Interest of Artificial Intelligence in Cancer Screening Colonoscopy
Ontology highlight
ABSTRACT:
DISEASE(S): Intestinal Polyps,Colonoscopy,Colorectal Neoplasms,Polyps,Deep Learning
PROVIDER: 2380194 | ecrin-mdr-crc |
REPOSITORIES: ECRIN MDR
Similar Datasets
Project description:Adenomatous polyps adjacent to colorectal cancer (CRC) were found to exhibit two distinct microRNAs (miRs) patterns from normal mucosa to low- and separately, to high-grade dysplasia; presence in screen-detected adenoma of non-cancer patients is unknown. Global miR expression was performed on biopsies obtained from 109 healthy patients undergoing screening/surveillance colonoscopy. Included were normal mucosa (NM); hyperplastic polyp (HP); tubular adenoma (TA), tubulovillous adenoma, with or without, high-grade dysplasia (TVHG) and serrated-polyps; sessile serrated adenoma (SSA) and traditional serrated adenoma (TSA). Logistic regression was used to model miRs predictive of histology and CRC risk. We identified 99 miRs that differed across five histologic groups (FDR=0.05) and that accurately separated on histology (Concordance Index (CI)=0.96). In HPNM, miRs-145, -143, -107a, -23b, and -24 were upregulated whereas miRs-663, -1268, -320b, -1275, and -671 where overexpressed in TVHGs (FDR P<0 .05). The expression of miR-145 and -30a showed high accuracy to separate low from high-risk polyps independent of serrated status (CI= 97.1%; AUC 93.4%). For TSAs, miR-125b and -199a were uniquely downregulated relative to HPNMs and miR-335 discriminated between non-serrated and serrated histology. Histologically advanced polyps from non-cancer patients share miR alterations with those reported for CRC and high-grade adenoma adjacent to tumor including downregulation of immune regulatory miRs-125 and -199a in TSAs; polyp that frequently present with in situ carcinoma. These data extend evidence that miR patterns of high-risk adenoma are detectable in subset of screen-detected adenoma for which measurement may be useful in in adenoma risk stratification.
2015-09-29 | GSE73487 | GEO
Project description:This preliminary clinical trial aims to evaluate the performance of the low-dose CT colonoscopy (CTC) with computer aided detection (CAD) on polyps detection compared with optical colonoscopy (OC), and explore the possible clinical routine for integrated use of CT colonoscopy and optical colonoscopy OC in colorectal cancer screening.
| 2237598 | ecrin-mdr-crc
Project description:The investigators hypothesize that the clinical implementation of an AI system is an optimal tool to monitor, audit and improve the detection and classification of polyps during colonoscopy. The purpose of this prospective clinical cohort study is to evaluate the performance of the SCALE-EYE virtual scale for measuring polyp size when used during live colonoscopies. The investigators also wish to evaluate CAD-eye for detection and classification of polyp histology. It is hypothesized that CAD-eye and SCALE-EYE can function in real-time practice with high accuracy.
| 2404446 | ecrin-mdr-crc
Project description:Colorectal cancer often arises from adenomatous polyps. Polyps can grow and progress to cancer, but may also remain static in size, regress, or resolve. Prediciting which progress to cancer and which remain benign is difficult. We developed a long-lived murine model of colorectal cancer with tumors that can be followed by colonoscopy. Our aim was to assess whether these tumors have similar growth patterns and histologic fates to human colorectal polyps to identify features to aid in risk stratification of colonic tumors. Long-lived Apcin/+ mice were treated with 4% dextran sodium sulfate to promote colonic tumorigenesis. Tumor growth patterns were characterized by serial colonoscopy, and pathology was determined. Serial biopsies of tumors were obtained for immunohistochemistry and gene expression profiling by microarray analysis with Affymetrix Whole Genome array. Tumors (n=424) grew, remained static, regressed, or resolved over time with different relative frequencies. Newly developed tumors demonstrated dynamic growth patterns with higher rates of growth and resolution, while more established tumors tended to remain static in size. Colonic tumors were hyperplastic lesions (3%), adenomas (73%), intramucosal carcinomas (20%), or adenocarcinomas (3%). Differentially expressed genes between adenomas and intramucosal carcinomas were identified. We did not identify differentially expressed genes between early and late biopsies from the same tumor. This novel murine model of intestinal tumorigenesis develops colonic tumors that can be monitored by serial colonoscopy, mirror growth patterns seen in human colorectal polyps, and progress to colorectal cancer. Further characterization of cellular and molecular features are needed to determine which features can be used to risk-stratify polyps for progress to colorectal cancer and potentially guide prevention strategies. F1 (SWR x C57BL/6) Apcin/+ mice who had been given two treatments of 4% dextran sodium sulfate in drinking water at weaning underwent colonoscopy around 80 days of age. Those with distal colonic tumors amenable to biopsy had two biopsies taken at that time. Tumors were monitored by colonoscopy every 14 days. Every 28 days, two additional biopsies were taken of each tumor. This protocol was repeated until mice were moribund. Six tumors were removed after sacrifice. The earliest and latest biopsies of each tumor were selected for RNA extraction and Affymetrix hybridization.
