Where: (1) Intensity&M-LBP, (2) Intensity&S-Textons, (3) Intensity&M-LBP&Gabor, (4) Intensity&M-LBP&S-Textons, (5) Intensity&M-LBP&Gabor&S-Textons and (6) Intensity&M-LBP&Gabor&Wavelets.

Figshare2016-02-29 更新50

Results using the Bagging classifier with colour model and feature combinations and a previous correlation analysis using a threshold of 97%.

图像纹理分类

特征融合优化

Where: (1) Intensity&M-LBP, (2) Intensity&S-Textons, (3) Intensity&M-LBP&Gabor, (4) Intensity&M-LBP&S-Textons, (5) Intensity&M-LBP&Gabor&S-Textons and (6) Intensity&M-LBP&Gabor&Wavelets.

Figshare2016-02-29 更新30

The accuracy and parameters of CNN models.

CNN优化

深度学习架构

Architecture as 200, 21, 4 describes one layer with 200 filters, filter length 21 and pooling size 4; ‘/’ separates data for two layers.

NIAID Data Ecosystem50

Marginal testing of fused social determinant of health features in the full HIDD data and selection frequency in resampled fusion models.

健康社会决定因素

特征融合优化

Marginal testing of fused social determinant of health features in the full HIDD data and selection frequency in resampled fusion models.

NIAID Data Ecosystem30