResultsAt the structural parsing level, PR-CoT supervision did not lead to a stable improvement in the element-wise overlap of syndrome pattern structural components. Compared with the corresponding baselines, neither Qwen3-32B nor Qwen3-14B showed consistent advantages in structural matching metrics after the introduction of PR-CoT supervision. In contrast, at the semantic similarity level, PR-CoT supervision produced stable positive gains across different model scales and evaluation systems. The average semantic score of Qwen3-32B increased from 6.425 8 in the baseline model to 6.585 0 after PR-CoT supervision, and that of Qwen3-14B increased from 5.870 0 to 5.964 2. At the expert blind review level, the overall score of Qwen3-32B (PR-CoT) was 7.026 0±0.107 7, higher than 6.416 3±0.288 9 for its baseline. In zero-shot cross-disease testing, the PR-CoT model still showed advantages in semantic evaluation and expert evaluation on both gynecological and heart-system disorder test sets, indicating a certain degree of transferability. ConclusionThe benefits of PR-CoT supervision are mainly reflected in TCM semantic consistency and clinical plausibility, rather than in improved hard matching of structural elements. These findings support understanding syndrome manifestation recognition as a process of generating and expressing latent pathogenesis structures, rather than as a classification task within a traditional fixed label space. By introducing pathogenesis reasoning as an explicit intermediate structure into the modeling process and combining it with a progressive multidimensional evaluation framework, this study provides a methodological pathway for intelligent TCM syndrome differentiation that integrates theoretical alignment, interpretability, and multi-level evaluation.

ObjectiveThe essence of syndrome manifestation recognition in traditional Chinese medicine (TCM) is to infer the body’s latent pathogenesis state from clinical observational information, rather than to perform simple label matching. However, previous studies have largely modeled this task as syndrome pattern classification within a fixed label space, which does not adequately reflect the cognition process of TCM syndrome differentiation centered on pathogenesis reasoning, and is also insufficient to capture the openness, semantic variability, and cross-disease reusability of syndrome manifestation expression. This study aimed to investigate whether introducing pathogenesis reasoning chain-of-thought (PR-CoT) supervision into large language models (LLMs) could improve the quality and cognitive consistency of syndrome manifestation recognition and support cross-disease transfer. MethodsSyndrome manifestation recognition was formulated as a conditional generation task under the framework of clinical observational information (X)→pathogenesis structure (Z)→syndrome pattern output (Y), where Z serves as an explicit intermediate structural variable linking the clinical evidence and syndrome judgment. Within this framework, a PR-CoT-supervised dataset for syndrome manifestation recognition was constructed based on medical case records of spleen-stomach disorders. After preprocessing, information extraction, manual proofreading, and data cleaning, the dataset comprised 4 800 training cases, 400 development cases, and 400 test cases. Each sample was annotated with a structured PR-CoT consisting of three progressive levels: clinical information summarization, comprehensive pathogenesis analysis, and syndrome pattern output. Supervised fine-tuning was conducted on open-source LLMs, with an end-to-end model serving as the baseline. Qwen3-32B was used as the primary experimental model, and Qwen3-14B as the scale comparison model. A progressive multidimensional evaluation framework was further established, comprising a structural parsing level, a semantic similarity level, and an expert blind review level. At the structural parsing level, syndrome pattern expressions were decomposed into structural elements and evaluated using Precision, Recall, F1 score, and Jaccard similarity. At the semantic similarity level, independent LLMs scored the theoretical proximity between predicted and reference syndrome patterns. At the expert blind review level, three TCM experts independently evaluated model outputs on two dimensions: syndrome differentiation consistency and terminology standardization of syndrome patterns. In addition, zero-shot cross-disease transfer evaluation was conducted on gynecological and heart-system disorder test sets. ResultsAt the structural parsing level, PR-CoT supervision did not lead to a stable improvement in the element-wise overlap of syndrome pattern structural components. Compared with the corresponding baselines, neither Qwen3-32B nor Qwen3-14B showed consistent advantages in structural matching metrics after the introduction of PR-CoT supervision. In contrast, at the semantic similarity level, PR-CoT supervision produced stable positive gains across different model scales and evaluation systems. The average semantic score of Qwen3-32B increased from 6.425 8 in the baseline model to 6.585 0 after PR-CoT supervision, and that of Qwen3-14B increased from 5.870 0 to 5.964 2. At the expert blind review level, the overall score of Qwen3-32B (PR-CoT) was 7.026 0±0.107 7, higher than 6.416 3±0.288 9 for its baseline. In zero-shot cross-disease testing, the PR-CoT model still showed advantages in semantic evaluation and expert evaluation on both gynecological and heart-system disorder test sets, indicating a certain degree of transferability. ConclusionThe benefits of PR-CoT supervision are mainly reflected in TCM semantic consistency and clinical plausibility, rather than in improved hard matching of structural elements. These findings support understanding syndrome manifestation recognition as a process of generating and expressing latent pathogenesis structures, rather than as a classification task within a traditional fixed label space. By introducing pathogenesis reasoning as an explicit intermediate structure into the modeling process and combining it with a progressive multidimensional evaluation framework, this study provides a methodological pathway for intelligent TCM syndrome differentiation that integrates theoretical alignment, interpretability, and multi-level evaluation.

