This paper summarizes Professor WANG Yaoxian's experience in treating diabetic kidney disease （DKD） from the perspective of spleen and stomach based on the "internal heat leading to concretions" theory. It is considered that internal heat leading to concretions constitutes the core pathogenesis of DKD， with the spleen and stomach serving as the source of internal heat； therefore， treatment should be based on regulating the spleen and stomach. In the early stage of DKD， dysfunction of the spleen and stomach leads to the initial generation of internal heat. Common syndrome patterns include gastrointestinal heat accumulation and constrained heat in the liver and stomach， for which modified Gegen Qinlian Decoction （葛根芩连汤） can be used to clear heat bind while modified Dachaihu Decoction （大柴胡汤） is used to clear stomach and soothe liver， respectively. In the middle stage of DKD， weakness of the spleen and stomach results in the initial formation of concretions and conglomerations. Common patterns include spleen deficiency with prevalence of dampness and deficiency of both the spleen and kidney. Treatment emphasizes strengthening the spleen and resolving dampness， raising yang and boosting the stomach with modified Shengyang Yiwei Decoction （升阳益胃汤）， or supplementing spleen and boosting kidney， dissipating bind and dispe-ring concretions with modified Shenqi Dihuang Decoction （参芪地黄汤）， respectively. In the late stage of DKD， it is characterized by spleen and stomach depletion， and rampant accumulation of turbidity and toxin， and the common syndrome patterns are damp-turbidity obstruction in the middle jiao （焦） and spleen-kidney yang deficiency. Treatment aims to remove turbidity and harmonize the stomach， or to warm the kidney and strengthen the spleen while elimina-ting turbidity， using modified Dahuang Gancao Decoction（大黄甘草汤） and Jupi Zhuru Decoction （橘皮竹茹汤） or modified Baoyuan Decoction （保元汤） and Lizhong Decoction （理中汤）， respectively. In clinical practice， appropriate formulas and medications are flexibly selected according to specific syndromes.

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.

With the rapid development of minimally invasive techniques in surgery, arti-ficial intelligence (AI), particularly deep learning, is playing an increasingly important role in mini-mally invasive surgery. By automated analysis of surgical videos, AI can efficiently perform key tasks such as instrument recognition, surgical phase identification, action analysis, anatomical structure recognition, intraoperative diagnosis, adverse event monitoring and smart desmoking. These appli-cations provide essential support for real-time monitoring, surgical navigation and skill assessment during surgery. The authors summarize the current research progress of AI in minimally invasive surgery, including its applications in the fields of hepatobiliary and pancreatic surgery, as well as gastrointestinal surgery. It also explores the potential of AI in enhancing surgical safety, efficiency and skill assessment. By synthesizing the latest research achievements of AI technology in the field of surgery, as well as analyzing its technical challenges and risks, it aims to provide guidance for future innovations and clinical applications, promoting the advancement and implementation of AI in minimally invasive surgery.

