Limits...
Empirical comparison of color normalization methods for epithelial-stromal classification in H and E images.

Sethi A, Sha L, Vahadane AR, Deaton RJ, Kumar N, Macias V, Gann PH - J Pathol Inform (2016)

Bottom Line: For the main MRS method, which relied on classification of super-pixels, the number of variables used was reduced using backward elimination without compromising accuracy, and test - area under the curves (AUCs) were compared for original and normalized images.Khan method reduced color saturation while Vahadane reduced hue variance.Color normalization can give a small incremental benefit when a super-pixel-based classification method is used with features that perform implicit color normalization while the gain is higher for patch-based classification methods for classifying epithelium versus stroma.

View Article: PubMed Central - PubMed

Affiliation: Department of Electronics and Electrical Engineering, Indian Institute of Technology Guwahati, Guwahati, Assam, India; Department of Pathology, University of Illinois, Chicago, IL, USA.

ABSTRACT

Context: Color normalization techniques for histology have not been empirically tested for their utility for computational pathology pipelines.

Aims: We compared two contemporary techniques for achieving a common intermediate goal - epithelial-stromal classification.

Settings and design: Expert-annotated regions of epithelium and stroma were treated as ground truth for comparing classifiers on original and color-normalized images.

Materials and methods: Epithelial and stromal regions were annotated on thirty diverse-appearing H and E stained prostate cancer tissue microarray cores. Corresponding sets of thirty images each were generated using the two color normalization techniques. Color metrics were compared for original and color-normalized images. Separate epithelial-stromal classifiers were trained and compared on test images. Main analyses were conducted using a multiresolution segmentation (MRS) approach; comparative analyses using two other classification approaches (convolutional neural network [CNN], Wndchrm) were also performed.

Statistical analysis: For the main MRS method, which relied on classification of super-pixels, the number of variables used was reduced using backward elimination without compromising accuracy, and test - area under the curves (AUCs) were compared for original and normalized images. For CNN and Wndchrm, pixel classification test-AUCs were compared.

Results: Khan method reduced color saturation while Vahadane reduced hue variance. Super-pixel-level test-AUC for MRS was 0.010-0.025 (95% confidence interval limits ± 0.004) higher for the two normalized image sets compared to the original in the 10-80 variable range. Improvement in pixel classification accuracy was also observed for CNN and Wndchrm for color-normalized images.

Conclusions: Color normalization can give a small incremental benefit when a super-pixel-based classification method is used with features that perform implicit color normalization while the gain is higher for patch-based classification methods for classifying epithelium versus stroma.

No MeSH data available.


Related in: MedlinePlus

Color normalization illustrated using inter-image standard deviation (error bars) of mean (bars) hue, saturation, and intensity for epithelium and stroma. Continuous color bars between epithelium and stroma illustrate the hue, saturation, and intensity range holding the other two at their means
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4837797&req=5

Figure 3: Color normalization illustrated using inter-image standard deviation (error bars) of mean (bars) hue, saturation, and intensity for epithelium and stroma. Continuous color bars between epithelium and stroma illustrate the hue, saturation, and intensity range holding the other two at their means

Mentions: The R, G, and B color channels are highly correlated in H and E stained images. Therefore, it is more insightful to examine the distribution of pixels in the HSI space, where H refers to hue, S refers to saturation, and I refer to intensity (brightness). This is closer to how humans perceive color. We expected color normalization to reduce inter-image variance of color measures. Across 30 images, as seen in Figure 3, variances of mean pixel intensity and saturation for each image were reduced significantly in images color-normalized using Khan. Moreover, the saturation was much lower for Khan, especially in epithelium. On the other hand, Vahadane reduced the variance of hue significantly while leaving the variance of intensity and saturation almost same as the original. This implies that while the brightness and saturation were normalized by Khan method, Vahadane mainly normalized hue. Further, mean saturation was significantly reduced by Khan method, especially for epithelium, giving an overall grayish appearance to each image.


