A study on the ability of AI to enhance the accuracy and efficiency of pathologists’ diagnosis of colorectal tumor
Ontology highlight
ABSTRACT:
DISEASE(S): Colorectal Adenoma And Colorectal Cancer
PROVIDER: 113777 | ecrin-mdr-crc |
REPOSITORIES: ECRIN MDR
Similar Datasets
Project description:Histologic diagnosis of T cell-mediated rejection in kidney transplant biopsies has limited reproducibility because it is based on non-specific lesions using arbitrary rules that are subject to differing interpretations. We used microarray results from 403 indication biopsies previously given histologic diagnoses to develop a molecular classifier that assigned a molecular T cell-mediated rejection score to each biopsy. Independent assessment of the biopsies by multiple pathologists confirmed considerable disagreement on the presence of TCMR features: 79-88% accuracy and 35-69% sensitivity. The agreement of the molecular T cell-mediated rejection score with the histology diagnosis was similar to agreement among individual pathologists: accuracy 89%, sensitivity 51%. However, the score also predicted the consensus among pathologists, being highest when all agreed. Many discrepancies between the scores and the histologic diagnoses were in situations where histology is unreliable e.g. scarred biopsies. The score correlated with histologic lesions and gene sets associated with T cell-mediated rejection. The transcripts most often selected by the classifier were expressed in effector T cells, dendritic cells, or macrophages or inducible by interferon-gamma. Thus the T cell-mediated rejection score offers an objective assessment of kidney transplant biopsies, predicting the consensus opinion among multiple pathologists, and offering insights into underlying disease mechanisms. Antibody-mediated rejection is a major cause of kidney transplant failure, but the current diagnostic system misses most cases due to dependency on subjective non-standardized tests. We hypothesized that molecular features could provide a test to address this problem. We classified 403 biopsies by a reference standard based on microcirculation lesions and donor-specific HLA antibody, and used microarray analysis to develop a classifier that assigned each biopsy a score reflecting the probability of antibody-mediated rejection. The scores correlated with donor-specific antibody and histologic lesions: 42/45 biopsies with antibody-mediated rejection scores >0.5 had both donor-specific antibody and microcirculation lesions. Intermediate scores (0.2-0.5) were more ambiguous, but became more specific combined with donor-specific antibody. Compared to diagnoses based on histology-plus-donor-specific antibody, the scores had sensitivity 0.67; specificity 0.90. Donor-specific antibody improved the specificity to 0.97. The score correlated not only with diagnoses of individual pathologists but with the consensus among multiple pathologists. The classifier used transcripts expressed in endothelial cells (e.g. CDH13, DARC, ROBO4) and NK cells (e.g. CX3CR1, FGFBP2), as well as IFNG-inducible transcripts e.g. CXCL11. Thus the molecular phenotype of antibody-mediated rejection provides not only an objective test that predicts microcirculation lesions and donor-specific HLA antibody, but also offers mechanistic insights.
2013-02-13 | GSE36059 | GEO
Project description:Histologic diagnosis of T cell-mediated rejection in kidney transplant biopsies has limited reproducibility because it is based on non-specific lesions using arbitrary rules that are subject to differing interpretations. We used microarray results from 403 indication biopsies previously given histologic diagnoses to develop a molecular classifier that assigned a molecular T cell-mediated rejection score to each biopsy. Independent assessment of the biopsies by multiple pathologists confirmed considerable disagreement on the presence of TCMR features: 79-88% accuracy and 35-69% sensitivity. The agreement of the molecular T cell-mediated rejection score with the histology diagnosis was similar to agreement among individual pathologists: accuracy 89%, sensitivity 51%. However, the score also predicted the consensus among pathologists, being highest when all agreed. Many discrepancies between the scores and the histologic diagnoses were in situations where histology is unreliable e.g. scarred biopsies. The score correlated with histologic lesions and gene sets associated with T cell-mediated rejection. The transcripts most often selected by the classifier were expressed in effector T cells, dendritic cells, or macrophages or inducible by interferon-gamma. Thus the T cell-mediated rejection score offers an objective assessment of kidney transplant biopsies, predicting the consensus opinion among multiple pathologists, and offering insights into underlying disease mechanisms. Antibody-mediated rejection is a major cause of kidney transplant failure, but the current diagnostic system misses most cases due to dependency on subjective non-standardized tests. We hypothesized that molecular features could provide a test to address this problem. We classified 403 biopsies by a reference standard based on microcirculation lesions and donor-specific HLA antibody, and used microarray analysis to develop a classifier that assigned each biopsy a score reflecting the probability of antibody-mediated rejection. The scores correlated with donor-specific antibody and histologic lesions: 42/45 biopsies with antibody-mediated rejection scores >0.5 had both donor-specific antibody and microcirculation lesions. Intermediate scores (0.2-0.5) were more ambiguous, but became more specific combined with donor-specific antibody. Compared to diagnoses based on histology-plus-donor-specific antibody, the scores had sensitivity 0.67; specificity 0.90. Donor-specific antibody improved the specificity to 0.97. The score correlated not only with diagnoses of individual pathologists but with the consensus among multiple pathologists. The classifier used transcripts expressed in endothelial cells (e.g. CDH13, DARC, ROBO4) and NK cells (e.g. CX3CR1, FGFBP2), as well as IFNG-inducible transcripts e.g. CXCL11. Thus the molecular phenotype of antibody-mediated rejection provides not only an objective test that predicts microcirculation lesions and donor-specific HLA antibody, but also offers mechanistic insights. All consenting renal transplant patients undergoing biopsies for cause as standard of care. 403 samples and 8 controls (nephrectomies). This dataset is part of the TransQST collection.
