当前位置: X-MOL 学术Microprocess. Microsyst. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Breast thermography based unsupervised anisotropic- feature transformation method for automatic breast cancer detection
Microprocessors and Microsystems ( IF 2.6 ) Pub Date : 2020-05-23 , DOI: 10.1016/j.micpro.2020.103137
P. Gomathi , C. Muniraj , P.S. Periasamy

The most crucial Infrared (IR) cameras provide temperature-sensitive images and chest vascular transitions. Hotspots can be used to emphasize that these images reveal new subtle changes due to pathology. The resulting images show clusters that appear to vary in shape and spatial distribution but carry class dependent information. Automated sensing techniques are challenging because of the location, size, and direction of these clusters. High-level combinations come with spectral invariant features that are suitable for the system to provide transformations stability and shape-dependent information extraction from acoustic images. In this work, the classification of bispectral invariant benefits, diagnostic classification of breast thermal images into malignant, benign, and standard types, participates, and these features are proposed as the basis of Unsupervised Anisotropic- Feature Transformation Method. As indicated by the outcomes, the proposed approach is promising for the location of cancer affected variation from the normal and abnormal women's. All the more imperatively, the results demonstrated the possibility of this structure in breast malignancy identify to open a legitimate path to encouraging methodological and trial to look in this analysis. Also, the proposed mammogram is segmented from the background, which improves the quality of the image by reducing noise followed by a filter implemented on MATLAB software. The proposed approach is to use screening as a diagnostic technique for the most effective breast cancer detection.



中文翻译:

基于乳房热成像的无监督各向异性特征转换方法,用于乳腺癌自动检测

最关键的红外(IR)摄像机提供对温度敏感的图像和胸部血管过渡。热点可以用来强调这些图像揭示了由于病理引起的新的细微变化。生成的图像显示出群集,这些群集的形状和空间分布似乎有所不同,但带有类相关信息。由于这些簇的位置,大小和方向,自动传感技术具有挑战性。高级组合具有频谱不变性特征,适用于系统以提供变换稳定性和从声像中提取形状相关的信息。在这项工作中,参与了双谱不变收益的分类,将乳房热图像的诊断分类分为恶性,良性和标准类型,提出了这些特征作为无监督各向异性特征变换方法的基础。结果表明,所建议的方法有望用于定位受癌症影响的正常女性和异常女性的变异。更重要的是,结果证明了这种结构在乳腺癌中的可能性可能为确定鼓励该研究方法和试验的合法方法开辟一条合法的道路。同样,从背景中分割出建议的乳房X线照片,通过减少噪声以及在MATLAB软件上实现的过滤器来提高图像质量。提出的方法是将筛查用作诊断技术,以最有效地检测乳腺癌。所提出的方法对于定位正常和异常女性的癌症影响变异是有希望的。更重要的是,结果证明了这种结构在乳腺癌恶性肿瘤中识别的可能性,从而为鼓励在该分析中使用方法论和试验方法开辟了一条合法的道路。同样,从背景中分割出建议的乳房X线照片,通过减少噪声以及在MATLAB软件上实现的过滤器来提高图像质量。提出的方法是将筛查用作诊断技术,以最有效地检测乳腺癌。所提出的方法对于定位正常和异常女性的癌症影响变异是有希望的。更重要的是,结果证明了这种结构在乳腺癌中的可能性可能为确定鼓励该研究方法和试验的合法方法开辟一条合法的道路。同样,从背景中分割出建议的乳房X线照片,通过减少噪声以及在MATLAB软件上实现的滤波器来提高图像质量。提出的方法是将筛查用作诊断技术,以最有效地检测乳腺癌。结果表明,这种结构在乳腺恶性肿瘤中识别出可能为鼓励分析方法和方法开辟一条合法途径。同样,从背景中分割出建议的乳房X线照片,通过减少噪声以及在MATLAB软件上实现的过滤器来提高图像质量。提出的方法是将筛查用作诊断技术,以最有效地检测乳腺癌。结果表明,这种结构在乳腺恶性肿瘤中识别出可能为鼓励分析方法和方法开辟一条合法途径。同样,从背景中分割出建议的乳房X线照片,通过减少噪声以及在MATLAB软件上实现的滤波器来提高图像质量。提出的方法是将筛查用作诊断技术,以最有效地检测乳腺癌。

更新日期:2020-05-23
down
wechat
bug