A Nonlinear Mapping Approach to Stain Normalization in Digital Histopathology Images Using Image-Specific Color Deconvolution

By A. Khan, Nasir M. Rajpoot, D. Treanor, D. Magee. IEEE Transactions on Biomedical Engineering, 2014

503 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Biophysics, Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/0eac0ed2c87910dbb7d9f622854efb7c7b7b6f0b

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Anant Madabhushi
2. J. Urgen Schmidhuber
3. Lin F. Yang
4. Fuyong Xing
5. Anne L. Martel
6. Ozan Ciga
7. Nassir Navab
8. Luca Maria Gambardella

In Biophysics

1. Fuyong Xing
2. Lin F. Yang
3. Mattias Rantalainen
4. Johan Hartman
5. Balazs Acs
6. Jakob Nikolas Kather
7. Faisal Mahmood
8. Anne L. Martel

In Computer Vision and Pattern Recognition

1. J. Urgen Schmidhuber
2. Luc Van Gool
3. Lin F. Yang
4. Dacheng Tao
5. Andrew Zisserman
6. Xiaogang Wang
7. Trevor Darrell
8. Alan L. Yuille

Citations in the corpus

• Locality Sensitive Deep Learning for Detection and Classification of Nuclei in Routine Colon Cancer Histology Images (2016, methodology): “The hematoxylin intensity was obtained using a recently proposed color deconvolution method [35].”
• Quantifying the effects of data augmentation and stain color normalization in convolutional neural networks for computational pathology (2019, methodology): “More recent methods use machine learning algorithms to detect certain morphological structures, e.g. cell nuclei, that are associated with certain stains, improving the result of the normalization process (Khan et al. (2014); Bejnordi et al. (2016)).”
• Deep neural network models for computational histopathology: A survey (2019, methodology): “The uniformity assumption is relaxed in more recent ML methods, where the type of chemical staining and morphological properties of an image are considered in generating stain matrices (Khan et al., 2014; Bejnordi et al., 2015).”
• New Colors for Histology: Optimized Bivariate Color Maps Increase Perceptual Contrast in Histological Images (2015, background): “combined a variant of color deconvolution with mapping to a new color space to achieve a standardized aspect of different histological images [16].”
• Large scale tissue histopathology image classification, segmentation, and visualization via deep convolutional activation features (2017, background): “Khan [53] proposes a nonlinear mapping approach to normalize staining.”
• The impact of pre- and post-image processing techniques on deep learning frameworks: A comprehensive review for digital pathology image analysis (2020, methodology): “The automated extraction of the matrix W has been done by using: (i) prior information of the stain vectors [52], (ii) singular value decomposition [51,53] (iii) support vector machines [25], (iv) gaussian mixture model [54] and (v) cellular structures segmentation [55 – 57].”
• Assessment of algorithms for mitosis detection in breast cancer histopathology images (2014, methodology): “Preprocessing: Staining normalization by non-linear color mapping (Khan et al., 2014) was performed in order to neutralize the inherent variation in the color of the staining.”
• Deep Adversarial Training for Multi-Organ Nuclei Segmentation in Histopathology Images (2018, methodology): “Normalizing the images can significantly improve the performance of these methods [55].”
• Deep learning-based breast cancer classification through medical imaging modalities: state of the art and research challenges (2019, methodology): “To eliminate these inconsistencies, previous studies used RGB histogram specification, Reinhard’s (Reinhard et al. 2001), Macenko’s (Macenko et al. 2009), and Khan’s methods (Khan et al. 2014) to normalize the HP images before classification (Fig.”
• Artificial intelligence as the next step towards precision pathology (2020, methodology): “During this time, methods have been developed to decrease variability in image quality, for example colour standardization, spatial filtering, denoising or enhancement [44].”
• Stain Specific Standardization of Whole-Slide Histopathological Images (2016, unknown): “vectors [9], [12].”
• Robust Nucleus/Cell Detection and Segmentation in Digital Pathology and Microscopy Images: A Comprehensive Review (2016, methodology): “The preprocessing techniques including color normalization and image denoising, which are presented in [15], [21], and [22], and extraction of regions of interest, which”
• Machine Learning Methods for Histopathological Image Analysis (2017, background): “[68] Khan AM, Rajpoot N, Treanor D, Magee D.”
• Structure-Preserving Color Normalization and Sparse Stain Separation for Histological Images (2016, methodology): “This creates difficulty in image interpretation by software trained on a particular stain appearance [3], [4], and further adds to already existing inter- and intra-expert variance in diagnosis and labeling among pathologists [5].”
• Staingan: Stain Style Transfer for Digital Histological Images (2018, methodology): “Our proposed StainGAN method shows a relative improvement of ( 80% , 36% , 22% , 15% , and 3% ) on AUC over the Un-normalized, Reinhard, Macenko, Khan, and Vahadane, respectively.”