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

Diversity of H and E stained images illustrated using four prostate cancer samples with Gleason Grade 3. The first two samples show range of epithelial brightness, and the last two show the range of stromal brightness
© Copyright Policy - open-access
Related In: Results  -  Collection

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

Figure 1: Diversity of H and E stained images illustrated using four prostate cancer samples with Gleason Grade 3. The first two samples show range of epithelial brightness, and the last two show the range of stromal brightness

Mentions: Some color variation can be useful in classifying images because such variation might reflect important contrasts in the underlying biochemical composition of the tissue. However, images of similar tissues that are colored using the same stain also suffer from unwanted color variation due to differences in stain manufacturing processes across vendors, staining protocols across labs, and color responses across digital scanners. This is especially true for the H and E stain that is universally used by surgical pathologists to reveal histopathological detail. Hematoxylin itself is a natural product extracted from logwood trees; standardization across batches is, therefore, difficult and the dye is prone to precipitation in storage, which can cause day-to-day variation even within a single lab.[345] In addition, the handling of the specimen during fixation and processing can alter the way in which the tissue interacts with the dyes, producing extraneous variation even in tissue microarray (TMA) cores stained on the same slide. Figure 1 shows an example of this diversity of stain appearances, among prostate cancer TMA cores scanned on the same digital microscope. The two cores shown in the left half of Figure 1 show extremes of epithelial appearance whereas the two cores in the right half show extremes of stromal appearance. Such variation in stain appearance can be problematic for algorithms in computational pathology that rely on tissue color.


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)

Diversity of H and E stained images illustrated using four prostate cancer samples with Gleason Grade 3. The first two samples show range of epithelial brightness, and the last two show the range of stromal brightness
© Copyright Policy - open-access
Related In: Results  -  Collection

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

Figure 1: Diversity of H and E stained images illustrated using four prostate cancer samples with Gleason Grade 3. The first two samples show range of epithelial brightness, and the last two show the range of stromal brightness
Mentions: Some color variation can be useful in classifying images because such variation might reflect important contrasts in the underlying biochemical composition of the tissue. However, images of similar tissues that are colored using the same stain also suffer from unwanted color variation due to differences in stain manufacturing processes across vendors, staining protocols across labs, and color responses across digital scanners. This is especially true for the H and E stain that is universally used by surgical pathologists to reveal histopathological detail. Hematoxylin itself is a natural product extracted from logwood trees; standardization across batches is, therefore, difficult and the dye is prone to precipitation in storage, which can cause day-to-day variation even within a single lab.[345] In addition, the handling of the specimen during fixation and processing can alter the way in which the tissue interacts with the dyes, producing extraneous variation even in tissue microarray (TMA) cores stained on the same slide. Figure 1 shows an example of this diversity of stain appearances, among prostate cancer TMA cores scanned on the same digital microscope. The two cores shown in the left half of Figure 1 show extremes of epithelial appearance whereas the two cores in the right half show extremes of stromal appearance. Such variation in stain appearance can be problematic for algorithms in computational pathology that rely on tissue color.

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