Applying artificial intelligence to the identification ofvariegated coloring in skin tumors

The importance of color information for the automatic diagnosis of skin tumors by computer vision is demonstrated. The utility of the relative color concept is proved by the results in identifying variegated coloring. A feature file paradigm is shown to provide an effective methodology for the independent development of software modules for expert system/computer vision research. An automatic induction tool is used effectively to generate rules for identifying variegated coloring. Variegated coloring can be identified at rates as high as 92% when using the automatic induction technique in conjunction with the color segmentation method.     

Automated assessment of thigh composition using machine learning for Dixon magnetic resonance images

Objectives

To develop and validate a machine learning based automated segmentation method that jointly analyzes the four contrasts provided by Dixon MRI technique for improved thigh composition segmentation accuracy.

Materials and methods

The automatic detection of body composition is formulized as a three-class classification issue. Each image voxel in the training dataset is assigned with a correct label. A voxel classifier is trained and subsequently used to predict unseen data. Morphological operations are finally applied to generate volumetric segmented images for different structures. We applied this algorithm on datasets of (1) four contrast images, (2) water and fat images, and (3) unsuppressed images acquired from 190 subjects.

Results

The proposed method using four contrasts achieved most accurate and robust segmentation compared to the use of combined fat and water images and the use of unsuppressed image, average Dice coefficients of 0.94 ± 0.03, 0.96 ± 0.03, 0.80 ± 0.03, and 0.97 ± 0.01 has been achieved to bone region, subcutaneous adipose tissue (SAT), inter-muscular adipose tissue (IMAT), and muscle respectively.

Conclusion

Our proposed method based on machine learning produces accurate tissue quantification and showed an effective use of large information provided by the four contrast images from Dixon MRI.

     

焦炭显微光学组织自动检测与提取方法研究

焦炭显微光学组织结构测定是一种重要的焦炭质量评价方式,针对焦炭显微图像边缘模糊对比度低且存在光晕伪影等问题,设计一种基于语义分割和全连接条件随机场的焦炭显微光学组织自动检测与提取方法。首先,利用偏光显微镜、工业相机和上位机等搭建焦炭显微光学组织测定平台;其次,利用残差模块和注意力模块改进Unet网络模型,加强显微光学组织区域的输出权重,实现对焦炭光学组织的自动检测与分割;最后,使用全连接条件随机场对显微光学组织的空间特性进行建模,细化分割边缘,精确提取焦炭显微光学组织。实验结果表明,所提方法的精确度、召回率、F1分数和准确率分别达到了0.967、0.959、0.963、0.965,优于其他对比语义分割网络,证明该方法具有较高的分割性能,能够实现对焦炭显微光学组织的自动检测与提取。     

基于匹配滤波和自动阈值的眼底血管分割方法

提出一种快速、简便、高效的眼底血管分割方法。分析眼底图像的灰度值分布和对比度变化,利用匹配滤波克服背景干扰,消除噪声影响,达到灰度均衡,实现眼底图像的亮度归一化。估计眼底图像中背景像素所占比例,利用直方图自动选择阈值,完成对眼底图像中血管的有效分割。在公开的眼底图像数据库上进行测试,该方法对眼底血管分割具有较好的性能指标。实验表明,提出的基于匹配滤波和阈值优化的眼底血管分割方法,准确率高、复杂度低,对眼科疾病的计算机辅助诊断有一定的实用价值。     

Volume measurements of individual muscles in human <Emphasis Type="Italic">quadriceps femoris</Emphasis> using atlas-based segmentation approaches

Objectives

Atlas-based segmentation is a powerful method for automatic structural segmentation of several sub-structures in many organs. However, such an approach has been very scarcely used in the context of muscle segmentation, and so far no study has assessed such a method for the automatic delineation of individual muscles of the quadriceps femoris (QF). In the present study, we have evaluated a fully automated multi-atlas method and a semi-automated single-atlas method for the segmentation and volume quantification of the four muscles of the QF and for the QF as a whole.

