. Author manuscript; available in PMC: 2024 Jul 31.

Published in final edited form as: J Biomed Inform. 2020 Dec 31;115:103671. doi: 10.1016/j.jbi.2020.103671

Table 2.

Correspondence between patient data types and deep learning models.

	Patient Input Data Type
DL Models	Structured	Unstructured	Combined
LSTM	[14,58,61,64–68]	[69]	[6,54,55,57]
CNN	[14,33,66,68,70–72]	[63,73]	[54,55]
GRU	[31,38,72,74–81]
Autoencoder	[61,67,74,75,82,83]	[84]	[11]
word2vec	[59,85–87]		[56]
FFNN	[88,89]		[6,54]
Tensor Decomposition	[48,58]
Deep Averaging Network		[62,90]
Graph NN	[27,32,91]
Restricted Boltzmann Machines	[60]
Transformer	[43]