Skip to main content

Table 1 Dataset input features

From: Improving the performance of support-vector machine by selecting the best features by Gray Wolf algorithm to increase the accuracy of diagnosis of breast cancer

Column

Feature

Description

1

Mass thickness

Benign cells tend to be in one layer, whereas cancer cells tend to be in multiple layers

2

Uniformity of cell size

Cancer cells have different sizes while benign cells are similar in size

3

Cell shape uniformity

Cancer cells have different shapes, whereas benign cells are similar in shape

4

Marginal bond

Benign cells apt to attach mutually but cancer cells tend to misplace this capability

5

Epithelial cell size

Epithelial cells that are significantly large may be a malignant cell

6

Bare nuclei

Bare nuclei are a term used for nuclei that are not surrounded by cytoplasm. They are usually observed in benign cells

7

Bland chromation

In benign cells, the nucleus (chromatin) is in the form of uniform texture, but cancer cells are not this way

8

Normal nucleoli

Normal nucleoli are small structures observed in cell nucleus. These nuclei are very small in benign cells (observable). However, they are large and distinctive in cancer cells

9

Mitosis division

Mitosis is the division of the cell nucleus into two daughter cells. Pathologists calculate the grade of cancer by counting the number of mitosis divisions