OBJECTIVE:To systematically evaluate the clinical effectiveness of rehabilitation robots in treating lower limb motor dysfunction in patients with cerebral palsy,and to compare the differences in therapeutic effects among different robots.METHODS:PubMed,Web of Science,Embase,Cochrane Library,CNKI,CBM,VIP and WanFang databases were searched to collect randomized controlled trials on rehabilitation robotics for the treatment of motor dysfunction in patients with cerebral palsy,published from database inception to April 10,2024.The main outcome indicators included muscle strength,muscle tension,balance function,step speed,step frequency,step length,walking endurance,lower limb motor function,and activities of daily living.The above indicators were coded according to the International Classification of Functioning,Disability and Health.Meta-analysis was performed to evaluate the clinical efficacy and compare the therapeutic efficacy of different rehabilitation robots.Literature search and screening were performed by two researchers,and the quality of the included literature was evaluated using the Cochrane 5.1.0 risk of bias assessment tool.Meta-analysis was performed using RevMan 5.4 software and Stata 16.0 software.RESULTS:(1)Fifteen articles were finally included,involving 512 patients with 260 in the experimental group and 252 in the control group.(2)The Meta-analysis results showed that rehabilitation robots could improve body structure and function[standardized mean difference=0.41,95%confidence interval(CI):0.24-0.58,P<0.05],activities(standardized mean difference=0.53,95%CI:0.41-0.65,P<0.05)and participation ability(mean difference=7.86,95%CI:1.54-14.18,P<0.05).In particular,the rehabilitation robot improved lower limb muscle strength,balance function,step speed,walking endurance,lower limb gross motor function,and activities of daily living in patients with cerebral palsy,but showed insignificant effects on step frequency,step length,and muscle tension.(3)The network Meta-analysis results showed that:step speed:Innowalkpro>Gait Trainer>Lokomat>3DCalt;6-minute walk test score:Gait Trainer>Lokomat>Lokohelp>Innowalkro;Gross Motor Function Measure-88D score:Lokohelp>Lokomat>KidGo>Innowalkpro>3DCalt;Gross Motor Function Measure-88E score:Lokomat>Lokohelp>KidGo>3DCalt>Innowalkpro.CONCLUSION:Based on the International Classification of Functioning,Disability,and Health,rehabilitation robot training can improve the lower limb motor function and activities of daily living in patients with cerebral palsy.The Innowalkpro robot was more effective in improving step speed;the Gait trainer robot was more effective in improving 6-minute walk test scores;the Lokohelp robot was more effective in improving Gross Motor Function Measure-88D zone scores;and the Lokomat robot was more effective in improving Gross Motor Function Measure-88E zone scores.

OBJECTIVE:To systematically evaluate the clinical effectiveness of rehabilitation robots in treating lower limb motor dysfunction in patients with cerebral palsy,and to compare the differences in therapeutic effects among different robots.METHODS:PubMed,Web of Science,Embase,Cochrane Library,CNKI,CBM,VIP and WanFang databases were searched to collect randomized controlled trials on rehabilitation robotics for the treatment of motor dysfunction in patients with cerebral palsy,published from database inception to April 10,2024.The main outcome indicators included muscle strength,muscle tension,balance function,step speed,step frequency,step length,walking endurance,lower limb motor function,and activities of daily living.The above indicators were coded according to the International Classification of Functioning,Disability and Health.Meta-analysis was performed to evaluate the clinical efficacy and compare the therapeutic efficacy of different rehabilitation robots.Literature search and screening were performed by two researchers,and the quality of the included literature was evaluated using the Cochrane 5.1.0 risk of bias assessment tool.Meta-analysis was performed using RevMan 5.4 software and Stata 16.0 software.RESULTS:(1)Fifteen articles were finally included,involving 512 patients with 260 in the experimental group and 252 in the control group.(2)The Meta-analysis results showed that rehabilitation robots could improve body structure and function[standardized mean difference=0.41,95%confidence interval(CI):0.24-0.58,P<0.05],activities(standardized mean difference=0.53,95%CI:0.41-0.65,P<0.05)and participation ability(mean difference=7.86,95%CI:1.54-14.18,P<0.05).In particular,the rehabilitation robot improved lower limb muscle strength,balance function,step speed,walking endurance,lower limb gross motor function,and activities of daily living in patients with cerebral palsy,but showed insignificant effects on step frequency,step length,and muscle tension.(3)The network Meta-analysis results showed that:step speed:Innowalkpro>Gait Trainer>Lokomat>3DCalt;6-minute walk test score:Gait Trainer>Lokomat>Lokohelp>Innowalkro;Gross Motor Function Measure-88D score:Lokohelp>Lokomat>KidGo>Innowalkpro>3DCalt;Gross Motor Function Measure-88E score:Lokomat>Lokohelp>KidGo>3DCalt>Innowalkpro.CONCLUSION:Based on the International Classification of Functioning,Disability,and Health,rehabilitation robot training can improve the lower limb motor function and activities of daily living in patients with cerebral palsy.The Innowalkpro robot was more effective in improving step speed;the Gait trainer robot was more effective in improving 6-minute walk test scores;the Lokohelp robot was more effective in improving Gross Motor Function Measure-88D zone scores;and the Lokomat robot was more effective in improving Gross Motor Function Measure-88E zone scores.

