Table 3.

Area under the curve (AUC) values by different models (–time: time information was excluded and +time: time information was included).

Method	AUCa–time	AUC+time
LRb	0.6447	0.6773
MLPc	0.6545	0.6753
LSTMd	0.6045	0.6567
ALSTMe	0.6346	0.6714
TLSTMf	—	0.6548
ATTAINg	0.6119	0.6597
RETAINh	0.6455	0.6882
TSANNi-I	0.6692	0.7003 j
TSANN-I-step	0.6463	—
TSANN-II	0.6827	0.6855

^aAUC: area under the receiver operating curve.

^bLR: logistic regression.

^cMLP: multilayer perceptron.

^dLSTM: long short-term memory.

^eALSTM: attention long short-term memory.

^fTLSTM: time-aware long short-term memory.

^gATTAIN: Attention-based Time-aware Disease Progression.

^hRETAIN: Reverse Time Attention model.

ⁱTSANN: time-sensitive attentive neural network.

^jThe optimal value for each column is italicized.