Fig. 5: Summary of quantum vs. classical performance metrics for synthesized evidence.
From: A systematic review of quantum machine learning for digital health
Selected test scores (accuracy, AUC and squared error ∈ [0, 1]) reported for quantum interventions deployed under realistic operating conditions using health data. Ranges reflected min and max test scores attained over different experimental configurations reported in each study; error bars for each test score are shown if originally reported. Synthesized studies with strong (weak) consensus are shaded in green (gray). Best possible scores for accuracy, fidelity and AUC metrics (unity) differ from squared error metrics (zero). Performance benchmarks that cannot be easily compared or interpreted alongside other metrics, or are not consistently reported, e.g. F1 scores and precision, are excluded for readability. NA Not applicable. NR Not reported. Est. Estimates of numerical values implied from visual/graphical data.
