基于纹理分析的带标记线心脏MR图像分割

背景：左心室边界的准确分割是对左心室运动及形变进行分析的前提。由于受带标记线心脏核磁共振图像中标记线强梯度的影响,对左心室内膜的提取变得非常困难。目的：为了抑制标记线对图像分割的影响,提出了一种基于最小值-方差能量图的纹理分析方法。方法：首先对局部最小值和方差进行加权求和,得到能量图;然后利用中值滤波滤除能量图中的伪影并保持边界;最后,应用GVF-snake模型提取左心室内膜。结果与结论：针对标记线在心脏MR图像中的分布特征,提出了一种基于最小值-方差的纹理分析方法,该方法有效地去除了标记线。结果提示,对使用该纹理分析方法生成的能量图应用GVF-snake模型可以较好地提取左心室内膜。     

基于纹理分析和双边滤波方法对加标记心脏核磁共振图像分割

背景:加标记心脏核磁共振成像方式提供了左心室内外心膜的边缘信息,该边缘信息可由分割图像得到.但是,所引入的标记线加大了这类图像边界分割的凼难.目的:针对目前在加标记心脏核磁共振图像中对左心室分割困难的问题,提出了一种新的自动分割的方法.方法;首先,使用全局直方图规定化方法增强标记和非标记区域的对比度;然后,利用一种简单的纹理分析方法区分血流充盈的心腔(非纹理)区域和加标记心肌(纹理)区域;再应用双边滤波在保持边界的同时滤掉图像的伪影;最后,用GVF-snake模型自动提取左心室图像的边界.结果与结论:提出了一种简单的纹理分析方法来移除标记线:用局部窗口中的最大灰度值与最小灰度值之差来代替原象素点灰度值,再运用双边滤波滤除图像伪影并保持边界,最后应用 GVF-snake模型实现了左心室边界的有效提取.实验结果显示,该方法能够较好地提取部分加标记心脏核磁共振图像中血流充盈区的边界.     

采用纹理分析特征用于肝脏超声图像的病患的辨识

采用纹理分析特征,定量分析肝脏图像,将一般的矢量“距离”判决——多维空间的判决转化为标量距离,即一维空间距离进行判决辨识分类肝脏病患。对正常肝、慢性肝炎和肝硬化图像辨识分类,正确率可达90%,为肝脏病患辨识分类的专家系统提供了一个有参考价值的分类方法。     

基于T2WI图像纹理分析评价大鼠心肺复苏后脑损伤初步研究

目的:探讨基于T2WI图像纹理分析评价大鼠心肺复苏（CPR）后脑损伤的可行性。方法:将18只SD大鼠随机（随机数字法）分为假手术组（ n=8）及模型组（ n=10）。模型组大鼠恢复自主循环（ROSC）6 h后行T2WI序列扫描,假手术组于术后6 h扫描。比较两组大鼠基于T2WI图像的纹理特征及AQP4及和NSE抗体免疫组织化学评分的差表达异。运用受试者工作特征曲线(ROC)评价两组间差异的纹理特征对脑损伤的诊断效能。运用Spearman等级相关系数分析两组间差异的纹理特征与免疫组织化学AQP4及NSE抗体表达之间的相关性。 结果:两组的最小强度值、标准偏差、差分熵、逆差矩及高灰度不均匀性特征的差异具有统计学意义（ P<0.05）模型组大鼠全脑T2WI纹理特征的最小强度值、标准偏差及逆差矩明显低于假手术组（ P<0.05）,而差分熵及高灰度不均匀性显著高于假手术组（ P<0.05）。差分熵诊断效能最优,ROC曲线下面积（area under curve, AUC）为0.922,敏感度100%及特异度75%。模型组AQP4及NSE表达免疫组织化学（AQP4及NSE）评分明显著高于假手术组（ P<0.05）。最小强度值与AQP4及NSE抗体表达均呈正相关( r=0.501、0.568, P=0.048、0.022);标准偏差与AQP4及NSE抗体表达均呈正相关( r=0.620、0.530, P=0.010、0.035);差分熵与AQP4抗体表达呈负相关( r=-0.535, P=0.033)。 结论:基于T2WI图像的纹理分析可以评估脑水肿及神经元受损情况程度,最小强度值、标准偏差、和差分熵是反映评估心脏骤停CPR后脑损伤程度的敏感指标,其中差分熵敏感度和特异度最高。     