2013-02-13 | E-GEOD-36059 | biostudies-arrayexpress
Project description:Cholangiocellular carcinoma (CCC) and pancreatic ductal adenocarcinoma (PDAC) are two highly aggressive cancer types which arise from epithelial cells of the pancreatobiliary system. Due to their histologic and morphologic similarity, differential diagnosis between CCC and metastasis of PDAC located in the liver frequently proves an unsolvable issue for pathologists. Biomarkers with high specificity and sensitivity for the differentiation of these tumour types would therefore be a valuable tool, but so far none are available. Here, we address this problem by comparing microdissected CCC and PDAC tumour cells in a label-free proteomics approach. The novel biomarker candidates were subsequently verified by immunohistochemical staining of primary and secondary tumour tissue.
2015-12-10 | PXD001833 | Pride
Project description:Women that underwent operation for what was expected to be epithelial ovarian cancer at Odense University Hospital, Denmark, from April 2004 to February 2005, were consecutively asked to participate in the study, and all gave written consent. In the operating theatre the tumors were removed as an entirety, and immediately transported to the pathology department without fixation. One senior pathologist was responsible for collection of samples. From each tumor three samples of about 200 300 mg each were taken with maximal mutual distance, taking into consideration that all samples should contain tumor tissue. This way of sampling was done without any kind of magnification and is mentioned as macro-dissection. The three samples were given the letters a, b and c. The three samples from each of the nine tumors are compared.
2007-02-01 | E-MEXP-935 | biostudies-arrayexpress
Project description:Genetic lesions characteristic for RCC subtypes can be identified by virtual karyotyping with SNP microarrays. In this study, we examined whether virtual karyotypes could be used to better classify a cohort of morphologically challenging/unclassified RCC. Tumor resection specimens from 17 patients were profiled by virtual karyotyping with Affymetrix 10K 2.0 or 250K Nsp SNP Mapping arrays and were also evaluated independently by a panel of seven genito-urinary pathologists. Tumors were classified by the established pattern of genomic imbalances based on a reference cohort of 98 cases with classic morphology and compared to the morphologic diagnosis of the pathologist panel. In 3 cases, samples from areas with different morphologic appearance were also tested (n=5).
2009-03-13 | E-GEOD-14670 | biostudies-arrayexpress
Project description:Low response rates to certain tumor types remain a major challenge for immune checkpoint blockade therapy. In this study, we first conducted an integrated biomarker evaluation of bladder cancer patients from confirmatory cohorts (IMvigor210) and found that no significant differences exist between sexes before acceptance of anti-PD-L1 treatment, whereas male patients showed a better response. Thus, we then focused on sex-related changes post anti-PD-L1 treatment and found no obvious impact on the gut microbiota in male mice but a significant decrease in the sex hormone levels. Further, castration dramatically enhanced the antitumor efficacy against murine colon adenocarcinoma in male mice. Moreover, a narrow-spectrum antibiotic, colistin was innovatively used for deregulation of testosterone levels to enhance the immunotherapy efficiency in male mice. These findings indicate that the impact on the sex hormone levels in males may contribute to the sexual dimorphism in response and provide a promising way to enhance immunotherapy efficiency.
| S-EPMC8477942 | biostudies-literature
Project description: Not available
| S-EPMC10912629 | biostudies-literature
Project description:The mastication stimuli during the weaning period can enhance the learning ability of rats by increasing the dendrite branches of hippocampal CA1 neurons and by regulating genes related to dendrite growth. Using DNA microarray analysis, differentially expressed genes were enriched in functional terms related to dendrite growth and included the Igf2, RhoA, and Rho GEF genes, most of which were upregulated in the chow diet group.
2022-09-08 | GSE210292 | GEO