BACKGROUND Kidney transplantation is the most effective means to treat patients with renal failure,but its postoperative problems such as rejection reactions,immunosuppressant poisoning,chronic transplant kidney nephr...BACKGROUND Kidney transplantation is the most effective means to treat patients with renal failure,but its postoperative problems such as rejection reactions,immunosuppressant poisoning,chronic transplant kidney nephropathy,etc.still have not been effectively solved.This study searched for literature on traditional Chinese medicine(TCM)syndromes after kidney transplantation in China,conducted statistical analysis of the results,and sought to identify the underlying patterns.AIM To understand the TCM syndromes after renal transplantation and associated rules and provide a theoretical basis for further clinical research.METHODS The literature pertaining to TCM syndromes in renal transplantation,published in the China National Knowledge Infrastructure,Wanfang database,and WIP database from 1970 to 2021,was meticulously searched and comprehensively and statistically analyzed.RESULTS Following the established inclusion and exclusion criteria,13 studies were selected for analysis.Post-renal transplantation,no significant discrepancy was noted among the groups based on the location of TCM viscera.However,when categorized according to TCM pathogenic factors,the groups with spleen and kidney yang deficiency,as well as liver and kidney yin deficiency,exhibited a statistically significant difference in the frequency.CONCLUSION Currently,the research on TCM syndromes pertaining to renal transplantation is in its nascent phase.It is imperative to conduct a multicentric,large-scale survey of TCM syndromes subsequent to renal transplantation in the ensuing years.展开更多
BACKGROUND Hypertrophic cardiomyopathy(HCM)is one of the most prevalent inherited myocardial disorders and is charac-terized by considerable genetic and phenotypic heterogeneity.A subset of patients with HCM progress ...BACKGROUND Hypertrophic cardiomyopathy(HCM)is one of the most prevalent inherited myocardial disorders and is charac-terized by considerable genetic and phenotypic heterogeneity.A subset of patients with HCM progress to a dilated phase of HCM(DPHCM),which is associated with a poor prognosis;however,the underlying pathogenesis remains inadequately understood.CASE SUMMARY In this study,we present a case involving a pedigree with familial DPHCM and conduct a retrospective review of patients with DPHCM with identified gene mutations.Through panel sequencing targeting the coding regions of 312 genes associated with inherited cardiomyopathy,a heterozygous missense mutation(c.746G>A,p.Arg249Glu)in the MYH7 gene was identified in the proband(III-5).Sanger sequencing subsequently confirmed this pathogenic mutation in three additional family members(II-4,III-4,and IV-3).A total of 26 well-documented patients with DPHCM were identified in the literature.Patients with DPHCM are commonly middle-aged and male.The mean age of patients with DPHCM was 53.43±12.79 years.Heart failure,dyspnoea,and atrial fibrillation were the most prevalent symptoms observed,accompanied by an average left ventricular end-diastolic size of 58.62 mm.CONCLUSION Our findings corroborate the pathogenicity of the MYH7(c.746G>A,p.Arg249Glu)mutation for DPHCM and suggest that the Arg249Gln mutation may be responsible for high mortality.展开更多
本文提出了一种基于双交叉注意力融合的Swin-AK Transformer(Swin Transformer based on alterable kernel convolution)和手工特征相结合的智能手机拍摄图像质量评价方法。首先,提取了影响图像质量的手工特征,这些特征可以捕捉到图像...本文提出了一种基于双交叉注意力融合的Swin-AK Transformer(Swin Transformer based on alterable kernel convolution)和手工特征相结合的智能手机拍摄图像质量评价方法。首先,提取了影响图像质量的手工特征,这些特征可以捕捉到图像中细微的视觉变化;其次,提出了Swin-AK Transformer,增强了模型对局部信息的提取和处理能力。此外,本文设计了双交叉注意力融合模块,结合空间注意力和通道注意力机制,融合了手工特征与深度特征,实现了更加精确的图像质量预测。实验结果表明,在SPAQ和LIVE-C数据集上,皮尔森线性相关系数分别达到0.932和0.885,斯皮尔曼等级排序相关系数分别达到0.929和0.858。上述结果证明了本文提出的方法能够有效地预测智能手机拍摄图像的质量。展开更多
近年来,Transformer在众多监督式计算机视觉任务中取得了显著进展,然而由于高质量医学标注图像的缺乏,其在半监督图像分割领域的性能仍有待提高。为此,提出了一种基于多尺度和多视图Transformer的半监督医学图像分割框架:MSMVT(multi-sc...近年来,Transformer在众多监督式计算机视觉任务中取得了显著进展,然而由于高质量医学标注图像的缺乏,其在半监督图像分割领域的性能仍有待提高。为此,提出了一种基于多尺度和多视图Transformer的半监督医学图像分割框架:MSMVT(multi-scale and multi-view transformer)。鉴于对比学习在Transformer的预训练中取得的良好效果,设计了一个基于伪标签引导的多尺度原型对比学习模块。该模块利用图像金字塔数据增强技术,为无标签图像生成富有语义信息的多尺度原型表示;通过对比学习,强化了不同尺度原型之间的一致性,从而有效缓解了由标签稀缺性导致的Transformer训练不足的问题。此外,为了增强Transformer模型训练的稳定性,提出了多视图一致性学习策略。通过弱扰动视图,以校正多个强扰动视图。通过最小化不同视图之间的输出差异性,使得模型能够对不同扰动保持多层次的一致性。实验结果表明,当仅采用10%的标注比例时,提出的MSMVT框架在ACDC、LIDC和ISIC三个公共数据集上的DSC图像分割性能指标分别达到了88.93%、84.75%和85.38%,优于现有的半监督医学图像分割方法。展开更多