With the rapid development of minimally invasive techniques in surgery, arti-ficial intelligence (AI), particularly deep learning, is playing an increasingly important role in mini-mally invasive surgery. By automated analysis of surgical videos, AI can efficiently perform key tasks such as instrument recognition, surgical phase identification, action analysis, anatomical structure recognition, intraoperative diagnosis, adverse event monitoring and smart desmoking. These appli-cations provide essential support for real-time monitoring, surgical navigation and skill assessment during surgery. The authors summarize the current research progress of AI in minimally invasive surgery, including its applications in the fields of hepatobiliary and pancreatic surgery, as well as gastrointestinal surgery. It also explores the potential of AI in enhancing surgical safety, efficiency and skill assessment. By synthesizing the latest research achievements of AI technology in the field of surgery, as well as analyzing its technical challenges and risks, it aims to provide guidance for future innovations and clinical applications, promoting the advancement and implementation of AI in minimally invasive surgery.

Objective We aimed to evaluate the efficacy and safety of Shengxuebao Mixture in the treatment of cancer-related anemia(CRA)presenting with syndrome of deficiency of liver and kidney combined with syndrome of deficiency of both qi and blood.Methods A randomized,double-blind,placebo-controlled,multicenter clinical trial was conducted.Eligible patients with malignant tumors meeting the inclusion and exclusion criteria were enrolled from 26 hospitals,including Dongzhimen Hospital,Beijing University of Chinese Medicine,Xiaogan Central Hospital,and Yangzhou Hospital of Traditional Chinese Medicine,from June 1,2022,to September 30,2024.Patients were allocated 1:1 to either the experimental group receiving Shengxuebao Mixture or the control group receiving its simulator(placebo)using a block randomization method under double-blind conditions.Both groups received 15 mL orally three times daily for 28 consecutive days.The primary efficacy indicators included the hemoglobin(Hb)improvement rate(RHb)and the traditional Chinese medicine(TCM)syndrome improvement rate(RTCM)at week 4 of treatment.The secondary efficacy indicators encompassed Hb and red blood cell(RBC)count,Karnofsky Performance Status(KPS)score,TCM syndrome score,individual TCM symptom scores,and changes in each of these indicators compared to the baseline period at weeks 2,4,and 6 of treatment.Safety evaluations were conducted at week 4 of treatment.Results A total of 239 patients were enrolled,with 225 cases included in the Full Analysis Set(FAS)(109 in the experimental group vs.116 control group),163 in the Per Protocol Set(PPS)(77 vs.86),and 225 in the Safety Set(SS)(109 vs.116).Baseline characteristics between groups showed no significant differences.Significant differences were observed between the experimental and control groups in RHb at week 4(FAS:49.51%vs.35.24%,P<0.05;PPS:53.25%vs.36.05%,P<0.05)and RTCM at week 4(FAS:61.54%vs.39.62%,P<0.01;PPS:64.94%vs.40.70%,P<0.01).At weeks 2,4,and 6,the experimental group showed greater improvements in Hb and RBC counts than the control group.Additionally,the TCM syndrome scores were lower in the experimental group than in the control group at these time points.Except for week 2 in PPS,the KPS improvement was better in the experimental group than in the control group(P<0.05).The experimental group also demonstrated a greater reduction in scores for individual TCM symptoms such as spiritlessness and weakness,poor appetite and reduced food intake at weeks 4 and 6 compared to the control group(P<0.05,P<0.01).Furthermore,the reduction in vertigo score was more pronounced in the experimental group at week 6(P<0.01).For the score of pale and lusterless complexion,only in the PPS was the reduction from baseline more significant in the experimental group than in the control group at weeks 4 and 6(P<0.05).No significant differences were observed between the experimental and control groups in the incidence of all adverse events or drug-related adverse reactions.Conclusion Shengxuebao Mixture demonstrates significant efficacy in patients with CRA presenting syndrome of deficiency of liver and kidney combined with syndrome of deficiency of both qi and blood,effectively increasing Hb levels,ameliorating TCM syndromes,alleviating clinical symptoms,and enhancing functional status,with no significant difference in adverse drug reactions compared to the placebo.

