132

Bioinformatics of the Brain

FIGURE 4.6

Edge-based techniques were applied on same brain MRI images. The source

image featured in this figure was selected from the dataset available as open

source on Kaggle [5].

This is known as the Laplacian of Gaussian (LoG) operation, and it realized

in Equation 4.33:

LoG(x, y) = −¹

πσ⁴



1 −^{x2 + y2}

2σ²



e^{−x2 + y2}

2σ²

(4.33)

Figure 4.6 depicts the resultant images obtained by employing various edge

detection methods mentioned above on a sample brain MRI image.

4.3.3.4

Clustering Techniques

Clustering technique is a type of unsupervised learning method and assigns

data without label information to classes according to certain attributes. In

clustering techniques, k-means and Fuzzy C-means algorithms are frequently

used in brain tumor segmentation tasks. These algorithms are mentioned in

the following sections.

k-means Clustering Algorithm

The k-means algorithm is an iterative method that assigns each data point to

exactly one cluster. Prior to the implementation of the clustering method, the

k-value must be determined. Each cluster has a centroid (center point) which

is initialized randomly. Distances (like Euclidean distance) between the data