影像组学资料

影像组学：它的研究对象是放射影像，它的研究方法则是将影像内包含的所有信息提取出来然后进行综合系统化分析。更确切的说，影像组学是采用自动化算法从影像的感兴趣区（ROI）内提取出大量的特征信息作为研究对象，并进一步采用多样化的统计分析和数据挖掘方法从大批量信息中提取和剥离出真正起作用的关键信息，最终用于疾病的辅助诊断、分类或分级。

pyradiomics官网

Feature Classes

Currently supports the following feature classes:

• First Order Statistics (19 features,一阶统计量)
  体素强度的一阶统计特征值，包括平均值，最大、小值等，
  1. Energy；
  2. Total Energy；
  3. Entropy；
  4. Minimum，numpy.min(Array)；
  5. 10th percentile，numpy.percentile(Array, 10)；
  6. 90th percentile，numpy.percentile(Array, 90)；
  7. Maximum，numpy.max(Array)；
  8. Mean，numpy.mean(Array)；
  9. Median，numpy.median(Array)；

  10. Interquartile Range，numpy.percentile(Array, 75)-numpy.percentile(Array, 25)；
  11. Range，numpy.max(Array)-numpy.min(Array)；
  12. Mean Absolute Deviation；
  13. Robust Mean Absolute Deviation；
  14. Root Mean Squared；
  15. Standard Deviation，numpy.std(Array)；
  16. Skewness；偏斜度？
  17. Kurtosis；峰度
  18. Variance，numpy.std(Array)**2；
  19. Uniformity；

• Shape-based (3D) (16 features，基于形状3D)
  3D上的形状特征值，包括三维方向上的大小及形状特征数据

  1. Volume  体积
  2. Surface Area  表面面积
  3. Surface Area to Volume ratio  表面积与体积之比
  4. Sphericity  球形度(病灶区域相对于球体形状的圆度的量度。这是一个无量纲测量，独立于尺度和方向)
  5. Compactness 1  紧凑度(病灶区域形状相对于球体的紧凑程度的量度。与球形度和冗余有关)
  6. Compactness 2  
  7. Spherical Disproportion  球面歧化(病灶区域表面积与球体表面积之比)
  8. Maximum 3D diamete  最大三维直径
  9. Maximum 2D diameter (Slice)  最大2D直径（切片）
  10. Maximum 2D diameter (Column)  最大2D直径(列)
  11. Maximum 2D diameter (Row)  最大2D直径(行)
  12. Major Axis
  13. Minor Axis
  14. Least Axis
  15. Elongation  延伸率
  16. Flatness  平面度

• Shape-based (2D) (10 features，基于形状2D)
  2D上的形状特征值，包括ROI的二维大小及形状特征值
  1. Mesh Surface；
  2. Pixel Surface；
  3. Perimeter；
  4. Perimeter to Surface ratio；
  5. Sphericity；
  6. Spherical Disproportion；
  7. Maximum 2D diameter；
  8. Major Axis Length；
  9. Minor Axis Length；
  10. Elongation；
   
• Gray Level Cooccurence Matrix (24 features,灰度共生矩阵,GLCM)
  能量、相关性、对比度、熵值等
  1. Autocorrelation；
  2. Joint Average；
  3. Cluster Prominence；
  4. Cluster Shade；
  5. Cluster Tendency；
  6. Contrast；
  7. Correlation；
  8. Difference Average；
  9. Difference Entropy；
  10. Difference Variance；
  11. Joint Energy；
  12. Joint Entropy；
  13. Informational Measure of Correlation (IMC) 1；
  14. Informational Measure of Correlation (IMC) 2；
  15. Inverse Difference Moment (IDM)；
  16. Maximal Correlation Coefficient (MCC)；
  17. Inverse Difference Moment Normalized (IDMN)；
  18. Inverse Difference (ID)；
  19. Inverse Difference Normalized (IDN)；
  20. Inverse Variance；
  21. Maximum Probability；
  22. Sum Average；
  23. Sum Entropy；
  24. Sum of Squares；
   
• Gray Level Size Zone Matrix (16 features,灰度大小区域矩阵,GLSZM)
  1. Small Area Emphasis (SAE);
  2. Large Area Emphasis (LAE);
  3. Gray Level Non-Uniformity (GLN):
  4. Gray Level Non-Uniformity Normalized (GLNN);
  5. Size-Zone Non-Uniformity (SZN);
  6. Size-Zone Non-Uniformity Normalized (SZNN);

