基于边界特征的乳腺肿瘤超声图像识别

通过对乳腺肿瘤边界特征的分析，得到边界的特征量似圆度，面积比率，长宽比组成的特征矢量，最后用反向传播人工神经网络（BP）的算法对经病理证实的119幅乳腺良、恶性肿块超声图像进行分类识别。BP神经网络对良、恶性肿瘤正确识别率分别为89．7％、73．5N。量化后的乳腺超声图像肿瘤轮廓特征结合BP神经网络可以比较有效的用于肿瘤的良、恶性识别。     

Noninvasive measurement of arterial diameters in humans using ultrasound echoes with prefiltered waveforms

A system for the direct measurement of the inner diameters of selected human arteriesin situ has been developed. Ultrasound pulses are emitted perdendicular to the vessel axis, and reflected from the wall material interfaces. Improved depth resolution is achieved by using a high-frequency transducer, and by inverse filtering of the signal. This is performed by emitting a waveform calculated beforehand to give an optimum time resolution in the received echo. Echoes from both innerand outer surfaces of the vessel walls are discriminated, and the inter-echo time intervals can be measured with an accuracy of 33ns, corresponding to approximately 26μm. By repeating the measurements through the cardiac cycle, the diameter variations can be accurately described.     

Ultrasonographic assessment of human skeletal muscle size

The measurement of human muscle size is essential when assessing the effects of training, disuse and ageing. The considered gold standard for cross-sectional area measurements of muscle size is magnetic resonance imaging (MRI). However, MRI is costly and often inaccessible. The aim of the present study was to test the reproducibility and validity of a more accessible alternative method using ultrasonography (ULT). We examined the cross-sectional areas in the vastus lateralis muscle of six individuals. Axial-plane ULT scans were taken at given levels along the entire muscle length. The ULT scanning was repeated on different days (reliability) and validated against MRI-based measurements. Mean intraclass correlation coefficients were 0.998 for the reliability of ULT and 0.999 for the validity of ULT against MRI. The coefficient of variation values for cross-sectional area measurements assessed by six different experimenters were 2.1% and 0.8% for images obtained with ULT and MRI, respectively. The ULT method is a valid and reliable alternative tool for assessing cross-sectional areas of large individual human muscles. The present findings justify the application of the ULT method for the detection of changes throughout large muscles in response to training, disuse or as a consequence of sarcopenia.     

基于F-score特征选择和支持向量机的P300识别算法

如何从脑电信号中快速准确地识别出P300成分是脑-机接口研究中的一个热点问题.针对P300的识别问题,我们提出了一种将F-score特征选择与支持向量机相结合的判别方法,该方法采用F-score特征选择减少输入特征的维数,以克服支持向量机算法判别速度慢的缺点;然后借助支持向量机算法良好的分类性能实现P300的识别.本文在BCI Competition 2003的P300实验数据集上对该方法进行了验证,结果表明,在5次重复实验中该方法的识别准确率达到了100%,且判别速度与未经特征选择的传统支持向量机算法相比提高了近2倍.     

Application of Scanning Acoustic Microscopy for Assessing Stress Distribution in Atherosclerotic Plaque

Scanning acoustic microscopy (SAM) was equipped to assess the acoustic properties of normal and atherosclerotic coronary arteries. The SAM image in the atherosclerotic lesion clearly demonstrated that the sound speed was higher than that in the normal intima, and that the variation of elasticity was found within the fibrous cap of the plaque. Young's elastic modulus of each region was calculated and the finite element analysis was applied to derive the stress distribution in these arterial walls. In a case of normal coronary artery, the stress was dominant in the intima and the distribution was rather homogeneous and in a case of atherosclerosis, high stress was concentrated to the relatively soft lesion in the fibrous cap overlying lipid pool. SAM provides information on the physical properties, which cannot be obtained by the optical microscope. The results would help in understanding the pathological features of atherosclerosis. © 2001 Biomedical Engineering Society. PAC01: 8764-t, 8763Df, 8719Xx, 8719Rr     

超声筛查胎儿发育异常的回归模型构建研究

目的 用Logistic多因素非条件回归分析，建立超声筛查胎儿发育异常回归模型。方法对影响胎儿发育异常的各个因素进行非条件Logistic回归分析，计算标准偏回归系数、0R值等。结果共有7个因素进入方程，从各影响因素的标准偏回归系及OR值大小分析，以股骨、孕妇年龄、羊水对回归模型影响较大。结论胎儿发育异常与多种因素有关，股骨短小的孕妇应作为超声重点监测对象。     

超声波辐照后家兔肝脏超微结构的变化

本文作者用诊断剂量的超声波辐照活体家兔肝脏，喂养３ｄ致死后取材，经处理后用电子显微镜观察其超微结构的变化情况。实验结果表明超声波辐照２０ｍｉｎ的家兔肝脏的超檄结构无明显变化，而辐照３０ｍｉｎ及６０ｍｉｎ的家兔肝脏，除细胞质内粗面内质网有程度不同的扩张外，肝组织及肝细胞的其它结构均无明显改变。由于热效应与辐照时间有关，而本实验辐照３０ｍｉｎ及６０ｍｉｎ的家兔肝脏的超微结构的变化是相同的。且本实验所用超声强度小而不致引起机械效应。因此作者认为本实验出现的兔肝细胞粗面内质网扩张的现象不会是热效应和机械效应引起，而只可能是空化效应所致，这说明诊断超声可能在动物组织中引起空化效应。     

The use of ultrasound in automated prostatectomy

A new way of applying transurethral ultrasound scanning in the common surgical procedure of the transurethral resection of the prostate is described. The scanning is incorporated as part of a robotic procedure for surgery, so that the overall time spent in an operation can be further shortened, and a safe and accurate operation can be achieved. The prostate dimensions obtained pre-operatively by the transrectal method and those obtained operatively are compared. A robotic system, which was developed specifically to remove prostatic adenoma automatically, is discussed. The system, called a motorised frame, is briefly described, together with its predecessor, a manual frame, in relation to ultrasound measurements. Sizing of the prostate pre-operatively using transrectal ultrasound methods is discussed, using both the manual and the motorised frame. The shortcomins of transrectal ultrasound for use in a robotic procedure are highlighted.     

Outlining the prostate boundary using the harmonics method

In a computerised ultrasound image guidance for automated prostatectomy system, it is necessary to identify a smooth, continuous contour for the prostate (boundary) from the ultrasound image. The radial bas-relief (RBR) method, which has been reported previously, can extract a skeletonised image from an ultrasound image automatically. After this process the prostate boundary is clearly revealed. However, analysis of the image is far from complete, as there are many spurious branches that create too much ambiguity for the system to define the actual boundary. There are also sections missing from the prostate boundary. Therefore further post-processing is required to describe and define the prostate boundary. In the paper, the harmonics method is used to describe the prostate boundary. The harmonics method uses Fourier information for noise removal and encodes a smooth boundary. The results of using the harmonics method after application of the RBR method on ultrasound images are presented. Factors that affect the performance are also highlighted and discussed.