Table 1.

Overview of different objectives in artificial intelligence–based trajectory analysis.

Objective	Description	Examples	Selected references
Risk scoring	The objective is to estimate the likelihood of future health outcomes (eg, mortality, readmission, and adverse drug reactions)	Predict the 10-year risk of developing coronary heart disease for patients as in the Framingham risk score Predict the need for an intensive care unit in an emergency ward through measurements from wearables	[3,17,23,26-32]
Subtyping	The objective is to cluster the patient cohort into different disease dynamics (ie, subtyping) while accounting for the longitudinal form of patient trajectories	Cluster disease progressions into “recurrent course” and “progressive decline”	[26]
Pathway discovery	The objective is to detect clinically meaningful subpatterns in patient trajectories	Identify frequent patterns in patient trajectories that are indicative of disease onset	[1,33,34]