共查询到20条相似文献,搜索用时 125 毫秒
1.
基于交叉熵的神经网络在病理图像分析中的应用 总被引:1,自引:0,他引:1
目的 探讨利用基于交叉熵的神经网络识别典型病理图像.方法 提取27个宫颈细胞及细胞核的形态学与色度学特征参数并进行标准化处理,运用(SNN)软件对700例宫颈细胞图片进行神经网络分析.结果 细胞按正常、低度和高度鳞状上皮细胞、癌细胞进行分类识别,训练集与测试集识别率分别为98.29%、96.57%.使用相同训练样本数量,识别率高于其他网络模型.结论 该方法可以很好对宫颈细胞进行分类识别,并可应用于其他病理图片定量分析研究. 相似文献
2.
基于独立分量分析的肝纤维化超声图像研究 总被引:1,自引:1,他引:0
目的研究独立分量分析(ICA)方法在区别肝纤维化疾病中的应用.方法采用FastICA算法,在基于正常肝组织与纤维化肝组织可以看作是独立信源的前提下,对同一病人不同纤维化时期和正常人的超声图像选取局部区域后分别进行独立分量分离.结果FastICA算法可以较快的分离出各组的独立分量,且异常组的独立分量数明显多于正常组.结论对肝纤维化超声图像的独立成分进行分析,并与相应部位的正常肝组织超声图像的独立成分进行比较,是一种值得尝试的新的超声图像分析方法. 相似文献
3.
目的:探讨利用K—medoids聚类分析方法来寻找典型病理图像。方法:对正常、低度和高度鳞状上皮内病变宫颈细胞的51个特征参数进行标准化处理,以消除原始变量数量及量纲影响。运用统计软件R对所有特征参数进行K—medoids聚类分析。结果:对于正常、低度和高度鳞状上皮内病变宫颈细胞的分类正确率分别为91.7%、98.0%和92,0%。作为类中心点的3个病理图片可作为典型病理图片指导临床及教学研究。结论:该方法简单、方便,客观性强,可应用到其他医学领域定量研究。 相似文献
4.
目的 观察计算机图像处理技术对妊娠期高血压患者眼底图像中视网膜血管的分割结果,探讨妊娠期高血压性视网膜病变的有效诊断方法.方法 选取55例妊娠期高血压患者的眼底图像,35 例血压正常的孕妇的眼底图像作正常对照,借助计算机图像处理技术,对其眼底图像中的视网膜血管进行分割与提取,分析眼底病理变化的特征参数.结果 计算机图像处理技术对正常和病变眼底图像中视网膜血管的分割均较完整,与眼底图像比较,处理后的视网膜血管更清晰、精确,能恰当地描述眼底视网膜血管的特征现象.结论 计算机图像处理技术对高血压患者眼底图像中视网膜血管的分割效果良好,有助于妊娠期高血压患者视网膜病变的临床无损诊断与健康普查. 相似文献
5.
目的:不同的医学影像为医生提供不同的医学信息,图像融合技术逐渐成为临床诊断和医学研究的重要手段.方法:本文采用了一种基于不可分离小波框架的图像融合算法.算法首先对已配准的医学图像进行非分离小波分解,把图像分解成低频和高频子图像,针对高频分量和低频分量不同的特点,采用不同的融合规则对其进行融合,最后进行小波逆变换,得到最... 相似文献
6.
1种视网膜眼底图像增强的新方法 总被引:1,自引:0,他引:1
目的:提高视网膜眼底图像中血管和病灶部位的可视性,便于临床诊断。方法:首先,对采集到的视网膜眼底图像进行数学形态学处理,以弱化背景,使图像得到初步增强;然后,运用有限对比度自适应直方图均衡化算法改善光照不均的现象;最后,利用二维匹配滤波算法进一步强化血管和病灶部位信息,提高血管与背景的对比度。结果:该方法能很好地弱化背景,增强对比度,凸显感兴趣区域,也对视盘区域血管的分割和病灶部位的识别有极大地帮助。结论:该方法可有效地增强正常和病变的眼底图像,辅助医师诊断。 相似文献
7.
8.
9.
10.
11.
Wireless Capsule Endoscopy (WCE) is a technology in the field of endoscopic imaging which facilitates direct visualization of the entire small intestine. Many algorithms are being developed to automatically identify clinically important frames in WCE videos. This paper presents a supervised method for automated detection of bleeding regions present in WCE frames or images. The proposed method characterizes the image regions by using statistical features derived from the first order histogram probability of the three planes of RGB color space. Despite being inconsistent and tiresome, manual selection of regions has been a popular technique for creating training data in the studies of capsule endoscopic images. We propose a semi-automatic region-annotation algorithm for creating training data efficiently. All possible combinations of different features are exhaustively analyzed to find the optimum feature set with the best performance. During operation, regions from images are obtained by applying a segmentation method. Finally, a trained neural network recognizes the patterns of the data arising from bleeding and non-bleeding regions. 相似文献
12.