文摘BACKGROUND Kidney transplantation is the most effective means to treat patients with renal failure,but its postoperative problems such as rejection reactions,immunosuppressant poisoning,chronic transplant kidney nephropathy,etc.still have not been effectively solved.This study searched for literature on traditional Chinese medicine(TCM)syndromes after kidney transplantation in China,conducted statistical analysis of the results,and sought to identify the underlying patterns.AIM To understand the TCM syndromes after renal transplantation and associated rules and provide a theoretical basis for further clinical research.METHODS The literature pertaining to TCM syndromes in renal transplantation,published in the China National Knowledge Infrastructure,Wanfang database,and WIP database from 1970 to 2021,was meticulously searched and comprehensively and statistically analyzed.RESULTS Following the established inclusion and exclusion criteria,13 studies were selected for analysis.Post-renal transplantation,no significant discrepancy was noted among the groups based on the location of TCM viscera.However,when categorized according to TCM pathogenic factors,the groups with spleen and kidney yang deficiency,as well as liver and kidney yin deficiency,exhibited a statistically significant difference in the frequency.CONCLUSION Currently,the research on TCM syndromes pertaining to renal transplantation is in its nascent phase.It is imperative to conduct a multicentric,large-scale survey of TCM syndromes subsequent to renal transplantation in the ensuing years.
基金Supported by National Natural Science Foundation of China,No.81770379.
文摘BACKGROUND Hypertrophic cardiomyopathy(HCM)is one of the most prevalent inherited myocardial disorders and is charac-terized by considerable genetic and phenotypic heterogeneity.A subset of patients with HCM progress to a dilated phase of HCM(DPHCM),which is associated with a poor prognosis;however,the underlying pathogenesis remains inadequately understood.CASE SUMMARY In this study,we present a case involving a pedigree with familial DPHCM and conduct a retrospective review of patients with DPHCM with identified gene mutations.Through panel sequencing targeting the coding regions of 312 genes associated with inherited cardiomyopathy,a heterozygous missense mutation(c.746G>A,p.Arg249Glu)in the MYH7 gene was identified in the proband(III-5).Sanger sequencing subsequently confirmed this pathogenic mutation in three additional family members(II-4,III-4,and IV-3).A total of 26 well-documented patients with DPHCM were identified in the literature.Patients with DPHCM are commonly middle-aged and male.The mean age of patients with DPHCM was 53.43±12.79 years.Heart failure,dyspnoea,and atrial fibrillation were the most prevalent symptoms observed,accompanied by an average left ventricular end-diastolic size of 58.62 mm.CONCLUSION Our findings corroborate the pathogenicity of the MYH7(c.746G>A,p.Arg249Glu)mutation for DPHCM and suggest that the Arg249Gln mutation may be responsible for high mortality.
文摘本文提出了一种基于双交叉注意力融合的Swin-AK Transformer(Swin Transformer based on alterable kernel convolution)和手工特征相结合的智能手机拍摄图像质量评价方法。首先,提取了影响图像质量的手工特征,这些特征可以捕捉到图像中细微的视觉变化;其次,提出了Swin-AK Transformer,增强了模型对局部信息的提取和处理能力。此外,本文设计了双交叉注意力融合模块,结合空间注意力和通道注意力机制,融合了手工特征与深度特征,实现了更加精确的图像质量预测。实验结果表明,在SPAQ和LIVE-C数据集上,皮尔森线性相关系数分别达到0.932和0.885,斯皮尔曼等级排序相关系数分别达到0.929和0.858。上述结果证明了本文提出的方法能够有效地预测智能手机拍摄图像的质量。
文摘近年来,Transformer在众多监督式计算机视觉任务中取得了显著进展,然而由于高质量医学标注图像的缺乏,其在半监督图像分割领域的性能仍有待提高。为此,提出了一种基于多尺度和多视图Transformer的半监督医学图像分割框架:MSMVT(multi-scale and multi-view transformer)。鉴于对比学习在Transformer的预训练中取得的良好效果,设计了一个基于伪标签引导的多尺度原型对比学习模块。该模块利用图像金字塔数据增强技术,为无标签图像生成富有语义信息的多尺度原型表示;通过对比学习,强化了不同尺度原型之间的一致性,从而有效缓解了由标签稀缺性导致的Transformer训练不足的问题。此外,为了增强Transformer模型训练的稳定性,提出了多视图一致性学习策略。通过弱扰动视图,以校正多个强扰动视图。通过最小化不同视图之间的输出差异性,使得模型能够对不同扰动保持多层次的一致性。实验结果表明,当仅采用10%的标注比例时,提出的MSMVT框架在ACDC、LIDC和ISIC三个公共数据集上的DSC图像分割性能指标分别达到了88.93%、84.75%和85.38%,优于现有的半监督医学图像分割方法。