ObjectiveTo compare the differences in fungal community composition between lesional and non-lesional scalp areas in patients suffering from severe alopecia areata （AA）， and compare these with healthy scalp areas in control subjects. Additionally， to preliminarily explore the changes in scalp fungal communities in severe AA patients and their potential underlying immunological mechanisms. MethodsA total of 20 severe AA patients and 18 healthy controls were enrolled. Skin swab samples were collected from lesional and non-lesional scalp areas of severe AA patients， as well as from the normal scalp of healthy controls. The fungal internal transcribed spacer （ITS） region was amplified and analyzed using high-throughput sequencing. ResultsThe lesional scalp areas of severe AA patients exhibited higher α-diversity and species richness in fungal communities. Notably， the relative abundance of Ascomycota， along with genera such as Mycosphaerella， Aspergillus， Penicillium， and Wallemia， significantly increased in the bald regions. In contrast， Acremonium and Schizophyllum were more predominant in the non-lesional areas of severe AA patients. ConclusionDistinct region-specific differences in scalp fungal microbiota in severe AA patients suggests that fungal dysbiosis may play a potential role in the pathogenesis of alopecia areata. These findings provide new insights into the disease characteristics of severe AA from the perspective of scalp microecology.

ObjectiveTo develop a nomogram-based risk prediction model for chronic obstructive pulmonary disease (COPD) incidence among the community-dwelling population aged 40 years old and above, so as to provide targeted references for the screening and prevention of COPD. MethodsBased on a natural population cohort in suburban Shanghai, a total of 3 381 randomly selected participants aged ≥40 years underwent pulmonary function tests between July and October 2021. Cox stepwise regression analysis was used to develop overall and gender-specific risk prediction models, along with the construction of corresponding risk nomograms. Model predictive performance was evaluated using the C-indice, area under the curve (AUC) values, and Brier score. Stability was assessed through 10-fold cross-validation and sensitivity analysis. ResultsA total of 3 019 participants were included, with a median follow-up duration of 4.6 years. The COPD incidence density was 17.22 per 1 000 person-years, significantly higher in males (32.04/1 000 person-years) than that in females (7.38/1 000 person-years) (P<0.001). The overall risk prediction model included the variables such as gender, age, education level, BMI, smoking, passive smoking, and respiratory comorbidities. The male-specific model incorporated the variables such as age, BMI, respiratory comorbidities, and smoking, while the female-specific model included age, marital status, respiratory comorbidities, and pulmonary tuberculosis history. The C-indices for the overall, male-specific, and female-specific models were 0.829, 0.749, and 0.807, respectively. The 5-year AUC values were 0.785, 0.658, and 0.811, with Brier scores of 0.103, 0.176, and 0.059, respectively. Both 10-fold cross-validated C-indices and sensitivity analysis (excluding participants with a follow-up duration of <6 months) yielded C-indices were above 0.740. ConclusionThis study developed concise and practical overall and gender-specific COPD risk prediction models and corresponding nomograms. The models demonstrated robust performance in predicting COPD incidence, providing a valuable reference for identifying high-risk populations and formulating targeted screening and personalized management strategies.

