diff --git a/book.org b/book.org
index 0d88bfe..4ec176b 100644
--- a/book.org
+++ b/book.org
@@ -2439,17 +2439,17 @@ The latter is also called sensitivity if used in combination with specificity (E
 
 #+NAME: eq:precision
 \begin{equation}
-\frac{TP}{TP+FP}
+\frac{\mathit{TP}}{\mathit{TP}+\mathit{FP}}
 \end{equation}
 
 #+NAME: eq:recall
 \begin{equation}
-\frac{TP}{TP+FN}
+\frac{\mathit{TP}}{\mathit{TP}+\mathit{FN}}
 \end{equation}
 
 #+NAME: eq:specificity
 \begin{equation}
-\frac{TN}{TN+FP}
+\frac{\mathit{TN}}{\mathit{TN}+\mathit{FP}}
 \end{equation}
 
 Many studies for pass/fail prediction use accuracy (Equation\nbsp{}[[eq:accuracy]]) as a single performance metric.
@@ -2459,7 +2459,7 @@ This is clearly a bad classifier, but it will nonetheless have an accuracy of 75
 
 #+NAME: eq:accuracy
 \begin{equation}
-\frac{TP+TN}{TP+TN+FP+FN}
+\frac{\mathit{TP}+\mathit{TN}}{\mathit{TP}+\mathit{TN}+\mathit{FP}+\mathit{FN}}
 \end{equation}
 
 In our study, we will therefore use two more complex metrics that take these effects into account: balanced accuracy and F_1-score.