Table 4 Diagnostic metrics used to evaluate the performance of the investigated CNN architectures.
From: Diagnostic performance of convolutional neural networks for dental sexual dimorphism
Metrics | Description |
|---|---|
Loss | A loss function indicates how well the model assimilates the dataset. The loss function will output a higher value if the predictions are off the actual target. Since our problem/question relies on a multi-class classification, we used cross-entropy within our loss function |
Accuracy | The accuracy of a machine learning classification algorithm is one way to measure how often the algorithm classifies a data point correctly. This can be understood as the number of items correctly identified as either true positive or true negative out of the total number of items |
F1-score | Represents the average of precision and recall and measures the effectiveness of identification when recall and precision have balanced importance |
Precision | Agreement of true class labels with machine’s predictions. It is calculated by summing all true positives and false positives in the system, across all classes |
Recall | Effectiveness of a classifier to identify class labels. It is calculated by summing all true positives and false negatives in the system, across all classes |
Specificity | Known as the true negative rate. This function calculates the proportion of actual negative cases that have gotten predicted as negative by our model |