乳腺X线图像纹理特征预测乳腺癌腋窝淋巴结转移

目的 探讨乳腺X线图像纹理特征预测乳腺癌腋窝淋巴结转移的价值。方法 回顾性分析171例病理证实为非特殊类型浸润性乳腺癌患者的X线及临床资料。所有患者均接受腋窝淋巴结清扫手术,并根据手术及病理结果将患者分为腋窝淋巴结转移组和非转移组。分析患者的X线头尾位图像纹理特征,应用灰度直方图及灰度共生矩阵纹理分析方法测定均值、标准偏差、偏度、峰度、方差、能量、熵、自相关、惯量、逆差距和反差等11个纹理参数。结果 171例患者中淋巴结转移组96例,非转移组75例。X线检出腋窝淋巴结阴性119例,阳性52例,其诊断腋窝淋巴结转移的敏感度和特异度分别为48.96%（47/96）和93.33%（70/75）。X线图像纹理分析结果显示腋窝淋巴结转移组能量、熵、逆差距、自相关值均高于非转移组,惯量、反差值均低于非转移组（P均< 0.05）;其余纹理特征参数值两组间差异无统计学意义（P均> 0.05）。纹理参数值能量、熵、惯量、逆差距、自相关和反差诊断腋窝淋巴结转移的ROC曲线下面积（AUC）值分别为0.610、0.610、0.374、0.599、0.612和0.421（P均< 0.05）。乳腺X线检查、纹理特征及X线联合纹理特征诊断腋窝淋巴结转移的AUC值分别为0.711、0.676和0.787（P均< 0.05）;纹理特征、乳腺X线联合纹理特征诊断腋窝淋巴结转移的敏感度分别为62.5%和64.6%,特异度分别为66.7%和82.7%。结论 乳腺X线图像纹理参数对腋窝淋巴结转移有一定的预测作用,且乳腺X线联合纹理特征可提高对腋窝淋巴结转移的诊断效能。     

基于T2WI图像的纹理分析对鼻咽癌与鼻咽部炎性增生性疾病的鉴别诊断价值

目的:探讨磁共振T2WI图像纹理分析鉴别鼻咽癌(nasopharyngeal carcinoma,NPC)与鼻咽部炎性增生(nasopharyngeal hyperplasia,NPH)的价值。方法:收集经手术及病理学检查证实为NPC和NPH的患者共48例,均经过磁共振成像(magnetic resonance imaging,MRI)检查,其中NPC 30例,NPH 18例。应用MaZda软件对两组病灶进行纹理分析,在T2WI图像上选取病灶范围最大层面勾画感兴趣区(region of interest,ROI),测得294组参数,进行统计学分析。最后,利用多参数联合鉴别NPC和NPH。结果:两组间S(0,5)熵、S(5,5)熵、S(5,-5)熵、45°方向游程长不均匀度、135°方向游程长不均匀度、梯度偏度及小波低高频转换系数s-3等7组参数差异有统计学意义(P0.05),其中135°方向游程长不均匀度、梯度偏度及小波低高频转换系数s-3这3组参数曲线下面积(area under curve,AUC)分别为0.759、0.803及0.731,诊断准确率中等。另外,分析发现,135°方向游程长不均匀度联合梯度偏度、梯度偏度联合小波低高频转换系数s-3鉴别两组病变的诊断效能分别为0.824和0.833,诊断准确率较单一参数高。结论:基于磁共振T2WI序列的纹理分析可用于鉴别NPC和NPH性疾病。     

基于MR图像三维纹理特征的阿尔茨海默病和轻度认知障碍的分类

目的 利用三维纹理特征对阿尔茨海默病(AD)患者和轻度认知障碍(MCI)患者进行分类识别,以探索AD早期诊断新途径。方法 对12例早期AD患者(AD组)、12例MCI患者(MCI组)及12名健康对照者(NC组)的MR图像进行三维纹理分析,采用灰度共生矩阵和游程长矩阵提取每位受试者左、右侧海马结构及胼胝体的三维纹理特征,选取三组间存在显著性差异的纹理参数作为特征变量,采用支持向量机(SVM)方法对各组进行分类,利用留一法估算分类准确率。结果 对NC组与MCI组、MCI组与AD组、NC组与AD组进行分类识别的最高准确率分别为79.17%、83.33%、91.67%。结论 利用三维纹理分析可分类识别早期AD患者及MCI患者,有助于AD的早期诊断。     

