骨龄X射线图像的预处理和改进k余弦的指骨关键点定位

背景：在骨龄自动化评价过程中,二值化阈值的选取直接关系到图像二值化效果以及对指骨关键点的准确定位。目的：解决骨龄X射线图像二值化阈值选取问题,以及k余弦算法定位指骨关键点速度过慢的弊端,从而达到快速准确定位指骨关键点的目的。方法：采用图像预处理技术,提出基于OSTU阈值的二次阈值算法,用该方法获取的阈值来二值化骨龄X射线图像;针对k余弦算法定位指骨关键点速度很慢的缺点,提出一种改进的k余弦算法来快速准确的定位指骨关键点。结果与结论：用改进后的k余弦算法来定位63例指骨关键点的总成功率在95%以上,此法运行速度比k余弦算法快20倍以上,其准确性和速度都达到比较满意的效果。     

指骨骨骺兴趣区域的自动提取方法

背景:在骨龄自动化评定中,许多方法对骨骺兴趣区域的提取不够理想。目的:通过k余弦算法定位出手指骨的特征点,解决对骨骺兴趣区域的提取的困难和提取方法的局限性。方法:提取指骨兴趣区域,并用各向异性扩散方法进行滤波,然后根据骨龄指骨的图像数据进行统计,寻找出定位指骨中心点的方法,通过3次多项式对中心点线进行拟合,定位出指骨的中心轴,在此基础上提取出指骨骨骺兴趣区域。结果与结论:用该方法定位出的指骨中心轴比较真实的反映指骨骨架,在此基础上能提取出指骨骨骺兴趣区域,成功率>95%,并能运用到后续对指骨骨骺兴趣区域相关的评定中。     

指骨骨骺兴趣区域的自动提取方法

背景：在骨龄自动化评定中,许多方法对骨骺兴趣区域的提取不够理想。目的：通过k余弦算法定位出手指骨的特征点,解决对骨骺兴趣区域的提取的困难和提取方法的局限性。方法：提取指骨兴趣区域,并用各向异性扩散方法进行滤波,然后根据骨龄指骨的图像数据进行统计,寻找出定位指骨中心点的方法,通过3次多项式对中心点线进行拟合,定位出指骨的中心轴,在此基础上提取出指骨骨骺兴趣区域。结果与结论：用该方法定位出的指骨中心轴比较真实的反映指骨骨架,在此基础上能提取出指骨骨骺兴趣区域,成功率〉95%,并能运用到后续对指骨骨骺兴趣区域相关的评定中。     

一种鲁棒的骨龄X射线平片自动轮廓提取方法

背景:骨龄X射线平片具有不均匀和复杂性,因而在骨龄自动评价的研究中,手掌轮廓提取的结果往往不理想.目的:采用计算机自动提取手掌轮廓,为骨龄自动评价中图像预处理阶段的研究奠定重要基础.为了解除骨龄评定带来的主观性和不确定性,提出用计算机进行自动评价.方法:在仔细分析骨龄X射线平片的基础上,提出了一种对图像背景的可行有效子采样点方案,并提出用二元三次线性回归方法来模拟图像背景,通过形态学以及二进制标记等～系列操作,最后成功提取出手掌轮廓.结果与结论:采用异常点移除和回归方法相结合来提取轮廓,采取固定阀值,不受阀值选取的困扰,并具有鲁棒性.大量的实验结果表明了该方法提取的手掌轮廓成功率在93%以上,能完全应用于骨龄自动识别的后续研究工作.     

深度学习模型在儿童TW3法骨龄评估的临床效能分析

目的 探讨深度学习模型在儿童应用TW3法进行骨龄评估的临床效能。方法 回顾性收集180例儿童左手X线片。评估基于Tanner-WhitehouseⅢ(TW3)法的掌指骨和腕骨骨龄。以3位高年资主任医师的评估均值为金标准组,计算并比较深度学习模型(模型组)及2位低年资医师(医师组,分别记作医师1、医师2)与金标准组评估时间差异,掌指骨骨龄、腕骨骨龄的均方误差(MSE)及平均绝对误差(MAE);采用组内相关系数(ICC)分析模型组和医师组与金标准组结果一致性。结果 骨龄评估所用时间模型组明显少于医师组(P<0.05)。掌指骨骨龄评估,模型组和医师1、医师2与金标准组相比MSE、MAE差异有统计学意义(P<0.05)。腕骨骨龄评估,模型组和医师1、医师2与金标准组相比MSE、MAE差异有统计学意义(P<0.05)。掌指骨骨龄评估,模型组与金标准组ICC为0.988,医师1、医师2与金标准组的ICC分别为0.986、0.977。腕骨骨龄评估,模型组与金标准组ICC为0.971,医师1、医师2与金标准组ICC分别为0.970、0.953。结论对于儿童,应用TW3法深度学习模型在...     

