该文旨在比较DeepSeek-V3.1与ChatGPT-5在结直肠癌肝转移(CRLM)多学科团队(MDT)决策中的应用表现,评估其与MDT专家意见的一致性,为大语言模型(LLMs)的临床实践提供循证依据与优化方向。该文基于真实世界数据与最新指南,设计了6例涵盖不...该文旨在比较DeepSeek-V3.1与ChatGPT-5在结直肠癌肝转移(CRLM)多学科团队(MDT)决策中的应用表现,评估其与MDT专家意见的一致性,为大语言模型(LLMs)的临床实践提供循证依据与优化方向。该文基于真实世界数据与最新指南,设计了6例涵盖不同肿瘤负荷、基因突变谱和体能状态的虚拟CRLM病例,通过结构化的提示策略,在DeepSeek-V3.1与ChatGPT-5模型中分别生成MDT治疗建议。由4名MDT专家采用7维度5级李克特量表对模型输出进行独立评审。并通过统计学分析对两款模型在各个病例、各项维度和各个学科的表现分别进行比较。2款大语言模型在所有病例中的综合得分均≥4.0分(满分5分),表明其在复杂的MDT决策场景下具备可接受的临床效能。在跨维度分析中,两者在清晰度、个体化程度、抗幻觉能力和伦理安全4项上得分较高,而在准确性、全面性和前沿性方面仍有一定提升空间。DeepSeek-V3.1在整体表现(4.27±0.77 vs 4.08±0.86)、前沿性(3.90±0.65 vs 3.24±0.72)与伦理安全(4.87±0.34 vs 4.58±0.65)方面显著优于ChatGPT-5(P<0.05);在放疗领域亦明显领先(4.55±0.67 vs 3.38±0.91,P<0.01)。ChatGPT-5则在胃肠外科领域表现优于DeepSeek-V3.1(4.48±0.67 vs 4.17±0.85,P=0.02)。DeepSeek-V3.1与ChatGPT-5均表现出为CRLM-MDT决策提供可靠建议的良好能力。其中,Deep Seek-V3.1在前沿知识整合、伦理安全性及放射肿瘤学领域展现出显著优势,而Chat GPT-5则在胃肠外科方面表现更优,二者形成优势互补。该文证实了大型语言模型作为“MDT协作者”的可行性,为缩小地域间诊疗水平差距、提升临床决策效率提供了一项便捷可靠的技术方案。展开更多
Background:Recent studies suggest an association between the expansion of Prevotella copri and the disease severity in children with metabolic dysfunction-associated steatotic liver disease(MASLD).We aimed to investig...Background:Recent studies suggest an association between the expansion of Prevotella copri and the disease severity in children with metabolic dysfunction-associated steatotic liver disease(MASLD).We aimed to investigate the causative role and molecular mechanisms of P.copri in pediatric MASLD.Methods:C57BL/6 J mice aged 3 weeks were fed a high-fat diet(HFD)and orally administered with P.copri for 5 weeks.We assessed the key features of MASLD and the gut microbiota profile.By untargeted metabolomics on mouse fecal samples and the supernatant from P.copri culture,we identified P.copriderived metabolite and tested its effects in vitro.Results:In HFD-fed mice,administration of P.copri significantly promoted liver steatosis.Genes associated with inflammation and fibrosis were significantly upregulated in the livers from the HFD+P.copri group compared with those in the livers from the HFD group.In addition,P.copri reduced gut microbial diversity,increased the proportion of Firmicutes and decreased Bacteroidota.Importantly,5-aminopentanoic acid(5-AVA)was significantly enriched in both mouse feces from the HFD+P.copri group and the culture supernatant of P.copri.In vitro,5-AVA aggravated palmitic acid-induced lipid accumulation in HepG2 cells and primary mouse hepatocytes.Mechanistically,P.copri-produced 5-AVA exacerbated hepatic steatosis by promoting lipogenesis and fatty acid uptake,while also reducing hepatic very-low-density lipoprotein export.Conclusions:Our findings demonstrated that P.copri promotes liver steatosis in HFD-fed juvenile mice through its metabolite 5-AVA,suggesting its potential as a therapeutic target for the management of pediatric MASLD.展开更多