Subjects and methods

The study was conducted in 32 young healthy males, using high-resolution magnetic resonance images (MRI) of the thigh. The multi-atlas-based segmentation method was conducted in 25 subjects. Different non-linear registration approaches based on free-form deformable (FFD) and symmetric diffeomorphic normalization algorithms (SyN) were assessed. Optimal parameters of two fusion methods, i.e., STAPLE and STEPS, were determined on the basis of the highest Dice similarity index (DSI) considering manual segmentation (MSeg) as the ground truth. Validation and reproducibility of this pipeline were determined using another MRI dataset recorded in seven healthy male subjects on the basis of additional metrics such as the muscle volume similarity values, intraclass coefficient, and coefficient of variation. Both non-linear registration methods (FFD and SyN) were also evaluated as part of a single-atlas strategy in order to assess longitudinal muscle volume measurements. The multi- and the single-atlas approaches were compared for the segmentation and the volume quantification of the four muscles of the QF and for the QF as a whole.

Results

Considering each muscle of the QF, the DSI of the multi-atlas-based approach was high 0.87 ± 0.11 and the best results were obtained with the combination of two deformation fields resulting from the SyN registration method and the STEPS fusion algorithm. The optimal variables for FFD and SyN registration methods were four templates and a kernel standard deviation ranging between 5 and 8. The segmentation process using a single-atlas-based method was more robust with DSI values higher than 0.9. From the vantage of muscle volume measurements, the multi-atlas-based strategy provided acceptable results regarding the QF muscle as a whole but highly variable results regarding individual muscle. On the contrary, the performance of the single-atlas-based pipeline for individual muscles was highly comparable to the MSeg, thereby indicating that this method would be adequate for longitudinal tracking of muscle volume changes in healthy subjects.

Conclusion

In the present study, we demonstrated that both multi-atlas and single-atlas approaches were relevant for the segmentation of individual muscles of the QF in healthy subjects. Considering muscle volume measurements, the single-atlas method provided promising perspectives regarding longitudinal quantification of individual muscle volumes.

     

Multi-atlas-based fully automatic segmentation of individual muscles in rat leg

Objective

To quantify individual muscle volume in rat leg MR images using a fully automatic multi-atlas-based segmentation method.

Materials and methods

We optimized a multi-atlas-based segmentation method to take into account the voxel anisotropy of numbers of MRI acquisition protocols. We mainly tested an image upsampling process along Z and a constraint on the nonlinear deformation in the XY plane. We also evaluated a weighted vote procedure and an original implementation of an artificial atlas addition. Using this approach, we measured gastrocnemius and plantaris muscle volumes and compared the results with manual segmentation. The method reliability for volume quantification was evaluated using the relative overlap index.

Results

The most accurate segmentation was obtained using a nonlinear registration constrained in the XY plane by zeroing the Z component of the displacement and a weighted vote procedure for both muscles regardless of the number of atlases. The performance of the automatic segmentation and the corresponding volume quantification outperformed the interoperator variability using a minimum of three original atlases.

Conclusion

We demonstrated the reliability of a multi-atlas segmentation approach for the automatic segmentation and volume quantification of individual muscles in rat leg and found that constraining the registration in plane significantly improved the results.

     

基于改进分水岭算法的菌落图像分割

随着计算机图像处理技术的快速发展,在菌落筛选环节上,传统的人工手动筛选已经逐渐被自动筛选仪器所替代。自动筛选仪器的核心是菌落图像识别模块,而菌落图像识别的关键在于图像分割技术,本文提出了一种改进的分水岭分割算法。该算法首先采用高斯滤波去除图像噪声,再对去噪后的图像进行形态学处理,然后进行倒角距离变换得到菌落距离图像,再采用形态学方法填补其空洞信息,接着对标记后的区域进行分水岭分割,最后利用区域合并算法聚集图像相似区域,从而得到最终的分割图像。采用本文提出的改进型分水岭算法进行菌落图像分割的准确率为93.4%,而传统的分水岭算法的分割准确率为75%,通过与传统分水岭对比实验的结果可以看出,改进后的算法较好地抑制了传统分水岭的过分割现象,极大提高了识别精度。     