基于智能计算的自动骨龄评估及其与TW3法比较

目的 寻找新的算法以提高自动化骨龄评估(ABAA)的准确性和实用性.方法 建立基于目标的兴趣区(ROD.按照Tanner-Whitehouse(TW3)法将ROI分为RUS(包括尺桡骨及掌指骨)ROI及腕骨ROI.按离子群优化(PSO)法,每个兴趣区提取5项特征(包括大小、形态及融合或比邻状态)输入人工神经网络(ANN)分类器,ANN建立在前馈的多层网络基础上,并以反向传播算法规则训练ANN以分别处理RUS及腕骨特征.约1046份左手及腕的数字X线片被随机分成两部分,一半用以训练ANN,另一半用以ABAA,而之前全部采用TW3法有两名小儿内分泌专家人工判读骨龄.结果 不同专家判读的骨龄间比较提示:RUS骨龄的标准差大于腕骨骨龄(4.40和2.42),但二者的变异系数(CV)均     

基于灰度对比和自适应小波变换的X射线图像增强

背景:X射线检查作为常规的检查方式得到了广泛的应用,然而由于现有技术的局限性,使得X射线图像往往具有灰度对比度低和噪声影响等缺点,因此,现有的X射线图像往往达不到医生的要求.目的:增强和去噪处理对比度较低且含有噪声的X射线图像,以达到易于医生理解和识别的目的.方法:针对空间域处理和变换域处理增强X射线图像的不足,提出了基于灰度对比和自适应小波变换的X射线图像增强算法.首先,选择需要增强和减弱的灰度范围,并根据八邻域灰度对比增强算法对X射线图像进行灰度变换,并用中值滤波算法对图像进行平滑;然后,对X射线图像进行小波分解,并运用相邻分解层之间相关系数的大小来确定细节信号和噪声.结果与结论:应用了灰度对比和自适应小波变换相结合的X射线图像增强算法,把基于空间域增强的方法和基于变换域的方法有机地结合起来,比传统的单一增强方法更为优越.实验结果证明它能自适应地增强X射线图像的灰度对比,使得图像细节的显示更加清晰,同时在一定程度上去除了噪声的干扰,对于灰度对比度较低的图像效果更加明显.     

基于空间自动配准手术计划系统的设计和实现

文章提出了一种基于手术空间和图像空间自动配准的手术计划系统设计和实现方法:①选取3个有一定间隔的横断面.②使用模板匹配法自动识别每一个横断面上的标志点.③用识别的数据建立手术空间和图像空间的刚体变换模型,再用最小二乘法解方程组得到两个空间的最优变换矩阵和平移矩阵.最后再通过建立脑空间和图像空间映射关系来计算功能靶点坐标.基于以上算法设计了手术计划系统软件,对比了手动和自动方法识别标志点的运行时间,并设计系统改进前后功能靶点计算值和手术空间实际测量值对比实验.结果证明以上方法有较好的靶点计算速度和精度.     

自适应超分辨率图像的重建算法

背景:超分辨率重建已经在视频、遥感等许多领域内的到广泛的研究与应用。目的:介绍一种自适应超分辨率重建算法,以期从序列低分辨率图像中重建出高分辨率图像。方法:采用常数λ=2/3作为正则化参数和自适应步长作为第一种方案。第二种方案充分考虑到低分辨率图像中的运动误差估计、点扩散函数以及加性高斯白噪声对重建算法的影响。实验构造出新的非线性自适应正则化函数,进而利用实验方法分析代价函数的凸性。通过数学理论,根据代价函数凸性实验得到自适应步长因子,从而改进了图像的空间分辨率和算法的收敛速度。结果与结论:为验证此算法的有效性,采用光学图像进行实验。方案二图像峰值信噪比增高,其收敛速度为方案一的2倍以上;方案二的平均计算需要的时间为68.25s。结果证实,自适应超分辨率图像重建算法对图像分辨率和迭代的收敛速度均改善显著,其稳定性较好。     

基于SAPSO优化三维Otsu方法的医学图像分割算法

医学图像具有内容丰富多样、特征丰富、多尺度等特性,因此对医学图像的分割比一般图像的难度更大.针对上述问题,提出了基于改进粒子群优化三维Otsu方法的医学图像分割算法.由于三维Otsu方法计算量大,采用粒了群优化算法来搜索阈值向量,每个粒子代表一个可行的阈值向量,通过粒子群之间的协作来获得最优阈值.由于粒子群优化算法容易陷入局部最优解的的缺点,提出了模拟退火的粒子群优化方法,使其能够快速准确得到整体最佳解,还能保持粒子群算法求解速度快的优点.最后通过仿真实验得出了结论表明,所提出的方法不仅能得到理想的结果,而且计算量人大减少.     

