Multi-class texture analysis in colorectal cancer histology.
Ontology highlight
ABSTRACT:
SUBMITTER: Kather JN
PROVIDER: S-EPMC4910082 | biostudies-other | 2016
REPOSITORIES: biostudies-other
Similar Datasets
Project description:PurposeThis paper proposes to characterize the continuum of colorectal cancer (CRC) using multiple texture features extracted from multispectral optical microscopy images. Three types of pathological tissues (PT) are considered: benign hyperplasia, intraepithelial neoplasia and carcinoma.Materials and methodsIn the proposed approach, the region of interest containing PT is first extracted from multispectral images using active contour segmentation. This region is then encoded using texture features based on the Laplacian-of-Gaussian (LoG) filter, discrete wavelets (DW) and gray level co-occurrence matrices (GLCM). To assess the significance of textural differences between PT types, a statistical analysis based on the Kruskal-Wallis test is performed. The usefulness of texture features is then evaluated quantitatively in terms of their ability to predict PT types using various classifier models.ResultsPreliminary results show significant texture differences between PT types, for all texture features (p-value < 0.01). Individually, GLCM texture features outperform LoG and DW features in terms of PT type prediction. However, a higher performance can be achieved by combining all texture features, resulting in a mean classification accuracy of 98.92%, sensitivity of 98.12%, and specificity of 99.67%.ConclusionsThese results demonstrate the efficiency and effectiveness of combining multiple texture features for characterizing the continuum of CRC and discriminating between pathological tissues in multispectral images.
| S-EPMC4764026 | biostudies-literature
Project description:Colorectal cancer (CRC) constitutes the third most commonly diagnosed cancer in males and the second in females. Precise histopathological classification of CRC tissue pathology is the cornerstone not only for diagnosis but also for patients' management decision making. An automated system able to accurately classify different CRC tissue regions may increase diagnostic precision and alleviate clinical workload. However, tissue classification is a challenging task due to the variability in morphological and textural characteristics present in histopathology images. In this study, an artificial neural network was trained to classify between eight classes of CRC tissue image patches derived from a public dataset with 5000 CRC histopathology image tiles. A total of 532 multi-level pathomics features examined at different scales were extracted by visual descriptors such as local binary patterns, wavelet transforms and Gabor filters. An exhaustive evaluation involving a variety of wavelet families and parameters was performed in order to shed light on the impact of scale on pathomics based CRC tissue differentiation. Our model achieved a performance accuracy of 95.3% with tenfold cross validation demonstrating superior performance compared to 87.4% reported in recent studies. Furthermore, we experimentally showed that the first and the second levels of the wavelet approximations can be used without compromising classification performance.
| S-EPMC8324876 | biostudies-literature
Project description:BackgroundPrevious studies indicate that alterations in Human Leukocyte Antigen (HLA) class I expression are frequent in colorectal tumors. This would suggest serious limitations for immunotherapy-based strategies involving T-cell recognition. Distinct patterns of HLA surface expression might conceal different immune escape mechanisms employed by the tumors and are worth further study.MethodWe applied four-color multiparameter flow cytometry (FCM), using a large panel of alloantigen-specific anti-HLA-A and -B monoclonal antibodies, to study membranous expression of individual HLA alleles in freshly isolated colorectal cancer cell suspensions from 21 patients.ResultsAlterations in HLA class I phenotype were observed in 8 (38%) of the 21 tumors and comprised loss of a single A or B alleles in 4 cases, and loss of all four A and B alleles in the other 4 cases. Seven of these 8 tumors were located on the right side of the colon, and those showing loss of both HLA-A and -B membranous expression were all of the MSI-H phenotype.ConclusionFCM allows the discrimination of complex phenotypes related to the expression of HLA class I. The different patterns of HLA class I expression might underlie different tumor behavior and influence the success rate of immunotherapy.