Empirical comparison of color normalization methods for epithelial-stromal classification in H and E images.

Sethi A, Sha L, Vahadane AR, Deaton RJ, Kumar N, Macias V, Gann PH - J Pathol Inform (2016)

Color normalization illustrated using inter-image standard deviation (error bars) of mean (bars) hue, saturation, and intensity for epithelium and stroma. Continuous color bars between epithelium and stroma illustrate the hue, saturation, and intensity range holding the other two at their means
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4837797&req=5

Figure 3: Color normalization illustrated using inter-image standard deviation (error bars) of mean (bars) hue, saturation, and intensity for epithelium and stroma. Continuous color bars between epithelium and stroma illustrate the hue, saturation, and intensity range holding the other two at their means
Mentions: The R, G, and B color channels are highly correlated in H and E stained images. Therefore, it is more insightful to examine the distribution of pixels in the HSI space, where H refers to hue, S refers to saturation, and I refer to intensity (brightness). This is closer to how humans perceive color. We expected color normalization to reduce inter-image variance of color measures. Across 30 images, as seen in Figure 3, variances of mean pixel intensity and saturation for each image were reduced significantly in images color-normalized using Khan. Moreover, the saturation was much lower for Khan, especially in epithelium. On the other hand, Vahadane reduced the variance of hue significantly while leaving the variance of intensity and saturation almost same as the original. This implies that while the brightness and saturation were normalized by Khan method, Vahadane mainly normalized hue. Further, mean saturation was significantly reduced by Khan method, especially for epithelium, giving an overall grayish appearance to each image.

Bottom Line: For the main MRS method, which relied on classification of super-pixels, the number of variables used was reduced using backward elimination without compromising accuracy, and test - area under the curves (AUCs) were compared for original and normalized images.Khan method reduced color saturation while Vahadane reduced hue variance.Color normalization can give a small incremental benefit when a super-pixel-based classification method is used with features that perform implicit color normalization while the gain is higher for patch-based classification methods for classifying epithelium versus stroma.

View Article: PubMed Central - PubMed

Affiliation: Department of Electronics and Electrical Engineering, Indian Institute of Technology Guwahati, Guwahati, Assam, India; Department of Pathology, University of Illinois, Chicago, IL, USA.

ABSTRACT

Context: Color normalization techniques for histology have not been empirically tested for their utility for computational pathology pipelines.

Aims: We compared two contemporary techniques for achieving a common intermediate goal - epithelial-stromal classification.

Settings and design: Expert-annotated regions of epithelium and stroma were treated as ground truth for comparing classifiers on original and color-normalized images.

Materials and methods: Epithelial and stromal regions were annotated on thirty diverse-appearing H and E stained prostate cancer tissue microarray cores. Corresponding sets of thirty images each were generated using the two color normalization techniques. Color metrics were compared for original and color-normalized images. Separate epithelial-stromal classifiers were trained and compared on test images. Main analyses were conducted using a multiresolution segmentation (MRS) approach; comparative analyses using two other classification approaches (convolutional neural network [CNN], Wndchrm) were also performed.

Statistical analysis: For the main MRS method, which relied on classification of super-pixels, the number of variables used was reduced using backward elimination without compromising accuracy, and test - area under the curves (AUCs) were compared for original and normalized images. For CNN and Wndchrm, pixel classification test-AUCs were compared.

Results: Khan method reduced color saturation while Vahadane reduced hue variance. Super-pixel-level test-AUC for MRS was 0.010-0.025 (95% confidence interval limits ± 0.004) higher for the two normalized image sets compared to the original in the 10-80 variable range. Improvement in pixel classification accuracy was also observed for CNN and Wndchrm for color-normalized images.

Conclusions: Color normalization can give a small incremental benefit when a super-pixel-based classification method is used with features that perform implicit color normalization while the gain is higher for patch-based classification methods for classifying epithelium versus stroma.

No MeSH data available.


Related in: MedlinePlus