Objective:To apply artificial intelligence（AI）in classifying ovarian tumors on ultrasound images，and compare the diagnostic results of several sonographers with varying seniority levels.Methods:A total of 645 patients diagnosed with adnexal masses via gynecological ultrasound examination at Qilu Hospital of Shandong University from January 2021 to December 2024 were enrolled. Three deep learning architectures，i.e.，Alexnet，Densenet121，and Resnet50 were developed and used to internally test the classification effectiveness of ovarian tumors，while the optimal model was selected for external testing. Two junior sonographers and two senior sonographers were recruited to independently diagnose ovarian tumors in the external test dataset. Subsequently，the benign and malignant results of the model's predictions were disclosed to each sonographer，and their revised diagnoses on the same external test data in combination with the best AI model were recorded.Results:The optimal model achieved an accuracy of 0.941，sensitivity of 0.936，and specificity of 0.944 on the internal test dataset，and maintained robust performance on the external test dataset with accuracy of 0.891，sensitivity of 0.880，and specificity of 0.907. Compared to junior sonographers，the optimal model demonstrated significantly higher sensitivity in discriminating benign from malignant ovarian tumors（0.880 vs. 0.723，0.602；all P<0.05）. No statistically significant difference was observed in diagnostic accuracy between the optimal model and senior sonographer 1（ P=0.05）. With assistance from the optimal model，junior sonographers achieved significant improvements in both sensitivity and specificity（sensitivity：0.723 vs. 0.843，0.602 vs. 0.819；specificity：0.778 vs. 0.833，0.685 vs. 0.741；all P<0.05）. Conclusions:The optimal model achieves comparable performance to that of senior sonographers in ovarian tumor classification. With model assistance，the diagnostic performance of junior sonographers is significantly improved.

OBJECTIVE: To review the clinical applications of functional perforator flaps in restoring human body functions. METHODS: An extensive literature review was conducted on both domestic and international publications to summarize the clinical use of functional perforator flaps for functional restoration. RESULTS: Perforator flaps are among the most commonly used flaps in reconstructive surgery. Beyond providing soft tissue repair, they are increasingly employed to reconstruct diverse bodily functions, leading us to propose the concept of the "functional perforator flap". Although various forms of functional perforator flaps are currently utilized, reports are predominantly scattered case studies, lacking systematic organization. Commonly used functional perforator flaps can be categorized into five types: chimeric perforator flaps, perforator flaps for nerve function restoration, perforator flaps for lymphatic drainage enhancement, flow-through perforator flaps, and perforator flaps for restoring bone and joint motion. These flaps significantly broaden the application scope of perforator flaps, elevating the goal of reconstruction from mere wound repair to achieving repair concurrent with functional reconstruction. CONCLUSION The application of various functional perforator flap designs significantly improves wound reconstruction outcomes and represents an effective approach for managing complex defects. Future developments will undoubtedly see more forms of functional perforator flaps reported to meet increasingly sophisticated reconstructive demands.
Humans ; Perforator Flap/blood supply* ; Plastic Surgery Procedures/methods* ; Soft Tissue Injuries/surgery* ; Skin Transplantation/methods* ; Wound Healing

Autism spectrum disorder(ASD),characterized by unknown etiology and high heterogeneity,ne-cessitates precise diagnostic and intervention strategies.Neuroimaging techniques have shown great promise in un-covering the neural mechanisms of ASD,providing a foundation for aided diagnosis and transcranial magnetic stim-ulation(TMS)interventions.This review highlights that integrating multimodal neuroimaging and developing indi-vidualized indices with developmental specificity can significantly improve the accuracy of ASD diagnosis and clas-sification.Furthermore,TMS interventions guided by functional connectivity derived from functional magnetic reso-nance imaging(fMRI)offer a personalized approach to ASD treatment.