| S-EPMC1599746 | biostudies-literature
Project description:To investigate the diagnostic value of texture analysis (TA) for differentiating between colorectal cancer (CRC), colonic lesions caused by inflammatory bowel disease (IBD), and normal thickened colon wall (NTC) on computed tomography (CT) and assess which scanning phase has the highest differential diagnostic value. In all, 107 patients with CRC, 113 IBD patients with colonic lesions, and 96 participants with NTC were retrospectively enrolled. All subjects underwent multiphase CT examination, including pre-contrast phase (PCP), arterial phase (AP), and portal venous phase (PVP) scans. Based on these images, classification by TA and visual classification by radiologists were performed to discriminate among the three tissue types. The performance of TA and visual classification was compared. Precise TA classification results (error, 2.03-12.48%) were acquired by nonlinear discriminant analysis for CRC, IBD and NTC, regardless of phase or feature selection. PVP images showed a better ability to discriminate the three tissues by comprising the three scanning phases. TA showed significantly better performance in discriminating CRC, IBD and NTC than visual classification for residents, but there was no significant difference in classification between TA and experienced radiologists. TA could provide useful quantitative information for the differentiation of CRC, IBD and NTC on CT, particularly in PVP images.
| S-EPMC7156692 | biostudies-literature
Project description:Most genetic variants for colorectal cancer (CRC) identified in genome-wide association studies (GWAS) are located in intergenic regions, implying pathogenic dysregulations of gene expression. However, comprehensive assessments of target genes in CRC remain to be explored. We conducted a multi-omics analysis using transcriptome and/or DNA methylation data from the Genotype-Tissue Expression, The Cancer Genome Atlas and the Colonomics projects. We identified 116 putative target genes for 45 GWAS-identified variants. Using summary-data-based Mendelian randomization approach (SMR), we demonstrated that the CRC susceptibility for 29 out of the 45 CRC variants may be mediated by cis-effects on gene regulation. At a cutoff of the Bonferroni-corrected PSMR < 0.05, we determined 66 putative susceptibility genes, including 39 genes that have not been previously reported. We further performed in vitro assays for two selected genes, DIP2B and SFMBT1, and provide functional evidence that they play a vital role in colorectal carcinogenesis via disrupting cell behavior, including migration, invasion and epithelial-mesenchymal transition. Our study reveals a large number of putative novel susceptibility genes and provides additional insight into the underlying mechanisms for CRC genetic risk loci.
| S-EPMC8485221 | biostudies-literature
Project description:Sound sources in the world are experienced as stable even when intermittently obscured, implying perceptual completion mechanisms that "fill in" missing sensory information. We demonstrate a filling-in phenomenon in which the brain extrapolates the statistics of background sounds (textures) over periods of several seconds when they are interrupted by another sound, producing vivid percepts of illusory texture. The effect differs from previously described completion effects in that 1) the extrapolated sound must be defined statistically given the stochastic nature of texture, and 2) the effect lasts much longer, enabling introspection and facilitating assessment of the underlying representation. Illusory texture biases subsequent texture statistic estimates indistinguishably from actual texture, suggesting that it is represented similarly to actual texture. The illusion appears to represent an inference about whether the background is likely to continue during concurrent sounds, providing a stable statistical representation of the ongoing environment despite unstable sensory evidence.
| S-EPMC6841952 | biostudies-literature
Project description:Matrix metalloproteinase-2 (MMP-2) plays a role in cell migration and invasion that involve degradation of extracellular matrix proteins as well as vascular remodeling. Increased expression of MMP-2 has often been observed in the tumors of colorectal cancer (CRC) patients. In order to unveil the perturbed proteomic signal during MMP-2-induced cancer progression, we analyzed MMP-2 dependent secretome change in HCT116 cancer cells by stable isotopic labeling with amino acids in cell culture (SILAC), together with the plasma proteome profile of CRC patients by label-free quantification according to disease progression from TNM stage I to IV. We also analyzed public database of CRC tissue proteome deposited by Clinical Proteomic Tumor Analysis Consortium (NCI/NIH). As a strategy for integrating the multi-layered proteomes, we first performed unsupervised hierarchical clustering on plasma and tissue proteomes according to TNM stage, and then Fisher’s exact test to find out which clusters were significantly represented by the secretome decreased by more than 1.5-fold upon MMP-2 knock down. The resultant cluster included CD9-ITGB1-ITGA7 protein complex as well as MMP-2. The proteins included in this cluster were mostly related to cell-cell communications and cell motility, and showed remarkable increased expression in tissue and plasma upon disease progression from TNM I to II. MMP-2 induced activation of FAK phospho-signaling in an activity-dependent manner. With the comparative and integrative multi-layered proteomic analysis, we suggest that high invasiveness featured in the disease progression of colorectal cancer with increased secretion of MMP-2 is because MMP-2 activates FAK signaling through CD9-integrin complex and consequently induces positive feedback of MMP-2 upregulation.
