Skip to main content

Table 4 Test independence of clusters generated from each pair of imaging modalities, using the original features (\(X_j\)) and DGCCA projected features (\(X_jU_j\))

From: Deep multiview learning to identify imaging-driven subtypes in mild cognitive impairment

 

Original features (\(\textit{p}=116\))

DGCCA Projected Features (\(k=20\))

 

Test Statistic

p-value

Test Statistic

p-value

VBM vs. AV45

1.089

0.99

11.959

<0.0001

VBM vs. FDG

1.742

0.38

102.786

<0.0001

AV45 vs. FDG

12.522

<0.0001

11.381

<0.0001

  1. Significant results (p-value ≤ 0.05) are shown in bold
Back to article page

BMC Bioinformatics

ISSN: 1471-2105

Contact us