Fig. 5
From: Risk-based evaluation of machine learning-based classification methods used for medical devices
Impact of risk-based approach – graphical comparison. Graphical representation of the results given in Table 3. Left side: Visualization of the expected risk (\(\:\stackrel{\sim}{ER}\)) values for the particular risk ratios \(\:{c}_{FN}\) in the range \(\:\frac{1}{8}={2}^{-3}\) to \(\:8={2}^{3}\). The same artificial model as given in Equation (1) and (2) with \(\:{\sigma\:}_{FP}={\sigma\:}_{FN}=0.3\) was used