目的 分析总结神经束膜瘤超声图像特征.方法 回顾性分析13例神经束膜瘤患者的超声图像特征.结果 神经内神经束膜瘤超声图像为神经"纺锤状"增大,跨度较长,肿瘤内部维持了神经结构,长轴及横断面均可见低回声的神经束增粗明显;软组织神经束膜瘤超声图像表现为软组织内椭圆形低回声区,形态规则,边界清晰,内部回声均质.结论 神经内神经束膜瘤的超声图像有一定特征性,能为临床诊断指明方向;软组织神经束膜瘤超声图像表现没有特异度.

目的:分析臂丛神经下干受压型胸廓出口综合征的超声声像图特征,探讨超声在诊断此类疾病中的价值.方法:回顾分析2012年5月至2020年5月之间于我院进行超声诊断并经手术证实为臂丛神经下干受压型胸廓出口综合征的23例患者.检查仪器为PHILIPS EPIQ5及iU22彩色多普勒超声诊断仪.以术中所见为金标准,归纳臂丛神经下干受压型胸廓出口综合征的超声声像图特征,分析超声诊断此类疾病的直接征象与间接征象.结果:臂丛神经下干受压型胸廓出口综合征超声声像图的直接征象为:颈8与胸1神经根或/与下干受压,19例,阳性率为82.6％.间接征象为:①锁骨下动脉上抬,18例,阳性率为78.3％;②异常突起的骨性结构,9例,阳性率为39.1％;③颈8或胸1神经根或/与下干增粗,且均未见直接受压征象,3例,阳性率为13.0％.结论:超声观察到颈8与胸1神经根或/与下干受压的直接征象对诊断臂丛神经下干受压型胸廓出口综合征具有较强的提示作用.间接征象如锁骨下动脉上抬、异常突起的骨性结构、颈8或胸1神经根或/与下干增粗有助于此类疾病的超声诊断.

超声成像作为常见的医学成像技术,在诊断和治疗疾病方面发挥着重要作用.人工判读超声图像依赖于医生主观经验知识,其结果存在观察者间和观察者内的差异,并且过程耗时耗力,难以满足快速、批量的临床诊断需求.随着人工智能的快速发展,基于深度学习的图像分割模型被广泛地应用于医学图像分割领域,并取得了不俗成果.基于U-Net深度神经网络,对其结构进行优化改进,构建适用于超声图像正中神经分割的卷积神经网络模型.在网络中添加了规范层和丢弃层,下采样的最大池化层替换成卷积层;同时,采用指数线性单元激活函数,使用循环学习率,以增加网络模型的鲁棒性.测试结果表明,改进的U-Net模型在超声正中神经图像的自动分割方面表现良好,横切面、纵切面的Dice系数分别达到了78％与89％.该方法有望用于临床超声神经图像的辅助分析.

目的:运用改进U-Net深度网络学习定量评价腕管综合征正中神经的超声图像,确定基于改进U-Net深度网络学习的卷积神经网络模型,探讨其在定量评估腕管综合征正中神经卡压中的应用价值.方法:搜集213例经肌电图确诊的腕管综合征正中神经卡压患者及104例健康志愿者,213例正中神经卡压患者中60例为双侧卡压.对317例受检者行超声检查,在腕管处保存超声图像,共得到正中神经图像377组.由擅长肌骨超声的医师对377组图像进行勾勒.应用基于改进U-net深度网络学习的卷积神经网络模型,分割腕管综合征卡压的正中神经超声图像,定量分析提取横切以及纵切的正中神经超声图像的影像组学量化特征.结果:改进的U-Net深度网络可以很好地识别切割正中神经;改进的U-Net深度网络可以定量表示CTS中卡压的正中神经回声减低,区域明暗参数A、明暗参数I、对比明暗参数RI以及纹理参数Homo、纹理不均匀参数Cont差异均有统计学意义(P=0.000).结论:改进的U-Net模型在超声正中神经图像自动分割方面表现良好,可以定量分析腕管综合征正中神经卡压时灰度以及神经纹理均匀性.

Purpose: To assess the clinical value of high-frequency ultrasonography in the diagnosis of traumatic brachial plexus injury. Methods: One hundred and forty-eight patients with suspected brachial plexus injury after trauma underwent high-frequency ultrasonography. The ultrasonographic findings were compared with operative findings. Results: The accuracy of ultrasonography in diagnosis of brachial plexus injury was 79.1％ and in determining the localization and type of injury was 93.9％ and 81.1％, respectively. Conclusion: High-frequency ultrasonography is useful in diagnosing traumatic brachial plexus injury and determining the localization and type of injury.

Objective The aim of this study was to determine the value of Doppler sonography in the evaluation of carpal tunnel syndrome (CTS).Methods A total of 71 patients with a clinical suspicion of UNE and 47 healthy controls were examined by Doppler and gray-scale sonography.Results Intraneural vascularization was found in 65 (91.5％) of 71 patients with CTS and in 25 (53.2％) of 47 healthy controls.The peak systolic velocity (PSV) of patients was higher than it of healthy volunteers.Combined Doppler and gray scale sonography,the sensitivity was raised to 100％.The prevalence of intraneural vascularization and Doppler scores of patients with moderate and severe CTS were higher than those with mild CTS and normal electrodiagnostic testing.Conclusions Doppler sonography is helpful to diagnose CTS and determine the treatment.

Objective To assess the value of high-resolution ultrasonography in diagnosis of cubital tunnel syndrome.Methods Forty-two patients(43 elbows) initially diagnosed as cubital tunnel syndrome underwent ultrasonography(US),while 15 healthy contralateral elbows of these patients taken as self controls,and 15 healthy volunteers as normal controls underwent US of the ulnar nerve.The findings of US measurements were compared with that of intra-operative results and pre-operative electromyography.Results High-resolution US displayed changes and some etiological factors of cubital tunnel syndrome.The measurements of ulnar nerve at the proximal part of the compression were higher than those in the control groups.The cut-off point of cross-sectional area(CSA) and CSA tumefaction ratio was 0.11 cm2 and 141.50%,respectively.The sensitivity of US was 92.86% compared with intra-operative results,and was 100% when combined with pre-operative electromyography.Conclusion High-frequency ultrasound can be used as an effective method to diagnose cubital tunnel syndrome.