摘要该文旨在比较DeepSeek-V3.1与ChatGPT-5在结直肠癌肝转移(CRLM)多学科团队(MDT)决策中的应用表现,评估其与MDT专家意见的一致性,为大语言模型(LLMs)的临床实践提供循证依据与优化方向。该文基于真实世界数据与最新指南,设计了6例涵盖不同肿瘤负荷、基因突变谱和体能状态的虚拟CRLM病例,通过结构化的提示策略,在DeepSeek-V3.1与ChatGPT-5模型中分别生成MDT治疗建议。由4名MDT专家采用7维度5级李克特量表对模型输出进行独立评审。并通过统计学分析对两款模型在各个病例、各项维度和各个学科的表现分别进行比较。2款大语言模型在所有病例中的综合得分均≥4.0分(满分5分),表明其在复杂的MDT决策场景下具备可接受的临床效能。在跨维度分析中,两者在清晰度、个体化程度、抗幻觉能力和伦理安全4项上得分较高,而在准确性、全面性和前沿性方面仍有一定提升空间。DeepSeek-V3.1在整体表现(4.27±0.77 vs 4.08±0.86)、前沿性(3.90±0.65 vs 3.24±0.72)与伦理安全(4.87±0.34 vs 4.58±0.65)方面显著优于ChatGPT-5(P<0.05);在放疗领域亦明显领先(4.55±0.67 vs 3.38±0.91,P<0.01)。ChatGPT-5则在胃肠外科领域表现优于DeepSeek-V3.1(4.48±0.67 vs 4.17±0.85,P=0.02)。DeepSeek-V3.1与ChatGPT-5均表现出为CRLM-MDT决策提供可靠建议的良好能力。其中,Deep Seek-V3.1在前沿知识整合、伦理安全性及放射肿瘤学领域展现出显著优势,而Chat GPT-5则在胃肠外科方面表现更优,二者形成优势互补。该文证实了大型语言模型作为“MDT协作者”的可行性,为缩小地域间诊疗水平差距、提升临床决策效率提供了一项便捷可靠的技术方案。
基金supported by grants from the National Key R&D Program of China(2022YFA1305600)National Natural Science Foundation of China(82100950,82470602,and 82470600)Natural Science Foundation of Shanghai(23ZR1452600)。
摘要Background:Recent studies suggest an association between the expansion of Prevotella copri and the disease severity in children with metabolic dysfunction-associated steatotic liver disease(MASLD).We aimed to investigate the causative role and molecular mechanisms of P.copri in pediatric MASLD.Methods:C57BL/6 J mice aged 3 weeks were fed a high-fat diet(HFD)and orally administered with P.copri for 5 weeks.We assessed the key features of MASLD and the gut microbiota profile.By untargeted metabolomics on mouse fecal samples and the supernatant from P.copri culture,we identified P.copriderived metabolite and tested its effects in vitro.Results:In HFD-fed mice,administration of P.copri significantly promoted liver steatosis.Genes associated with inflammation and fibrosis were significantly upregulated in the livers from the HFD+P.copri group compared with those in the livers from the HFD group.In addition,P.copri reduced gut microbial diversity,increased the proportion of Firmicutes and decreased Bacteroidota.Importantly,5-aminopentanoic acid(5-AVA)was significantly enriched in both mouse feces from the HFD+P.copri group and the culture supernatant of P.copri.In vitro,5-AVA aggravated palmitic acid-induced lipid accumulation in HepG2 cells and primary mouse hepatocytes.Mechanistically,P.copri-produced 5-AVA exacerbated hepatic steatosis by promoting lipogenesis and fatty acid uptake,while also reducing hepatic very-low-density lipoprotein export.Conclusions:Our findings demonstrated that P.copri promotes liver steatosis in HFD-fed juvenile mice through its metabolite 5-AVA,suggesting its potential as a therapeutic target for the management of pediatric MASLD.