Table 2.
Selected radiomic features for predicting laterality after univariate filtering and penalization method.
| Feature type | Number of features | Filter |
|---|---|---|
| (A) Hippocampus ( n = 48) | ||
| First order | 1 | None |
| GLCM dist = 1/dist = 2/dist = 3 | 6 | None |
| GLCM dist = 1/dist = 2/dist = 3 | 10 | LLL |
| GLRLM | 1 | LLL |
| GLCM dist = 1/dist = 2/dist = 3 | 14 | LLH |
| GLRLM | 1 | LLH |
| GLCM dist = 1/dist = 2/dist = 3 | 15 | LHL |
| Radiomic feature | Filter | Feature type |
| (B) Extrahippocampal region ( n = 99) | ||
| First order | 4 | None |
| GLCM dist = 1/dist = 2/dist = 3 | 8 | None |
| GLRLM | 4 | None |
| GLCM dist = 1/dist = 2/dist = 3 | 21 | LLL |
| First order | 1 | LLH |
| GLCM dist = 1/dist = 2/dist = 3 | 20 | LLH |
| GLRLM | 2 | LLH |
| First Order | 1 | LHL |
| GLCM dist = 1/dist = 2/dist = 3 | 20 | LHL |
| GLRLM | 2 | LHL |
| GLCM dist = 1/dist = 2/dist = 3 | 16 | LHH |
FLAIR, fluid-attenuated inversion recovery; std, standard deviation; GLCM, gray-level co-occurrence matrix; GLRLM, gray-level run-length matrix; H, high-pass filter; L, low-pass filter.