Skip to main content
. 2022 Jul 13;13:4050. doi: 10.1038/s41467-022-31339-8

Fig. 5. Automated diagnosis with deep convolutional neural networks (CNNs) on femto-SRH images.

Fig. 5

a Schematics of CNN predictions; b, c Training, validation and test results of the neural network for the classification of non-cancer/cancer tissues; d, e Training, validation and test results of the neural network for the classification of differentiated/undifferentiated cancers. f Confusion matrix of the three diagnostic subtypes. Scale bar: 200 µm. Source data are provided in the Source data file.