From: Thyroid nodule recognition in computed tomography using first order statistics
Feature type | Equations | Description |
---|---|---|
Entropy | \( e = - \mathop \sum \limits_{{l = 1}}^{k} \left[ {p\left( l \right)} \right]\log _{2} \left[ {p\left( l \right)} \right] \) | Describes the randomness and irregularity of all pixel intensity |
Uniformity | \( u = \mathop \sum \limits_{{l = 1}}^{k} \left[ {p\left( l \right)} \right]^{2} \) | Describes the distribution of gray level degree |
Mean intensity | \( m = \frac{1}{\text{n}}\mathop \sum \limits_{{{\text{i}} = 1}}^{\text{n}} p(i) \) | Describes the mean intensity value of all pixels |
Standard deviation | \( sd = \left( {\frac{1}{{\left( {n - 1} \right)}}\mathop \sum \limits_{{\left( {x,y} \right) \in R}} \left[ {{\text{a}}\left( {x,y} \right) - \overline{\text{a}} } \right]^{2} } \right)^{{\frac{1}{2}}} \) | Describes the off variation from the mean pixel value |
Kurtosis | \(\begin{aligned} k &= \frac{{n\left( {n + 1} \right)}}{{\left( {n - 1} \right)\left( {n - 2} \right)\left( {n - 3} \right)}}\frac{{\mathop \sum \nolimits_{{\left( {x,y} \right) \in R}} \left[ {{\text{a}}\left( {x,y} \right) - \overline{\text{a}} } \right]^{4} }}{{\left[ {sd\left( a \right)} \right]^{4} }} \\ & \quad - 3\frac{{\left( {n - 1} \right)^{2} }}{{\left( {n - 2} \right)\left( {n - 3} \right)}} \end{aligned}\) | Indicates the bulging or peakedness |
Skewness | \( s = \frac{n}{{\left( {n - 1} \right)\left( {n - 2} \right)}}\frac{{\mathop \sum \nolimits_{{\left( {x,y} \right) \in R}} \left[ {{\text{a}}\left( {x,y} \right) - \overline{\text{a}} } \right]^{3} }}{{sd\left( a \right)^{3} }} \) | Indicates the asymmetry |