Objective We aimed to evaluate the efficacy and safety of Shengxuebao Mixture in the treatment of cancer-related anemia(CRA)presenting with syndrome of deficiency of liver and kidney combined with syndrome of deficiency of both qi and blood.Methods A randomized,double-blind,placebo-controlled,multicenter clinical trial was conducted.Eligible patients with malignant tumors meeting the inclusion and exclusion criteria were enrolled from 26 hospitals,including Dongzhimen Hospital,Beijing University of Chinese Medicine,Xiaogan Central Hospital,and Yangzhou Hospital of Traditional Chinese Medicine,from June 1,2022,to September 30,2024.Patients were allocated 1:1 to either the experimental group receiving Shengxuebao Mixture or the control group receiving its simulator(placebo)using a block randomization method under double-blind conditions.Both groups received 15 mL orally three times daily for 28 consecutive days.The primary efficacy indicators included the hemoglobin(Hb)improvement rate(RHb)and the traditional Chinese medicine(TCM)syndrome improvement rate(RTCM)at week 4 of treatment.The secondary efficacy indicators encompassed Hb and red blood cell(RBC)count,Karnofsky Performance Status(KPS)score,TCM syndrome score,individual TCM symptom scores,and changes in each of these indicators compared to the baseline period at weeks 2,4,and 6 of treatment.Safety evaluations were conducted at week 4 of treatment.Results A total of 239 patients were enrolled,with 225 cases included in the Full Analysis Set(FAS)(109 in the experimental group vs.116 control group),163 in the Per Protocol Set(PPS)(77 vs.86),and 225 in the Safety Set(SS)(109 vs.116).Baseline characteristics between groups showed no significant differences.Significant differences were observed between the experimental and control groups in RHb at week 4(FAS:49.51%vs.35.24%,P<0.05;PPS:53.25%vs.36.05%,P<0.05)and RTCM at week 4(FAS:61.54%vs.39.62%,P<0.01;PPS:64.94%vs.40.70%,P<0.01).At weeks 2,4,and 6,the experimental group showed greater improvements in Hb and RBC counts than the control group.Additionally,the TCM syndrome scores were lower in the experimental group than in the control group at these time points.Except for week 2 in PPS,the KPS improvement was better in the experimental group than in the control group(P<0.05).The experimental group also demonstrated a greater reduction in scores for individual TCM symptoms such as spiritlessness and weakness,poor appetite and reduced food intake at weeks 4 and 6 compared to the control group(P<0.05,P<0.01).Furthermore,the reduction in vertigo score was more pronounced in the experimental group at week 6(P<0.01).For the score of pale and lusterless complexion,only in the PPS was the reduction from baseline more significant in the experimental group than in the control group at weeks 4 and 6(P<0.05).No significant differences were observed between the experimental and control groups in the incidence of all adverse events or drug-related adverse reactions.Conclusion Shengxuebao Mixture demonstrates significant efficacy in patients with CRA presenting syndrome of deficiency of liver and kidney combined with syndrome of deficiency of both qi and blood,effectively increasing Hb levels,ameliorating TCM syndromes,alleviating clinical symptoms,and enhancing functional status,with no significant difference in adverse drug reactions compared to the placebo.

With the rapid development of generative artificial intelligence(GAI)technologies,their widespread application in academic research and writing is continuously expanding the boundaries of sci-entific inquiry.However,this trend has also raised a series of ethical and regulatory challenges,inclu-ding issues related to authorship,content authenticity,citation accuracy,and accountability.In light of the growing involvement of AI in generating academic content,establishing an open,controllable,and trustworthy ethical governance framework has become a key task for safeguarding research integrity and maintaining trust within the academic community.This expert consensus outlines ethical requirements across key stages of AI-assisted academic writing-including topic selection,data management,citation practices,and authorship attribution.It aims to clarify the boundaries and ethical obligations surrounding AI use in academic writing,ensuring that technological tools enhance efficiency without compromising in-tegrity.The goal is to provide guidance and institutional support for building a responsible and sustainable research ecosystem.