Skip to main content
Fig. 4 | BMC Biology

Fig. 4

From: Morphology-based deep learning enables accurate detection of senescence in mesenchymal stem cell cultures

Fig. 4

Cascade R-CNN system performance at doxorubicin-induced iMSC senescence. A Protocol for the development of the Cascade R-CNN in doxorubicin-induced iMSC senescence. B The AP, mAP, and AR showed the performance of the trained Cascade R-CNN trained in doxorubicin data. C The mAP and AR for small, medium, and large objects in doxorubicin data. D The precision, recall, and F1 score for non-senescent and senescent cells detection in doxorubicin data. E A heatmap showed the mAP of Cascade R-CNN prediction in each test dataset. F Linear correlation between the senescence proportion output of Cascade R-CNN trained by doxorubicin data and the doxorubicin duration. G The Pearson correlation coefficient between the senescence proportion output of the Cascade R-CNN trained by doxorubicin data and senescence-related indicators. Data were representative of three independent experiments, n = 3

Back to article page