癫痫是一种慢性神经系统疾病,其分类广泛、机制复杂,具有反复发作和不可预测的特点,对患者的生活造成了一定的影响。深度了解其病理生理机制对疾病的治疗和提高患者生活质量十分重要,静息态功能磁共振成像(resting-state functional mag...癫痫是一种慢性神经系统疾病,其分类广泛、机制复杂,具有反复发作和不可预测的特点,对患者的生活造成了一定的影响。深度了解其病理生理机制对疾病的治疗和提高患者生活质量十分重要,静息态功能磁共振成像(resting-state functional magnetic resonance imaging,rs-fMRI)现已成为探究癫痫脑功能改变的有效方法。目前基于rs-fMRI研究癫痫的数据分析方法主要有低频振幅(amplitude of low frequency fluctuation,ALFF)、局部一致性(regional homogeneity,ReHo)、功能连接(function connection,FC)和图论分析。本文将对rs-fMRI的各种分析方法在癫痫中的应用作一综述,为癫痫的病理生理机制及术前定位、治疗提供重要参考。展开更多
目的探讨平均表观传播扩散MRI(mean apparent propagator-MRI,MAP-MRI)与动态对比增强MRI(dynamic contrast enhanced-MRI,DCE-MRI)在鉴别胶质母细胞瘤(glioblastoma,GBM)与脑转移瘤(brain metastases,BMs)中的临床应用价值。材料与方...目的探讨平均表观传播扩散MRI(mean apparent propagator-MRI,MAP-MRI)与动态对比增强MRI(dynamic contrast enhanced-MRI,DCE-MRI)在鉴别胶质母细胞瘤(glioblastoma,GBM)与脑转移瘤(brain metastases,BMs)中的临床应用价值。材料与方法对经手术病理确诊的GBM[异柠檬酸脱氢酶野生型(isocitrate dehydrogenase-wildtype,IDH-wt)]患者27例及经手术病理确诊或临床随访证实的BMs患者24例行常规MRI序列及扩散频谱成像(diffusion spectrum imaging,DSI)与DCE-MRI检查,DSI经解析得到MAP-MRI的各参数图,DCE-MRI经西门子工作站处理后得到多个参数图。在各参数图上分别测量两组患者肿瘤实质区、瘤周水肿区及对侧正常脑组织区的参数值。为了最小化个体差异,将各参数值除以对侧正常脑组织的值,得到各参数的相对值。采用卡方检验对两组患者的性别进行组间比较;采用两独立样本t检验及Mann-Whitney U检验对两组患者的年龄、MAP-MRI及DCE-MRI各参数值及其相对值进行组间比较,P<0.05为差异有统计学意义,并绘制受试者工作特征(receiver operating characteristic,ROC)曲线,采用DeLong检验以评估各参数值鉴别诊断的效能。结果GBM(IDH-wt)组与BMs组患者的年龄、性别差异均无统计学意义(P=0.327和P=0.247)。GBM(IDH-wt)组患者肿瘤实质区的非高斯轴向(non-Gaussianity axial,NGAx)、非高斯垂直(non-Gaussianity vertical,NGRad)、返回轴概率(return to the axis probability,RTAP)、返回平面概率(return to the plane probability,RTPP)均高于BMs组,均方位移(mean square displacement,MSD)低于BMs组,且差异有统计学意义(P<0.05)。GBM(IDH-wt)组患者瘤周水肿区的相对转运常数(relative volume transfer constant,rK^(trans))高于BMs,而相对渗出速率常数(relative the rate constant,rK_(ep))低于BMs组,且差异有统计学意义(P<0.05)。肿瘤实质区RTPP与NGAx是鉴别GBM(IDH-wt)与BMs时曲线下面积(area under the curve,AUC)较高的单一参数,AUC分别为0.985、0.937,敏感度分别为0.963、0.926,特异度分别为0.917、0.833。结论MAP-MRI与DCE-MRI在鉴别GBM(IDH-wt)与BMs时表现出了较好的诊断价值,且肿瘤实质区的RTPP与NGAx可作为较好的影像学标记。展开更多
Due to the lack of long-range association and spatial location information,fine details and accurate boundaries of complex clothing images cannot always be obtained by using the existing deep learning-based methods.Th...Due to the lack of long-range association and spatial location information,fine details and accurate boundaries of complex clothing images cannot always be obtained by using the existing deep learning-based methods.This paper presents a convolutional structure with multi-scale fusion to optimize the step of clothing feature extraction and a self-attention module to capture long-range association information.The structure enables the self-attention mechanism to directly participate in the process of information exchange through the down-scaling projection operation of the multi-scale framework.In addition,the improved self-attention module introduces the extraction of 2-dimensional relative position information to make up for its lack of ability to extract spatial position features from clothing images.The experimental results based on the colorful fashion parsing dataset(CFPD)show that the proposed network structure achieves 53.68%mean intersection over union(mIoU)and has better performance on the clothing parsing task.展开更多
