Skip to main content
Fig. 4 | BMC Medical Imaging

Fig. 4

From: MRI-based radiomics approach for differentiation of hypovascular non-functional pancreatic neuroendocrine tumors and solid pseudopapillary neoplasms of the pancreas

Fig. 4

Quantitative nomogram for the discrimination of SPNs and hypo-NF-pNETs. The closer the risk is to 0.9, the more likely the tumor is to be SPN, and the closer the risk is to 0.1, the more likely it is to be hypo-NF-pNETs. Radscore was calculated by summing the selected features weighted by their coefficients. The final formula of radscore is: Radscore = -0.601 × wavelet_HLH_glszm_GrayLevelNonUniformityNormalized + 0.326 × log_sigma_5_0_mm_3D_gldm_DependenceNonUniformityNormalized + -1.391 × wavelet_HHH_glszm_SizeZoneNonUniformityNormalized + 1.138 × wavelet_HLL_glcm_ClusterShade + -1.778 × log_sigma_5_0_mm_3D_gldm_LargeDependenceLowGrayLevelEmphasis + 1.074 × wavelet_LHL_firstorder_Skewness + 0.199 × log_sigma_4_0_mm_3D_firstorder_90Percentile + 0.002 × wavelet_HHL_glszm_GrayLevelNonUniformityNormalized + 3.734 × wavelet_LLH_glszm_GrayLevelNonUniformityNormalized + 0.76 × wavelet_HLL_firstorder_Skewness + 0.259 × wavelet_HHH_glcm_Imc1 + 0.322 × log_sigma_5_0_mm_3D_glszm_LargeAreaLowGrayLevelEmphasis + 1.754 × log_sigma_3_0_mm_3D_firstorder_Maximum + -0.495*original_shape_Flatness + 2.127

Back to article page