Virtual histological staining of unlabelled tissue-autofluorescence images via deep learning
By Yair Rivenson, Hongda Wang, Zhensong Wei, K. de Haan, Yibo Zhang, Yichen Wu, Harun Günaydın, J. Zuckerman, Thomas Chong, A. Sisk and 3 more. Nature Biomedical Engineering, 2018
612 citations recorded by the source corpus.
The histological analysis of tissue samples, widely used for disease diagnosis, involves lengthy and laborious tissue preparation. Here, we show that a convolutional neural network trained using a generative adversarial-network model can transform wide-field autofluorescence images of unlabelled tissue sections into images that are equivalent to the bright-field images of histologically stained versions of the same samples. A blind comparison, by board-certified pathologists, of this virtual staining method and standard histological staining using microscopic images of human tissue sections of the salivary gland, thyroid, kidney, liver and lung, and involving different types of stain, showed no major discordances. The virtual-staining method bypasses the typically labour-intensive and costly histological staining procedures, and could be used as a blueprint for the virtual staining of tissue images acquired with other label-free imaging modalities. Deep learning can be used to virtually stain autofluorescence images of unlabelled tissue sections, generating images that are equivalent to the histologically stained versions.
Subfields: Biophysics, Media Technology
View Paper: https://www.semanticscholar.org/paper/15b087d4d92aa5452046002328add94850ea86a3
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Biophysics
In Media Technology
Citations in the corpus
- Large-scale pancreatic cancer detection via non-contrast CT and deep learning (2023, positive): “Unlike generative deep learning methods to synthesize contrast or color 18–21 , we train supervised learning models, which effectively capture subtle image details and directly learn downstream lesion detection and diagnosis tasks based on these detailed characteristics.”
- Inference in artificial intelligence with deep optics and photonics (2020, background): “Applications of deep learning in optical microscopy cover brightfield microscopy 80 , lensless microscopy 72,81,87 , fluorescence microscopy 82,83,85,86,88,89 , super-resolution microscopy 80,86,90–92 , confocal microscopy 85,86,93 , structured illumination microscopy 86 and many others 84,94–96 .”
- Deep learning-based transformation of H&E stained tissues into special stains (2020, neutral): “ldiscriminator = D(G(xinput, ?̃?), ?̃?) 2 + (1 − D(zlabel , ?̃?)) 2 (7)”
- Deep learning-enabled virtual histological staining of biological samples (2022, methodology): “For instance, to avoid intensity variations caused by potential photobleaching in autofluorescence imaging, Rivenson et al. normalized the input autofluorescence images by subtracting the mean value across the entire tissue slide and dividing it by the standard deviation of the pixel values18.”