Figure 3 | Scientific Reports

Figure 3

From: Nuquantus: Machine learning software for the characterization and quantification of cell nuclei in complex immunofluorescent tissue images

Figure 3

Performance of the swine and mouse prediction models.

(a) Swine model obtained precision recall curve that describes the trade-off between precision and recall. This model was trained on 88 cardiac images of post MI tissue obtained from N = 4 swine. (b) Mouse model obtained precision recall curve. The training was performed on 84 cardiac images of post MI tissue obtained from N = 19 mice. The contribution of VZ, BZ and IA images to the prediction model is shown by dashed lines. The overall performance of all cardiac sub regions is shown in red continuous line. (c) Example of swine prediction model on unseen images. The prediction was compared to the ground truth labels to identify true positive versus true negatives, false positive and false negative CM nuclei.

Back to article page