Contribution of the extrinsic and intrinsic hand muscles to the moments in finger joints

OBJECTIVE: The purpose of this current work is to develop a method of estimating force produced by the extrinsic and intrinsic hand muscles, and to estimate the contribution of these muscles to the finger joint moments. DESIGN: Experimental methods and a biomechanical model were developed for the estimation of (a) moments produced at finger joints, and (b) contribution of the intrinsic and extrinsic muscles to the moments, (c) forces of the extrinsic and intrinsic muscles within individual fingers. BACKGROUND: Because of the differential insertions of the extrinsic flexors, it is possible to isolate their mechanical effect at finger joints. METHODS: During the experiment, the location of force application was varied in parallel along individual fingers. The points of force application were on the distal phalanx, at the distal interphalangeal joint, or at the proximal interphalangeal joint. RESULTS: When the point of force application was varied in the proximal direction from the distal phalanx to the proximal interphalangeal joint the moment at a given joint decreased. The intrinsic and extrinsic muscle forces were dependent on the experimental conditions. The extrinsic muscles were the major contributors in counterbalancing finger joint moments when the point of force application was distal beyond the proximal interphalangeal joint. CONCLUSION: This current work provides both an experimental protocol and a biomechanical model that allows estimation of the contribution of the intrinsic and extrinsic muscles to finger joint moments. RELEVANCE: This study suggests ways of identifying the source of functional deficiency in the hand.     

基于改进SUSAN算法的医学图像边缘检测

背景:医学图像的边缘检测是医学图像处理中的一项重要的技术,也是医学图像进一步处理的基础。目的:运用改进的SUSAN算法对医学图像进行边缘检测,取得更丰富的医学图像边缘信息,以便于医学图像的进一步处理。方法:运用Sobel算子对SUSAN算法进行了改进,采用C++语言编程,并在VC++6.0开发平台上实现了改进算法。结果与结论:实验结果表明,该算法能实现阈值的自适应选取,对医学图像中的低对比度的图像边缘有较好的检测效果。     

模拟定位X线片和电脑采集图比较分析

目的：探讨射野模拟定位时X线片和电脑采集图各自效果及特点。方法：住院需放疗的肿瘤病人经模拟定位后，即拍X线片，同时电脑采集定位图像。结果：X线片的图像质量明显优于电脑采集图，而电脑采集图的资料内客明显多于X线片。结论：模拟定位X线片和电脑采集图各有特点，肢体薄的部位（如头颈部、胸部）多使用电脑采集图清晰、方便；肢体厚的腹部和脊拄侧住采取X线片图像质量高。     

Sensory characteristics of tender points in the lower back

Palpation of tender points in superficial tissue is commonly undertaken in the management of musculoskeletal pain. The sensory characteristics of digitally tender points (DTPs) have not been defined. This study had two major aims: 1) to characterise ‘Strain–Counterstrain’ DTPs, using quantitative sensory testing (QST) in participants with low back pain (LBP); 2) to compare corresponding points at lumbar sites in participants with LBP to those without LBP. Fifteen participants with LBP (9 females), mean (SD) Oswestry scores 20.8 (10.1)) and 15 participants without LBP (6 females) were included. QST was undertaken by a single examiner blind to the location of DTPs and included measurement of electrical detection and electrical pain threshold, thermal (hot/cold) detection and thermal pain threshold, vibration detection threshold and pressure-pain threshold. In participants with LBP, DTPs demonstrated significantly lower electrical detection and electrical pain thresholds compared to contralateral non-tender points (p < 0.0001). These findings may be indicative of altered central processing of Aβ afferents with terminal receptors at DTPs. Participants with LBP demonstrated elevated cold pain thresholds at lower back sites and at the peripheral shoulder site compared to participants without LBP (p < 0.001). This may also indicate augmented central pain processing in participants with LBP.     

Binarization of microarray data on the basis of a mixture model

Although gathered as continuous data, expression measurements from gene microarrays may be quantized before downstream analysis and modeling. This is especially true for modeling gene prediction and genetic regulatory networks. Coarse quantization results in lower computational requirements, lower data requirements for model inference, and easier conceptualization. This paper proposes a mixture model for binarization. For each gene, the model, composed of a sum of two distributions, is fit to expression data for that gene, and data points are binarized according to the model. The mixture model is based on the assumption of multiplicative up-regulation. The proposed method is compared with mean and median binarization by comparing classification performance based on the binary data from the different methods. Classification is performed for simulated data generated from a microarray model studied previously and for cancer data arising from two studies involving hereditary breast cancer and small, round blue-cell tumors of childhood.     

Automatic threshold selection algorithm to distinguish a tissue chromophore from the background in photoacoustic imaging

The adaptive matched filter (AMF) is a method widely used in spectral unmixing to classify different tissue chromophores in photoacoustic images. However, a threshold needs to be applied to the AMF detection image to distinguish the desired tissue chromophores from the background. In this study, we propose an automatic threshold selection (ATS) algorithm capable of differentiating a target from the background, based on the features of the AMF detection image. The mean difference between the estimated thickness, using the ATS algorithm, and the known values was 0.17 SD (0.24) mm for the phantom inclusions and -0.05 SD (0.21) mm for the tissue samples of malignant melanoma. The evaluation shows that the thickness and the width of the phantom inclusions and the tumors can be estimated using AMF in an automatic way after applying the ATS algorithm.