|
1. A computer-implemented method for determining an intervention relatedness measure for a predictive entity with respect to a target intervention, the computer-implemented method comprising: identifying, by one or more processors, an ordered sequence of one or more event codes associated with the predictive entity; determining, by the one or more processors and a cross-temporal encoding machine learning model and based at least in part on the ordered sequence, a cross-temporal encoding of the predictive entity based at least in part on the ordered sequence; determining, by the one or more processors and based at least in part on the cross-temporal encoding of the predictive entity and a target intervention cross-temporal encoding for the target intervention, a cross-temporal similarity measure for the predictive entity; determining, by the one or more processors and based at least in part on the cross-temporal similarity measure, the intervention relatedness measure; and performing, by the one or more processors, one or more prediction-based actions based at least in part on the intervention relatedness measure, wherein performing the one or more prediction-based actions comprises: determining whether the intervention relatedness measure satisfies an intervention relatedness measure threshold; and in response to determining that the intervention relatedness measure satisfies the intervention relatedness measure threshold, performing the one or more prediction-based actions based at least in part on a clinical intervention associated with the target intervention.
|
