Table 1 Extracted morphological and textural features of collagen fibers from SHG processed images.
From: Deep learning enables automated scoring of liver fibrosis stages
No. | Feature Descriptions |
|---|---|
Morphological Features | |
1 | Total number of collagen fibers |
2–3 | Median and variance of fiber orientation |
4–6 | Median, total and variance of fiber length |
7–9 | Median, total and variance of fiber width |
10 | Total perimeter of collagen fibers |
11–14 | No. of long & short and thick & thin fibers |
15–16 | Ratio of short/long fibers and thin/thick fibers |
17–19 | Median, total and variance of fiber area |
20 | Collagen mean intensity |
21 | Fiber proportionate area (CPA) |
Texture Features | |
22 | Entropy |
23–34 | Contrast, correlation, energy and homogeneity from the GLCM given three different pixel distances at two, four, eight pixels |
35–40 | Energy, entropy, mean, standard deviation, third moment and fourth moment of the coefficients from Fouriers transform |
41–100 | Energy, entropy, mean, standard deviation, third moment and fourth moment of the wavelet decomposition coefficients from ten sub-images generated by Daubechies wavelet transform |
101–130 | Energy, entropy, mean, standard deviation, third moment and fourth moment of the magnitude of the convolution over the image with Gabor filters at five scales |
