Learning dynamic treatment strategies for coronary heart diseases by artificial intelligence: real-world data-driven study

Table 2 Performance comparison on the test dataset

Method	Estimated mortality		Jaccard
Method	Trajectory-wise	State-wise	Jaccard
Clinician’s policy	0.0956	0.0959	–
Dual-LSTM	0.0887	0.0935	0.3171
AMANet	0.0827	0.0895	0.3250
DPO-LSTM	0.0919	0.0930	0.2069
SRL-Multimorbidity	0.0948	0.0953	0.2610
SL-LSTM (\(\varepsilon =0\))	0.0956	0.0967	0.3432
RL-LSTM (\(\varepsilon =1.0\))	0.0752	0.0721	0.0342
SRL-LSTM (\({\varvec{\varepsilon}}=0.4\))	0.0643	0.0878	0.3110
SRL-LSTM(\(\varepsilon =0.4\), w/o diagnosis codes)	0.0741	0.086	0.2233

The bold indicates the prefered AI model to learn dynamic treatment strategies for CHD patients

ISSN: 1472-6947