Skip to main content

Table 7 P value matrix of various model performances (AUROC) for 30-day readmission prediction. U, T, S represents unstructured data, temporal signals, and static information respectively

From: Combining structured and unstructured data for predictive models: a deep learning approach

 

LR

RF

Fusion-CNN

Fusion-LSTM

T + S

U

U + T + S

T + S

U

U + T + S

T + S

U

U + T + S

T + S

U

U + T + S

LR

\(T + S\)

1

< 0.001

< 0.001

< 0.001

< 0.001

< 0.001

0.0208

< 0.001

< 0.001

0.0158

0.0076

< 0.001

U

< 0.001

1

< 0.001

< 0.001

< 0.001

< 0.001

< 0.001

< 0.001

< 0.001

< 0.001

0.2449

< 0.001

\(U + T + S\)

< 0.001

< 0.001

1

< 0.001

< 0.001

< 0.001

0.3156

0.0954

< 0.001

< 0.001

< 0.001

< 0.001

RF

\(T + S\)

< 0.001

< 0.001

< 0.001

1

0.073

0.116

< 0.001

< 0.001

< 0.001

< 0.001

< 0.001

< 0.001

U

< 0.001

< 0.001

< 0.001

0.073

1

0.7452

< 0.001

< 0.001

< 0.001

< 0.001

< 0.001

< 0.001

\(U + T + S\)

< 0.001

< 0.001

< 0.001

0.116

0.7452

1

< 0.001

< 0.001

< 0.001

< 0.001

< 0.001

< 0.001

Fusion-CNN

\(T + S\)

0.0208

< 0.001

0.3156

< 0.001

< 0.001

< 0.001

1

0.0661

0.0011

0.1757

0.0012

< 0.001

U

< 0.001

< 0.001

0.0954

< 0.001

< 0.001

< 0.001

0.0661

1

0.0011

< 0.001

< 0.001

< 0.001

\(U + T + S\)

< 0.001

< 0.001

< 0.001

< 0.001

< 0.001

< 0.001

0.0011

0.0011

1

< 0.001

< 0.001

0.07

Fusion-LSTM

\(T + S\)

0.0158

< 0.001

< 0.001

< 0.001

< 0.001

< 0.001

0.1757

< 0.001

< 0.001

1

< 0.001

< 0.001

U

0.0076

0.2449

< 0.001

< 0.001

< 0.001

< 0.001

0.0012

< 0.001

< 0.001

< 0.001

1

< 0.001

\(U + T + S\)

< 0.001

< 0.001

< 0.001

< 0.001

< 0.001

< 0.001

< 0.001

< 0.001

0.07

< 0.001

< 0.001

1