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

Table 3 P value matrix of various model performances (AUROC) for in-hospital mortality prediction. U, T, S represents unstructured data, temporal signals, and static information respectively

		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	< 0.001	< 0.001	< 0.001	< 0.001	< 0.001	< 0.001
	U	< 0.001	1	< 0.001	< 0.001	< 0.001	< 0.001	0.5644	0.7432	< 0.001	0.0047	0.3918	< 0.001
	\(U + T + S\)	< 0.001	< 0.001	1	< 0.001	< 0.001	< 0.001	< 0.001	< 0.001	0.0018	< 0.001	< 0.001	< 0.001
RF	\(T + S\)	< 0.001	< 0.001	< 0.001	1	< 0.001	1	< 0.001	< 0.001	< 0.001	< 0.001	< 0.001	< 0.001
	U	< 0.001	< 0.001	< 0.001	< 0.001	1	< 0.001	< 0.001	< 0.001	< 0.001	< 0.001	< 0.001	< 0.001
	\(U + T + S\)	< 0.001	< 0.001	< 0.001	1	< 0.001	1	< 0.001	< 0.001	< 0.001	< 0.001	< 0.001	< 0.001
Fusion-CNN	\(T + S\)	< 0.001	0.5644	< 0.001	< 0.001	< 0.001	< 0.001	1	0.5428	< 0.001	< 0.001	0.6591	< 0.001
	U	< 0.001	0.7432	< 0.001	< 0.001	< 0.001	< 0.001	0.5428	1	< 0.001	< 0.001	0.232	< 0.001
	\(U + T + S\)	< 0.001	< 0.001	0.0018	< 0.001	< 0.001	< 0.001	< 0.001	< 0.001	1	< 0.001	< 0.001	0.6062
Fusion-LSTM	\(T + S\)	< 0.001	0.0047	< 0.001	< 0.001	< 0.001	< 0.001	< 0.001	< 0.001	< 0.001	1	0.0011	< 0.001
	U	< 0.001	0.3918	< 0.001	< 0.001	< 0.001	< 0.001	0.6591	0.232	< 0.001	0.0011	1	< 0.001
	\(U + T + S\)	< 0.001	< 0.001	< 0.001	< 0.001	< 0.001	< 0.001	< 0.001	< 0.001	0.6062	< 0.001	< 0.001	1

ISSN: 1472-6947