Different evaluation measures assess different characteristics of machine learning algorithms. The empirical evaluation of algorithms and classifiers is a matter of on-going debate between researchers. Although most measures in use today focus on a classifier's ability to identify classes correctly, we suggest that, in certain cases, other properties, such as failure avoidance or class discrimination may also be useful. We suggest the application of measures which evaluate such properties. These measures -Youden's index, likelihood, Discriminant power -are used in medical diagnosis. We show that these measures are interrelated, and we apply them to a case study from the field of electronic negotiations. We also list other learning problems which may benefit from the application of the proposed measures.
Respiratory failure in the acute respiratory syndrome coronavirus 2 (SARS-CoV-2) pandemic is hypothesized to be driven by an overreacting innate immune response, where the complement system is a key player. In this prospective cohort study of 39 hospitalized coronavirus disease COVID-19 patients, we describe systemic complement activation and its association with development of respiratory failure. Clinical data and biological samples were obtained at admission, days 3 to 5, and days 7 to 10. Respiratory failure was defined as PO2/FiO2 ratio of ≤40 kPa. Complement activation products covering the classical/lectin (C4d), alternative (C3bBbP) and common pathway (C3bc, C5a, and sC5b-9), the lectin pathway recognition molecule MBL, and antibody serology were analyzed by enzyme-immunoassays; viral load by PCR. Controls comprised healthy blood donors. Consistently increased systemic complement activation was observed in the majority of COVID-19 patients during hospital stay. At admission, sC5b-9 and C4d were significantly higher in patients with than without respiratory failure (P = 0.008 and P = 0.034). Logistic regression showed increasing odds of respiratory failure with sC5b-9 (odds ratio 31.9, 95% CI 1.4 to 746, P = 0.03) and need for oxygen therapy with C4d (11.7, 1.1 to 130, P = 0.045). Admission sC5b-9 and C4d correlated significantly to ferritin (r = 0.64, P < 0.001; r = 0.69, P < 0.001). C4d, sC5b-9, and C5a correlated with antiviral antibodies, but not with viral load. Systemic complement activation is associated with respiratory failure in COVID-19 patients and provides a rationale for investigating complement inhibitors in future clinical trials.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.