基于多重分形的医学图像分割方法

背景:由于人体解剖结构的复杂性、组织器官形状的不规则性及不同个体间的差异性,所以比较适合用多重分形来分析.目的:采用多重分形理论对医学图像进行图像分割.方法:采用基于容量测度的多重分形谱计算及基于概率测度的多重分形谱计算方法对图像进行分割.对于待处理图片分别进行传统的区域生长分割,max容量测度图像分割,sum容量测度图像分割,概率测度图像分割等4种分割,并加入噪声后再进行同样的分割处理作为比较.结果与结论:采用的两种基于多重分形谱的计算法中,基于容量测量的多重分形谱计算方法的关键是定义合适的测度μα;基于概率测度的多重分形谱计算方法的关键是定义合适的归一化概率Pi,不同的测度(概率)和不同的阈值对结果的影像比较大.基于概率测度的方法对噪声比较敏感,但是在滤过噪声时对图像象素大小变化比较大、比较复杂的图像有较好的分割效果.实验表明基于多重分形谱的医学图像分割方法在选择合适的测度(概率)和阈值时是可行的,特别是在较为复杂的图像处理中对于纹理和边缘的区别上有较大的优势,在准确地分割的同时能保留更多的细节,具有重要的实际意义.同时,多重分形也可以作为一种图像的特征,为特征提取多提供一种有力的数据.     

B超图像纹理分析参数的研究

Ｂ超图像纹理分析参数的研究连娟①周康源②姚文俊胡耀辉罗福成①１Ｂ超图像的定量分析Ｂ超图像的定理分析包括组织的定量分析与图像的定量分析。组织的定量分析应当与Ｂ超仪器本身无关，而图像的定量分析则不仅取决于组织的特性，还取决于Ｂ超仪器及其设置。目前常用的体...     

MR动态增强图像纹理分析判断乳腺结节良恶性的价值

目的 探讨MR动态增强图像纹理分析鉴别诊断乳腺结节良恶性的价值。方法 回顾性分析经手术病理证实的78例患者共80个乳腺结节的MR动态增强图像,每个结节获得63个纹理特征参数。绘制纹理参数鉴别诊断良恶性乳腺结节的ROC曲线,并与MR乳腺影像报告和数据系统（BI-RADS）的诊断效能比较。结果 78例患者的80个乳腺结节中,纹理参数中灰度游程长不均匀度判断乳腺结节良恶性的AUC值（0.836）最大且诊断准确率高,其诊断恶性乳腺结节的敏感度为82.93%（34/41）、特异度为94.87%（37/39）、准确率为88.75%（71/80）、阳性预测值为94.44%（34/36）、阴性预测值为84.09%（37/44）。MR BI-RADS分类诊断恶性乳腺结节的敏感度为95.12%（39/41）、特异度为87.18%（34/39）、准确率为91.25%（73/80）、阳性预测值为88.63%（39/44）、阴性预测值为94.44%（34/36）。MR BI-RADS分类和纹理分析判断恶性乳腺结节准确率差异无统计学意义（P=0.11）。与单独应用BI-RADS分类比较,两者联合应用可明显提高诊断恶性乳腺结节的特异度（P<0.001）。结论 MR纹理分析可作为传统诊断乳腺良恶性结节的补充。     

Towards ultrasound cardiac image segmentation based on the radiofrequency signal

In echocardiography, the radio-frequency (RF) image is a rich source of information about the investigated tissues. Nevertheless, very few works are dedicated to boundary detection based on the RF image, as opposed to envelope image. In this paper, we investigate the feasibility and limitations of boundary detection in echocardiographic images based on the RF signal. We introduce two types of RF-derived parameters: spectral autoregressive parameters and velocity-based parameters, and we propose a discontinuity adaptive framework to perform the detection task. In classical echographic cardiac acquisitions, we show that it is possible to use the spectral contents for boundary detection, and that improvement can be expected with respect to traditional methods. Using the system approach, we study on simulations how the spectral contents can be used for boundary detection. We subsequently perform boundary detection in high frame rate simulated and in vivo cardiac sequences using the variance of velocity, obtaining very promising results. Our work opens the perspective of a RF-based framework for ultrasound cardiac image segmentation and tracking.     