  7. Zone Percentage (ZP);
  8. Gray Level Variance (GLV);
  9. Zone Variance (ZV);
  10. Zone Entropy (ZE);
  11. Low Gray Level Zone Emphasis (LGLZE);
  12. High Gray Level Zone Emphasis (HGLZE);
  13. Small Area Low Gray Level Emphasis (SALGLE);
  14. Small Area High Gray Level Emphasis (SAHGLE);
  15. Large Area Low Gray Level Emphasis (LALGLE);
  16. Large Area High Gray Level Emphasis (LAHGLE)

• Gray Level Run Length Matrix (16 features,灰度游程矩阵,GLRLM)
  1. Short Run Emphasis (SRE);
  2. Long Run Emphasis (LRE);
  3. Gray Level Non-Uniformity (GLN);
  4. Gray Level Non-Uniformity Normalized (GLNN);
  5. Run Length Non-Uniformity (RLN);
  6. Run Length Non-Uniformity Normalized (RLNN);
  7. Run Percentage (RP);
  8. Gray Level Variance (GLV);
  9. Run Variance (RV);
  10. Run Entropy (RE);
  11. Low Gray Level Run Emphasis (LGLRE);
  12. High Gray Level Run Emphasis (HGLRE);
  13. Short Run Low Gray Level Emphasis (SRLGLE);
  14. Short Run High Gray Level Emphasis (SRHGLE);
  15. Long Run Low Gray Level Emphasis (LRLGLE);
  16. Long Run High Gray Level Emphasis (LRHGLE)

• Neighbouring Gray Tone Difference Matrix (5 features,相邻灰度差分矩阵,NGTDM)
  1.getCoarsenessFeatureValue():粗度。Coarseness is a measure of average difference between the center voxel and its neighbourhood and is an indication of the spatial rate of change. A higher value indicates a lower spatial change rate and a locally more uniform texture.
  2. getContrastFeatureValue():对比度。Contrast is a measure of the spatial intensity change, but is also dependent on the overall gray level dynamic range. Contrast is high when both the dynamic range and the spatial change rate are high, i.e. an image with a large range of gray levels, with large changes between voxels and their neighbourhood.
  3.getBusynessFeatureValue():Busyness。A measure of the change from a pixel to its neighbour. A high value for busyness indicates a ‘busy’ image, with rapid changes of intensity between pixels and its neighbourhood.
  4.getComplexityFeatureValue():复杂度。An image is considered complex when there are many primitive components in the image, i.e. the image is non-uniform and there are many rapid changes in gray level intensity.
  5.getStrengthFeatureValue():强度。Strength is a measure of the primitives in an image. Its value is high when the primitives are easily defined and visible, i.e. an image with slow change in intensity but more large coarse differences in gray level intensities.

• Gray Level Dependence Matrix (14 features，灰度依赖矩阵，GLDM)
  1. Small Dependence Emphasis (SDE);
  2. Large Dependence Emphasis (LDE);
  3. Gray Level Non-Uniformity (GLN);
  4. Dependence Non-Uniformity (DN);
  5. Dependence Non-Uniformity Normalized (DNN);
  6. Gray Level Variance (GLV);
  7. Dependence Variance (DV);
  8. Dependence Entropy (DE);
  9. Low Gray Level Emphasis (LGLE);
  10. High Gray Level Emphasis (HGLE);
  11. Small Dependence Low Gray Level Emphasis (SDLGLE);
  12. Small Dependence High Gray Level Emphasis (SDHGLE);
  13. Large Dependence Low Gray Level Emphasis (LDLGLE);
  14. Large Dependence High Gray Level Emphasis (LDHGLE)；
•  
•  
•  
• 每个特征参数代表含义随官网查看。
• 
• 
   

 

 

Reference:

影像组学简述
影像组学学习笔记
图像特征提取（纹理特征）
基于LBP纹理特征计算GLCM的纹理特征统计量+SVM/RF识别纹理图片
医学影像组学数据特征值介绍
《图像纹理特征提取方法综述》论文笔记
PyRadiomics 提取特征类之First Order
PyRadiomics 提取特征类之Shape
pyradiomics官网

 

道阻且长，行且将至；行而不辍，未来可期~~~