Skip to main content

Table 5 Model performance on the test set. MCC: Matthews Correlation Coefficient; Acc: accuracy; \(F_1\): \(F_1\)-score; \(\mu\) mean of the metric over \(10^{5}\) bootstrap resampling; \(\rho\) radius of the corresponding 5%-95% CI

From: Endoscopy-based IBD identification by a quantized deep learning pipeline

Category

MCC

Acc

\(F_1\)

\(\mu\)

\(\rho\)

\(\mu\)

\(\rho\)

\(\mu\)

\(\rho\)

per-patient

0.90

\(5\cdot 10^{-4}\)

0.95

\(2\cdot 10^{-4}\)

0.96

\(2\cdot 10^{-4}\)

per-image

0.84

\(2\cdot 10^{-4}\)

0.94

\(10^{-4}\)

0.94

\(7\cdot 10^{-5}\)