EMOCS: evolutionary multi-objective optimisation for clinical scorecard generation

摘要: Clinical scorecards of risk factors associated with disease severity or mortality outcome are used by clinicians to make treatment decisions and optimize resources. This study develops an automated tool framework based on evolutionary algorithms for the derivation from clinical data. The techniques employed NSGA-II Multi-objective Optimization Genetic Algorithm (GA) which optimizes Pareto-front two clinically-relevant scorecard objectives, size accuracy. Three methods presented improve previous manually derived scorecards. first is a hybrid algorithm uses GA feature selection decision tree generation. In second, generates full scorecard. third extended scoring system in also scores. this combinations features thresholds each point selected process discover near-optimal Pareto-fronts exploration expert makers. shown produce that upon human example C.Difficile, important infection found globally communities hospitals, although described applicable any where required data available.