复杂城市交通场景下的自动驾驶语义分割方法

多尺度特征金字塔可以缓解语义分割在复杂交通场景下通常存在漏分割、错误分割、边界分割不清晰等问题,但现有的多尺度特征金字塔在获取丰富的语义信息时,不得不下采样特征图,牺牲空间细节信息,而这导致了最终的分割结果仍然精度受限。针对该问题,本文提出了特征强化模块,使得有利于正确分类像素的相似特征在下采样过程之前得到基于不同矢量间余弦相似度的进一步加强,降低下采样带来的负面影响。另外,结合空洞卷积和条带卷积原理,本文对大卷积核进行了改造,并构建新的多尺度特征金字塔模块,以获取尺度不同且具备更大感受野的语义信息。该分割方法实时高效,能够满足自动驾驶语义分割要求,在VOC2012数据集上的实验表明,所提出的方法分割结果mIoU达到了74.36%,FPS达到了43,优于目前的主流语义分割方法。     

OFPTB—550型SF_6断路器罐体端部电场的有限元计算

采用有限元法,对OFPTB—550型SF_6断路器罐体端部的三维电场进行了数值计算,针对计算区域的结构特点,在三维电场的计算中把子块结构、等参坐标变换、坐标的平移和旋转、自动剖分和手剖有机地结合在一起。给出了一种复杂结构中三维电场的计算方法。通过对计算结果的分析,确定了最大电场强度值的位置,对断路器的绝缘强度进行了评估。     

M a t l a b与 C#. n e t 混合编程的电缆优化分割研究

针对电力工程电缆优化分割下料问题,建立了以电缆原材料总长最小为优化目标,采用Matlab与C#.net混合编程实现该优化的数学模型。在此基础上研制开发了基于B/S模式的电缆优化分割下料系统,包括自动生成可行下料方案、选定最优分割方案和计算原材料利用率等功能。最后通过实例分析与测试,表明该系统能够较好地实现电力电缆优化分割下料,也验证了Matlab和C#.net混合编程的特点和优势。     

基于嵌入式平台和FPGA的自动抄表系统

提出用嵌入式平台和FPGA实现自动抄表的一种方案,介绍了各个模块的设计方法。FPGA用于功率的测量,嵌入式平台UP-NETARM300显示并处理FPGA测量的数据或将数据送到局域网上,还用FPGA设计了串口通信模块,并实现了与嵌入式平台的通信。本方案主要解决的是电能计量自动抄表系统中前端采集子系统的设计,FPGA功能模块实现的是抄表式电度表的功能,嵌入式平台相当于采集子系统中的电量集中器。通过采用嵌入式开放设计技术,很好地克服了以往自动抄表系统的开放性及扩展性差的缺点。     

Automatic segmentation and plaque characterization in atherosclerotic carotid artery MR images

In vivo MRI provides a means to non-invasively image and assess the morphological features of atherosclerotic carotid arteries. To assess quantitatively the degree of vulnerability and the type of plaque, the contours of the lumen, outer boundary of the vessel wall and plaque components, need to be traced. Currently this is done manually, which is time-consuming and sensitive to inter- and intra-observer variability. The goal of this work was to develop an automated contour detection technique for tracing the lumen, outer boundary and plaque contours in carotid MR short-axis black-blood images. Seventeen patients with carotid atherosclerosis were imaged using high-resolution in vivo MRI, generating a total of 50 PD- and T1-weighted MR images. These images were automatically segmented using the algorithm presented in this work, which combines model-based segmentation and fuzzy clustering to detect the vessel wall, lumen and lipid core boundaries. The results demonstrate excellent correspondence between automatic and manual area measurements for lumen (r=0.92) and outer (r=0.91), and acceptable correspondence for fibrous cap thickness (r=0.71). Though further optimization is required, our algorithm is a powerful tool for automatic detection of lumen and outer boundaries, and characterization of plaque in atherosclerotic vessels.     

Right ventricular strain and volume analyses through deep learning-based fully automatic segmentation based on radial long-axis reconstruction of short-axis cine magnetic resonance images

Magnetic Resonance Materials in Physics, Biology and Medicine - We propose a deep learning-based fully automatic right ventricle (RV) segmentation technique that targets radially reconstructed...     

