Table 1.

Features for Segmentation.

Features	Explanation
Area $A_r$	Number of pixels in boundary area of ROI
Perimeter $P_r$	Number of pixels on boundary of ROI.
Circularity $C_r$	Number of pixels in boundary area of ROI. When there is a circular shape, circularity has a value of zero. By assuming $A_r$ for area and $P_r$ for the perimeter, circularity will be $C_r=1-\frac{4\pi A_r}{P_r^2}$.
Shape factor	Number of pixels on boundary of ROI. The count of burr around tumors will show the feature that is, region-of-interest as $S_r=\frac{P_r^2}{A_r}$. Assume $A_r$ for area, $P_r$ for the perimeter and $S_r$ for Shape factor.
Normalization radial length	$nrl\left(r\right)=\frac{rl\left(r\right)]}{max\left(rl\right(r\left)\right)}$. Here, $rl$ is radial length value meaning it is Euclidean Distance. So $rl\left(r\right)=\sqrt{{(b_i-t)}^2{(b_j-u)}^2}$. Here, $(b_i,b_j)$ is centre position and $(t,u)$ is boundary pixel position.
Mean-value of normalization based radial length	$nr{l}_{mean}=\frac{1}{P_r}\sum nrl\left(r\right)$
. Standard deviation value	$sigma=\sqrt{\frac{1}{P_r}{\sum }_{r=1}^P{\left(nrl\left(r\right)-nr{l}_{mean}\left(r\right)\right)}^2}$
. Entropy value	$E_r={\sum }_{r=1}^{P_r}{p}_{k}log{p}_k$. $p_k$ =probability of a certain $nrl$ to the number of whole radials and $P_r$ perimeter.
The normalization value of central position shift	$NCPS=\frac{\sqrt{{(b_i-c)}^2+{\left(b_j-d\right)}^2}}{A}$. The pixels coordination position $(c,d)$ is denoted with a minimum gray value inside the ROI. The Euclidian distance is calculated from the ROI centre $(c_i-c_j)$ at the position of the pixel with the lowest gray value, the ROI is divided.
Gradient	the gray value alteration among the boundary pixel and the 10th pixel from this pixel with the radial direction $g_r=I(t,u)-I(i_0,j_0)$, Where $I(t,u)$ is the gray value of the boundary pixel and $I(i_0,j_0)$ is the gray value of the 10th radial pixel.