Table 4.
Features identified with other classification methods. The most contributive features were identified with LASSO logistic regression, linear kernel SVM, and random forest classification.
| Pathway | Elastic net LR | LASSO LR | Linear SVM | RF |
|---|---|---|---|---|
| GCC | ||||
| BCC | FA | FA | ||
| SCC | FA | FA | FA | FA, MD, RD |
| Cg_left | AD | AD | FA, MD, AD, RD | |
| Cg_right | RD | FA | ||
| CST_left | FA | FA | ||
| CST_right | AD | |||
| IFO_left | FA | |||
| IFO_right | FA, AD | |||
| ILF_left | ||||
| ILF_right | FA, MD, AD, RD | |||
| MCP | FA | FA | FA | FA, MD, AD, RD |
| OR_left | AD | MD, RD | ||
| OR_right | FA,AD | AD | ||
| UNC_left | FA | FA | FA | MD, AD, RD |
| UNC_right | FA | FA | FA | MD, AD, RD |
| PV-V4_left | ||||
| PV-V4_right | AD | |||
| PV-MT_left | MD, AD, RD | |||
| PV-MT_right | MD, AD, RD | |||
| Pre-primaryMC_left | ||||
| Pre-primaryMC_right | FA | FA | FA | |
| Thal-PSC_left | ||||
| Thal-PSC_right | FA | |||
| AR_left | FA | |||
| AR_right | FA, MD, RD | FA |
Abbreviations: FA, fractional anisotropy; MD, mean diffusivity; AD, axial diffusivity; RD, radial diffusivity; GCC, genu of the corpus callosum; BCC, body of the corpus callosum; SCC, splenium of the corpus callosum; CG, cingulum; CST, corticospinal tract; IFO, inferior fronto-occipital fasciculus; ILF, inferior longitudinal fasciculus; MCP, middle cerebellar peduncle; OR, optic radiation; UNC, uncinate fasciculus; PV-V4, tract connecting v1 + v2 and v4; PV-MT, tract connecting the v1 + v2 and MT; Pre-primaryMC, tract connecting the premotor and primary motor cortices; Thal-PSC, Tract connecting the sensory thalamus and primary sensory cortex; AR, acoustic radiation.