CSELT隧道代理的实现

研究隧道代理技术在IPv4向IPv6过渡阶段的应用。阐述了隧道代理系统的构成,详细介绍了在一个内部网络中CSELT隧道代理系统的实现方法。此系统完成了自动隧道的配置和管理。实验结果表明,本隧道代理技术相比手动的隧道技术提高了操作效率。     

Sources of variation in multi-centre brain MTR histogram studies: body-coil transmission eliminates inter-centre differences

Object: 1. Identify sources of variation affecting Magnetisation Transfer Ratio (MTR) histogram reproducibility between-centres. 2. Demonstrate complete elimination of inter-centre difference. Materials and methods: Six principle sources of variation were summarised and analysed. These are:the imager coil used for radiofrequency (RF) transmission, imager stability, the shape and other parameters describing the Magnetisation Transfer (MT) pulse, the MT sequence used (including its parameters), the image segmentation methodology, and the histogram generation technique. Transmit field nonuniformity and B1 errors are often the largest factors. PLUMB (Peak Location Uniformity in MTR histograms of the Brain) plots are a convenient way of visualising differences. Five multi-centres studies were undertaken to investigate and minimise differences. Results: Transmission using a body coil, with a close-fitting array of surface coils for reception, gave the best uniformity. Differences between two centres, having MR imagers from different manufacturers, were completely eliminated by using body coil excitation, making a small adjustment to the MT pulse flip angle, and carrying out segmentation at a single centre. Histograms and their peak location and height values were indistinguishable. Conclusions: Body coil excitation is preferred for multi-centre studies. Analysis (segmentation and histogram generation) should ideally be carried out at a single site.     

基于组态软件的棚车车体自动检测系统

详细介绍了棚车车体自动检测装置的结构和工作原理,阐述了自动检测系统的软、硬件设计。通过超声波传感器和涡电流传感器检测车体的相关尺寸,并将实时数据上传给上位机进行计算、分析后与基准数据进行对比,再根据比较结果对车体进行整改,从而确保产品质量。采用车体自动检测装置后,不仅提高了工作效率,保证了产品的质量,而且提升了检测手段,实现了棚车车体检测的规范化和标准化。     

燃料电池极化曲线分段辨识算法

基于区间分段思想,将极化曲线非线性参数辨识转化为两个线性最小二乘辨识子问题.MATLAB仿真表明,拟合曲线与模型数据及实验数据的平均二次误差,分别为开路电压的0.31%和0.75%,精度与牛顿法相当.     

一种新的视频运动对象分割技术

本文将用于自然静态图像抠图领域的Knockout技术用于视频运动对象的精确分割,给出了一种新的视频运动对象分割方法。在利用积累差异技术自适应建立背景模型、采用背景差法初步提取当前帧视频对象区域的基础上,用本文提出的方法,自动标记原本采用Knockout技术进行抠图时,需手动进行标记的前景轮廓区域、背景轮廓区域及未知区域。在用本文方法完成对3个区域的自动标记后,再利用Knockout技术对当前帧的视频运动对象进行精确分割。本文还设计了一种新的变系数空域滤波器,该滤波器能有效地对背景差图像进行显著增强。同时,对Otsu自适应阈值化方法进行了改进,改进的方法能更准确地对背景差图像进行阈值化。     

基于Grabcut分割和填充物面积判别的复合绝缘子断串诊断

针对传统 GrabCut 算法需要人工初始化而引起图像分割效率低的问题,结合Otsu方法,提出了一种新的Grabcut自动化算法对复合绝缘子进行分割。首先,对原始图像进行HSV空间转换和加权的灰度化处理。其次,对V通道图像和灰度化后的图像进行Otsu分割并进行或逻辑融合,以此来确定目标绝缘子区域,并结合最大连通域定位绝缘子位置坐标完成Grabcut框取初始化,实现Grabcut的自动化。最后,针对绝缘子断串判别准确率低的问题,通过对绝缘子分割图像的最小外接矩形加框、填充、去原图的方式,提出一种新的面积判别方式来诊断并定位故障位置。实验结果表明：Grabcut自动化算法可以很好地分割出目标绝缘子,分割准确率可以达到96.6%以上。所提出的面积判别方法对于具有断串故障的绝缘子检测率可以达到96.6%以上,对于无故障的误检率为6.7%以下。