基于图像分割的Tamura纹理特征算法的研究与实现

背景：ITK主要提供医学图像处理、分割与配准算法,但其缺少可视化的功能,缺乏灵活实用的用户界面,VTK提供了丰富的医学影像处理与分析工具,具有强大的图形处理和可视化功能。目的：利用以前的确诊病例和医生的诊断经验以及患者的相关病史,对确诊的医学影像资源进行管理,归档,并检索,以减少人工干预,提高图像的查全率和查准率。方法：以视觉感知机制为基础,在ITK平台上进行图像平滑去噪和分割的预处理过程,利用Tamura算法完成纹理特征提取,最后通过实验采集、计算数据,完成对比分析。结果与结论：基于图像分割的Tamura纹理特征算法在基于图像纹理检索应用上便于相似性度量,进而可提高检索的准确率。     

Muscles of mastication model-based MR image segmentation

Objective The muscles of mastication play a major role in the orodigestive system as the principal motive force for the mandible. An algorithm for segmenting these muscles from magnetic resonance (MR) images was developed and tested. Materials and methods Anatomical information about the muscles of mastication in MR images is used to obtain the spatial relationships relating the muscle region of interest (ROI) and head ROI. A model-based technique that involves the spatial relationships between head and muscle ROIs as well as muscle templates is developed. In the segmentation stage, the muscle ROI is derived from the model. Within the muscle ROI, anisotropic diffusion is applied to smooth the texture, followed by thresholding to exclude bone and fat. The muscle template and morphological operators are employed to obtain an initial estimate of the muscle boundary, which then serves as the input contour to the gradient vector flow snake that iterates to the final segmentation. Results The method was applied to segmentation of the masseter, lateral pterygoid and medial pterygoid in 75 images. The overlap indices (κ) achieved are 91.4, 92.1 and 91.2%, respectively. Conclusion A model-based method for segmenting the muscles of mastication from MR images was developed and tested. The results show good agreement between manual and automatic segmentations.     

Enhancing MR image segmentation with realistic adversarial data augmentation

The success of neural networks on medical image segmentation tasks typically relies on large labeled datasets for model training. However, acquiring and manually labeling a large medical image set is resource-intensive, expensive, and sometimes impractical due to data sharing and privacy issues. To address this challenge, we propose AdvChain, a generic adversarial data augmentation framework, aiming at improving both the diversity and effectiveness of training data for medical image segmentation tasks. AdvChain augments data with dynamic data augmentation, generating randomly chained photo-metric and geometric transformations to resemble realistic yet challenging imaging variations to expand training data. By jointly optimizing the data augmentation model and a segmentation network during training, challenging examples are generated to enhance network generalizability for the downstream task. The proposed adversarial data augmentation does not rely on generative networks and can be used as a plug-in module in general segmentation networks. It is computationally efficient and applicable for both low-shot supervised and semi-supervised learning. We analyze and evaluate the method on two MR image segmentation tasks: cardiac segmentation and prostate segmentation with limited labeled data. Results show that the proposed approach can alleviate the need for labeled data while improving model generalization ability, indicating its practical value in medical imaging applications.     

Multi-level semantic adaptation for few-shot segmentation on cardiac image sequences

Obtaining manual labels is time-consuming and labor-intensive on cardiac image sequences. Few-shot segmentation can utilize limited labels to learn new tasks. However, it suffers from two challenges: spatial-temporal distribution bias and long-term information bias. These challenges derive from the impact of the time dimension on cardiac image sequences, resulting in serious over-adaptation. In this paper, we propose the multi-level semantic adaptation (MSA) for few-shot segmentation on cardiac image sequences. The MSA addresses the two biases by exploring the domain adaptation and the weight adaptation on the semantic features in multiple levels, including sequence-level, frame-level, and pixel-level. First, the MSA proposes the dual-level feature adjustment for domain adaptation in spatial and temporal directions. This adjustment explicitly aligns the frame-level feature and the sequence-level feature to improve the model adaptation on diverse modalities. Second, the MSA explores the hierarchical attention metric for weight adaptation in the frame-level feature and the pixel-level feature. This metric focuses on the similar frame and the target region to promote the model discrimination on the border features. The extensive experiments demonstrate that our MSA is effective in few-shot segmentation on cardiac image sequences with three modalities, i.e. MR, CT, and Echo (e.g. the average Dice is 0.9243), as well as superior to the ten state-of-the-art methods.