OBJECTIVES: To compare perioperative and mid-term results of multiple versus single arterial off-pump coronary artery bypass grafting (OPCABG) in patients with impaired left ventricular function. METHODS: This study was conducted among 86 patients with a left ventricular ejection fraction (LVEF) <50%, who underwent OPCABG at our hospital between January, 2018 and December, 2021. Of these patients, 22 underwent OPCABG with multiple arterial grafts (multiple graft group) and 64 received a single arterial graft in OPCABG (single graft group). The preoperative, intraoperative, and perioperative data were collected, and the patients were followed up for a mean of 29.28±14.84 months. The perioperative outcomes and follow-up results of the patients were compared, and the factors influencing major adverse cardiovascular events (MACE) were identified using logistic regression. Kaplan-Meier analysis was used to compare the postoperative survival rate without MACE. RESULTS: The patients in multiple graft group had a significantly younger age than those in single graft group (P<0.05), but the other baseline data were similar between the two groups (P>0.05). Perioperative mortality, 24-h postoperative drainage volume, length of ICU stay, intubation time, and the incidence of new-onset atrial fibrillation were all similar between the two groups (P>0.05), but the rate of postoperative hypotension was significantly higher in multiple graft group (34.78% vs 11.54%, P=0.009). No significant differences were found in the incidence of MACE or echocardiographic data during the follow-up. Logistic regression identified the female sex (OR: 0.191, 95% CI: 0.049-0.075) and creatinine level (OR: 1.016, 95% CI: 1.000-1.033) as factors affecting postoperative MACE occurrence. Kaplan-Meier analysis showed no significant difference in MACE-free survival rate between the two groups. CONCLUSIONS OPCABG with multiple arterial grafts does not increase severe perioperative complications or the risk of mid-term MACE in patients with impaired left ventricular function.
Humans ; Follow-Up Studies ; Postoperative Complications/epidemiology* ; Ventricular Dysfunction, Left/physiopathology* ; Coronary Artery Bypass, Off-Pump/adverse effects* ; Male ; Female ; Ventricular Function, Left ; Middle Aged ; Risk Factors ; Aged ; Perioperative Period ; Stroke Volume

OBJECTIVE: Rhodiolae Crenulatae Radix et Rhizoma (Hongjingtian in Chinese, RCRR), the roots and rhizomes of Rhodiola crenulata and its application in the medicinal market is very chaotic. In this study, DNA barcoding database and identification engine of Rhodiola species were established, decoction pieces from the medicinal market were identified, and the application and challenges of DNA barcoding in the rapid radiation of Rhodiola species were analyzed. This study provides reference for the protection, rational development, and utilization of endangered resources within Rhodiola species. METHODS: A total of 50 original plant samples from 20 species of the genus Rhodiola from Hebei, Xinjiang, Tibet, Jilin, and other major production areas were collected. Theses samples cover the typical distribution area (Qinghai-Tibetan Platea) of Rhodiola species and other scattered alpine regions (Changbai Mountain, Taibai Mountain, Lushan Mountain, etc.), it encompasses all Rhodiola species with thick rhizomes in China. ITS2 and psbA-trnH barcode of Rhodiola database (BORD) were established and an identification engine named Rhodiola-IDE was developed. The stability and accuracy of the standard DNA barcoding database were evaluated using two datasets. Rhodiola-IDE identified 31 decoction pieces of RCRR from the medicinal material market. RESULTS: The BORD containing 1 532 sequences of 88 Rhodiola species has been established, and the identification efficiency results showed good accuracy and stability. According to the Chinese Pharmacopoeia (2020 edition), 23 samples (74.2%) were identified as authentic R. crenulata, while the rest of the marketed varieties were R. kirilowii, R. dumulosa, and R. fastigiata. The product label "Larger flower, Hongjingtian" was identified as R. crenulata. Samples labeled as "Smaller flower, Hongjingtian" were identified as R. crenulata, R. kirilowii, and R. fastigiata. CONCLUSION ITS2 and psbA-trnH barcodes can identify monophyletic groups represented by R. crenulata. However, for non-monophyletic species, it is necessary to collect as many samples as possible and combine them with multiple markers for joint identification. This study discussed the application and challenges of DNA barcodes in Rhodiola under rapid radiation conditions, providing a scientific basis for the rational development and utilization of Rhodiola varieties.

To investigate the pharmacological substances basis and anti-vitiligo mechanism of Carum carvi L. (caraway) fruits, chemical and blood-absorbed components were identified using mass spectrometry combined with literature study and database analysis. A “blood-absorbed components–target genes–pathways” network was constructed through network pharmacology. The pharmacological effects of Carum carvi L. extract and its key blood-absorbed component, acacetin, were validated in vitro. 72 chemical components were identified in the extract, with 11 prototype blood-absorbed components and 26 metabolites being detected in the plasma of SD rats. 14 key active components and 24 key targets were predicted. In vitro experiments demonstrated that acacetin at 10 μmol/L exhibited melanogenesis-promoting and tyrosinase-activating effects compared with the positive control, significantly upregulating the expression of microphthalmia-associated transcription factor (MITF) and tyrosinase (tyrosinase, TYR). This study comprehensively analyzes the chemical and blood-absorbed components of Carum carvi L. and elucidates its pharmacological substances basis, which provides a theoretical foundation for the anti-vitiligo effects of acacetin.