Skip to main content

Table 3 The 11 top performing features obtained from the feature selection scheme from the learning set D 2 that included 3T T2w, ADC MRI sequences

From: Radiomics based targeted radiotherapy planning (Rad-TRaP): a computational framework for prostate cancer treatment planning with MRI

#

mpMRI sequence

feature

1

T2w

Signal intensity

2

T2w

Standard deviation of signal intensity

3

T2w

Sobel gradient (x)

4

T2w

Haralick correlation

5

T2w

Laws energy (kernel = R5E5)

6

ADC

Signal intensity

7

ADC

Haralick energy

8

ADC

Haralick correlation

9

ADC

Haralick differential entropy

10

ADC

Laws energy (kernel = L5W5)

11

ADC

Laws energy (kernel = W5L5)