文摘癫痫是一种慢性神经系统疾病,其分类广泛、机制复杂,具有反复发作和不可预测的特点,对患者的生活造成了一定的影响。深度了解其病理生理机制对疾病的治疗和提高患者生活质量十分重要,静息态功能磁共振成像(resting-state functional magnetic resonance imaging,rs-fMRI)现已成为探究癫痫脑功能改变的有效方法。目前基于rs-fMRI研究癫痫的数据分析方法主要有低频振幅(amplitude of low frequency fluctuation,ALFF)、局部一致性(regional homogeneity,ReHo)、功能连接(function connection,FC)和图论分析。本文将对rs-fMRI的各种分析方法在癫痫中的应用作一综述,为癫痫的病理生理机制及术前定位、治疗提供重要参考。
文摘目的探讨平均表观传播扩散MRI(mean apparent propagator-MRI,MAP-MRI)与动态对比增强MRI(dynamic contrast enhanced-MRI,DCE-MRI)在鉴别胶质母细胞瘤(glioblastoma,GBM)与脑转移瘤(brain metastases,BMs)中的临床应用价值。材料与方法对经手术病理确诊的GBM[异柠檬酸脱氢酶野生型(isocitrate dehydrogenase-wildtype,IDH-wt)]患者27例及经手术病理确诊或临床随访证实的BMs患者24例行常规MRI序列及扩散频谱成像(diffusion spectrum imaging,DSI)与DCE-MRI检查,DSI经解析得到MAP-MRI的各参数图,DCE-MRI经西门子工作站处理后得到多个参数图。在各参数图上分别测量两组患者肿瘤实质区、瘤周水肿区及对侧正常脑组织区的参数值。为了最小化个体差异,将各参数值除以对侧正常脑组织的值,得到各参数的相对值。采用卡方检验对两组患者的性别进行组间比较;采用两独立样本t检验及Mann-Whitney U检验对两组患者的年龄、MAP-MRI及DCE-MRI各参数值及其相对值进行组间比较,P<0.05为差异有统计学意义,并绘制受试者工作特征(receiver operating characteristic,ROC)曲线,采用DeLong检验以评估各参数值鉴别诊断的效能。结果GBM(IDH-wt)组与BMs组患者的年龄、性别差异均无统计学意义(P=0.327和P=0.247)。GBM(IDH-wt)组患者肿瘤实质区的非高斯轴向(non-Gaussianity axial,NGAx)、非高斯垂直(non-Gaussianity vertical,NGRad)、返回轴概率(return to the axis probability,RTAP)、返回平面概率(return to the plane probability,RTPP)均高于BMs组,均方位移(mean square displacement,MSD)低于BMs组,且差异有统计学意义(P<0.05)。GBM(IDH-wt)组患者瘤周水肿区的相对转运常数(relative volume transfer constant,rK^(trans))高于BMs,而相对渗出速率常数(relative the rate constant,rK_(ep))低于BMs组,且差异有统计学意义(P<0.05)。肿瘤实质区RTPP与NGAx是鉴别GBM(IDH-wt)与BMs时曲线下面积(area under the curve,AUC)较高的单一参数,AUC分别为0.985、0.937,敏感度分别为0.963、0.926,特异度分别为0.917、0.833。结论MAP-MRI与DCE-MRI在鉴别GBM(IDH-wt)与BMs时表现出了较好的诊断价值,且肿瘤实质区的RTPP与NGAx可作为较好的影像学标记。
文摘Due to the lack of long-range association and spatial location information,fine details and accurate boundaries of complex clothing images cannot always be obtained by using the existing deep learning-based methods.This paper presents a convolutional structure with multi-scale fusion to optimize the step of clothing feature extraction and a self-attention module to capture long-range association information.The structure enables the self-attention mechanism to directly participate in the process of information exchange through the down-scaling projection operation of the multi-scale framework.In addition,the improved self-attention module introduces the extraction of 2-dimensional relative position information to make up for its lack of ability to extract spatial position features from clothing images.The experimental results based on the colorful fashion parsing dataset(CFPD)show that the proposed network structure achieves 53.68%mean intersection over union(mIoU)and has better performance on the clothing parsing task.