Using sequential patterns as features for classification models to make accurate predictions on ICU events

Pattern mining algorithms have previously been utilized to extract informative rules in various clinical contexts. However, the number of generated patterns are numerous. In most cases, the extracted rules are directly investigated by clinicians for understanding disease diagnoses. The elicitation of important patterns for clinical investigation places a significant demand for precision and interpretability. Hence, it is essential to obtain a set of informative interpretable patterns for building advanced learning models about a patient's physiological condition, specially in critical care units. In this study, a two stage sequential contrast patterns based classification framework is presented, which is used to detect critical patient events like hypotension. In the first stage, we obtain a set of sequential patterns by using a contrast mining algorithm. These sequential patterns undergo post-processing, for conversion to binary valued and frequency based features for developing a classification model, in the second stage. Our results on eight critical care datasets demonstrate better predictive capabilities, when sequential patterns are used as features.