Figure 4 | Scientific Reports

Figure 4

From: Unsupervised segmentation of low-contrast multichannel images: discrimination of tissue components in microscopic images of unstained specimens

Figure 4

Human liver with metastasis of gastric cancer.

(a) red-green-blue (RGB) color microscopic image of unstained specimen. μ(A) = 0.9999, μ_average(A) = 0.9988. (b) “ground truth – subsequent slides” RGB color microscopic image of the specimen stained by Hep Par. (c) color-coded (digitally stained) image of the segmentation result obtained by EKM-NMF_L0 algorithm (Gaussian kernel variance = 0.1 and D = 50): blue: metastasis of gastric cancer, red: border area of inflammation, green: hepatocytes. μ(B) = 0.9994, μ_average(B) = 0.9917. (d) RGB color microscopic image of specimen (a) stained subsequently by H&E. (e) “ground truth – subsequent slides” RGB color microscopic image of the specimen stained by CDX2. (f) “ground truth - subsequent slides” RGB color microscopic image of the specimen stained by LCA.

Back to article page