Table 5 Confusion matrix formulas.
From: An AI-based automatic leukemia classification system utilizing dimensional Archimedes optimization
Measure | Formula | Definition |
|---|---|---|
Precision (P) | \(\frac{TP}{{\left( {TP + FP} \right)}}\) | The accuracy rate of positive predictions |
Specificity | \(\frac{TN}{{\left( {TN + FP} \right)}}\) | Specificity refers to the capacity to accurately identify individuals who do not have a particular disease as negative |
Recall/Sensitivity (R) | \(\frac{TP}{{\left( {TP + FN} \right)}}\) | The proportion of correctly identified positive instances among all instances labed as positive |
Accuracy(A) | \(\frac{{\left( {TP + TN} \right) }}{{ \left( {TP + TN + FP + FN} \right)}}\) | The accuracy ratepredictions |
ror(E) | \(1 - A\) | The percentage of inaccurate predictions |
F-measure | \(\frac{2*PR}{{\left( {P + R} \right)}}\) | Recall and precision are weighted harmonically averaged |
Dice similarity coefficient (DSC) | \(DSC = \frac{2TP}{{\left( {2TP + FP + FN} \right)}}\) | DSC is a statistical measure used to quantify the similarity between two samples |
Jaccard Index (JI) | \(JI = \frac{{\left( {TL \cap PL} \right)}}{{\left( {TL \cup PL} \right)}}\) | JI is a quantitative measure, ranging from 0 to 1 (equivalent to 0% to 100%), which assesses the similarity between two sets of samples |