Bleeding Detection in Wireless Capsule Endoscopy Based on Probabilistic Neural Network 总被引:1,自引:0,他引:1
Wireless Capsule Endoscopy (WCE), which allows clinicians to inspect the whole gastrointestinal tract (GI) noninvasively,
has bloomed into one of the most efficient technologies to diagnose the bleeding in GI tract. However WCE generates large
amount of images in one examination of a patient. It is hard for clinicians to leave continuous time to examine the full WCE
images, and this is the main factor limiting the wider application of WCE in clinic. A novel intelligent bleeding detection
based on Probabilistic Neural Network (PNN) is proposed in this paper. The features of bleeding region in WCE images distinguishing
from non-bleeding region are extracted. A PNN classifier is built to recognize bleeding regions in WCE images. Finally the
intelligent bleeding detection method is implemented through programming. The experiments show this method can correctly recognize
the bleeding regions in WCE images and clearly mark them out. The sensitivity and specificity on image level are measured
as 93.1% and 85.6% respectively. 相似文献
13.
基于模糊核聚类的MR图像分割新算法 总被引:1,自引:0,他引:1
在传统的模糊聚类算法中引入了核函数,同时引入了控制邻域作用的约束项,提出了改进的基于模糊核聚类的MR图像分割新算法.通过对模拟图和仿真的脑部MR图像的分割实验,证明本算法可以有效地分割含有噪声的图像. 相似文献
14.
目的 为减少脑灌注CT检查的辐射剂量,本文提出了一种基于先验图像约束扩散张量的低剂量脑灌注CT图像恢复方法。方法 脑灌注CT图像之间存在丰富的结构冗余信息,高质量的先验图像可以作为结构互补信息引入到低剂量脑灌注CT图像恢复过程中,抑制低剂量脑灌注CT图像中的噪声和伪影。首先,分别计算出先验图像和低剂量脑灌注CT图像的扩散张量,然后构造先验图像约束的扩散张量,并利用新构建的扩散张量对低剂量脑灌注CT图像进行滤波。结果 在数值体膜实验中,本文方法得到的CBF参数图像的SSIM值与FBP算法相比提高了63%。在临床实验中,本文方法得到的CBF参数图像的SSIM值 与FBP算法相比提高了45%。结论 数值体膜和临床数据实验结果表明本文方法在抑制低剂量脑灌注CT图像噪声和伪影的同时可以保持图像的结构细节特征,并且可以获取准确的脑血流动力学参数图像。 相似文献
15.
本文利用弥散张量磁共振成像(DT-MRI)的弥散张量信息,引入Frobenius范数,提出了一种基于马尔可夫随机场(MRF)的DT-MRI图像分割新算法.本文实验分别使用该算法和K均值算法对DT-MRI图像进行分割,并将分割结果与FA图像予以评估比较.结果 表明,新算法分割DT-MRI图像的精确度高于K均值算法.另外,新算法分割DT-MRI图像效果优于常规MRI(T2WI)图像. 相似文献
16.
The shape prior segmentation procedure and pruned association rule with ImageApriori algorithm has been used to develop an
improved brain image classification system are presented in this paper. The CT scan brain images have been classified into
three categories namely normal, benign and malignant, considering the low-level features extracted from the images and high
level knowledge from specialists to enhance the accuracy in decision process. The experimental results on pre-diagnosed brain
images showed 97% sensitivity, 91% specificity and 98.5% accuracy. The proposed algorithm is expected to assist the physicians
for efficient classification with multiple key features per image. 相似文献
17.
18.
The image feature detection is widely used in image registration, image stitching and object recognition. The feature detection algorithm can be applied to the detection of artificial images, and can be used to detect the energy spectrum CT image. A new algorithm of phase consistency detection based on dimensionality reduction is proposed in this paper. We mainly focus on the phase congruency of the spectral CT images in the paper and try to use dimensionality reduction to integrate the information of phase congruency detected in the image. The experimental results show that the algorithm can detect the energy spectrum CT image with clear edge and contour, which is beneficial to the subsequent processing. Meanwhile, the algorithm presented is more effective in diagnosis of disease for medical professionals. 相似文献
19.