2021-08-26 | PXD018304 | Pride
Project description:Histomorphological measurements can be used to identify microstructural changes related to disease pathomechanics, in particular, plantar soft tissue changes with diabetes. However, these measurements are time-consuming and susceptible to sampling and human measurement error. We investigated two approaches to automate segmentation of plantar soft tissue stained with modified Hart's stain for elastin with the eventual goal of subsequent morphological analysis. The first approach used multiple texture- and color-based features with tile-wise classification. The second approach used a convolutional neural network modified from the U-Net architecture with fewer channel dimensions and additional downsampling steps. A hybrid color and texture feature, Fourier reduced histogram of uniform improved opponent color local binary patterns (f-IOCLBP), yielded the best feature-based segmentation, but still performed 3.6% worse on average than the modified U-Net. The texture-based method was sensitive to changes in illumination and stain intensity, and segmentation errors were often in large regions of single tissues or at tissue boundaries. The U-Net was able to segment small, few-pixel tissue boundaries, and errors were often trivial to clean up with post-processing. A U-Net approach outperforms hand-crafted features for segmentation of plantar soft tissue stained with modified Hart's stain for elastin.
| S-EPMC8263502 | biostudies-literature
Project description:Mucinous colorectal carcinomas (MC) constitute 10% of colorectal malignancies. Recently, an increased risk of colorectal cancer has been demonstrated in germline BRCA1/2 mutation carriers. Furthermore, BRCA1/2 germline mutation carriers have exhibited a higher-than-expected frequency of MC tumors. Here, we investigate the relationship between BRCA mutations and mucinous histology in colorectal carcinoma patients, using both an existing cohort of sequenced colorectal tumors and a prospective case-control study comparing MC and conventional adenocarcinoma (AC) patients tested for BRCA mutations. We discovered that MC tumors exhibit a statistically significantly higher incidence of BRCA mutations in addition to a higher average mutation count when compared to AC tumors in the existing cohort. The strongest predictor of the mutation count was mucinous histology, independently of other variables including microsatellite instability. Contrary to our hypothesis, the first association did not recur in the prospective case-control study, likely due to our pathological definition of MC tumors and small sample size. Finally, we observed a higher tumor mutational burden (TMB) in MC tumors compared with AC tumors. We suggest that the association between MC histology, BRCA mutations, and increased TMB may open the door to the utilization of simple tests (such as histopathologic characterization) to detect patients who may benefit from immunotherapy in colorectal cancer.
| S-EPMC7196997 | biostudies-literature
Project description:We aimed to test the ability of texture analysis to differentiate the spatial heterogeneity of (125)I-A5B7 anti-carcinoembryonic antigen antibody distribution by nano-single photon emission computed tomography (SPECT) in well-differentiated (SW1222) and poorly differentiated (LS174T) hepatic metastatic colorectal cancer models before and after combretastatin A1 di-phosphate anti-vascular therapy.Nano-SPECT imaging was performed following tail vein injection of 20 MBq (125)I-A5B7 in control CD1 nude mice (LS174T, n=3 and SW1222, n=4), and CA1P-treated mice (LS174T, n=3; SW1222, n=4) with liver metastases. Grey-level co-occurrence matrix textural features (uniformity, homogeneity, entropy and contrast) were calculated in up to three liver metastases in 14 mice from control and treatment groups.Before treatment, the LS174T metastases (n=7) were more heterogeneous than SW1222 metastases (n=12) (uniformity, P=0.028; homogeneity, P=0.01; contrast, P=0.045). Following CA1P, LS174T metastases (n=8) showed less heterogeneity than untreated LS174T controls (uniformity, P=0.021; entropy, P=0.006). Combretastatin A1 di-phosphate-treated SW1222 metastases (n=11) showed no difference in texture features compared with controls (all P>0.05).Supporting the potential for novel imaging biomarkers, texture analysis of (125)I-A5B7 SPECT shows differences in spatial heterogeneity of antibody distribution between well-differentiated (SW1222) and poorly differentiated (LS174T) liver metastases before treatment. Following anti-vascular treatment, LS174T metastases, but not SW1222 metastases, were less heterogeneous.
| S-EPMC4580400 | biostudies-other