2014-05-03 | E-GEOD-46626 | biostudies-arrayexpress
Project description:Screening and surveillance of colorectal cancers (CRCs) using advanced colonoscopy technologies have significantly reduced the incidence and mortality rates of CRCs in recent years. However, a significant portion of CRCs are still remained undiagnosed, especially those involving sessile serrated adenomas/polyps (SSA/P), most likely due to their flat shape and the excessive amounts of secreted mucin that cover the polyps, making them invisible for colonoscopy. Here, a potential alternative solution is the application of molecular markers enabling unambiguous characterization of SSPs. However, full implementation of this strategy requires availability of robust markers which are still lacking. In this work by comprehensive molecular analysis of several malignant and normal samples at the genome, methylome and transcriptome levels we show that activating mutation of BRAF-V600E drives the generation of a unique SSP-specific DNA methylation profile. As the result a robust set of DNA methylation markers showing significant (~3 to 30 fold) increase in their methylation levels, exclusively in SSP samples, are introduced. These markers can be of important clinical relevance, especially in early diagnosis of SSPs using non-invasive approaches such as fecal DNA testing.
2018-04-09 | GSE110537 | GEO
Project description:Screening and surveillance of colorectal cancers (CRCs) using advanced colonoscopy technologies have significantly reduced the incidence and mortality rates of CRCs in recent years. However, a significant portion of CRCs are still remained undiagnosed, especially those involving sessile serrated adenomas/polyps (SSA/P), most likely due to their flat shape and the excessive amounts of secreted mucin that cover the polyps, making them invisible for colonoscopy. Here, a potential alternative solution is the application of molecular markers enabling unambiguous characterization of SSPs. However, full implementation of this strategy requires availability of robust markers which are still lacking. In this work by comprehensive molecular analysis of several malignant and normal samples at the genome, methylome and transcriptome levels we show that activating mutation of BRAF-V600E drives the generation of a unique SSP-specific DNA methylation profile. As the result a robust set of DNA methylation markers showing significant (~3 to 30 fold) increase in their methylation levels, exclusively in SSP samples, are introduced. These markers can be of important clinical relevance, especially in early diagnosis of SSPs using non-invasive approaches such as fecal DNA testing.
2018-04-09 | GSE110536 | GEO
Project description:Screening and surveillance of colorectal cancers (CRCs) using advanced colonoscopy technologies have significantly reduced the incidence and mortality rates of CRCs in recent years. However, a significant portion of CRCs are still remained undiagnosed, especially those involving sessile serrated adenomas/polyps (SSA/P), most likely due to their flat shape and the excessive amounts of secreted mucin that cover the polyps, making them invisible for colonoscopy. Here, a potential alternative solution is the application of molecular markers enabling unambiguous characterization of SSPs. However, full implementation of this strategy requires availability of robust markers which are still lacking. In this work by comprehensive molecular analysis of several malignant and normal samples at the genome, methylome and transcriptome levels we show that activating mutation of BRAF-V600E drives the generation of a unique SSP-specific DNA methylation profile. As the result a robust set of DNA methylation markers showing significant (~3 to 30 fold) increase in their methylation levels, exclusively in SSP samples, are introduced. These markers can be of important clinical relevance, especially in early diagnosis of SSPs using non-invasive approaches such as fecal DNA testing.
2018-04-09 | GSE110535 | GEO
Project description:<p>In the US, CRC incidence has declined with uptake in colonoscopy for early detection and removal of polyps, the precursor lesion, can stop a cancer from developing but in spite of this, CRC remains the second leading cause of cancer death in the United States. One third of people who undergo screening colonoscopy will have adenomatous polyps, but less than 5% of the time are these polyps presumed to go on to develop into cancer. Why does one polyp develop into cancer while another one that looks very similar does not. In this proposal we will identify what molecular genetic changes in the genome, in the mRNA expression and genetic methylation patterns will distinguish a polyp that has already transformed into cancer from one that has not.</p>
| phs001384 | dbGaP
Project description:Colorectal cancer often arises from adenomatous polyps. Polyps can grow and progress to cancer, but may also remain static in size, regress, or resolve. Prediciting which progress to cancer and which remain benign is difficult. We developed a long-lived murine model of colorectal cancer with tumors that can be followed by colonoscopy. Our aim was to assess whether these tumors have similar growth patterns and histologic fates to human colorectal polyps to identify features to aid in risk stratification of colonic tumors. Long-lived Apc^Min/+ mice were treated with 4% dextran sodium sulfate to promote colonic tumorigenesis. Tumor growth patterns were characterized by serial colonoscopy, and pathology was determined. Serial biopsies of tumors were obtained for immunohistochemistry and gene expression profiling by microarray analysis with Affymetrix Whole Genome array. Tumors (n=424) grew, remained static, regressed, or resolved over time with different relative frequencies. Newly developed tumors demonstrated dynamic growth patterns with higher rates of growth and resolution, while more established tumors tended to remain static in size. Colonic tumors were hyperplastic lesions (3%), adenomas (73%), intramucosal carcinomas (20%), or adenocarcinomas (3%). Differentially expressed genes between adenomas and intramucosal carcinomas were identified. We did not identify differentially expressed genes between early and late biopsies from the same tumor. This novel murine model of intestinal tumorigenesis develops colonic tumors that can be monitored by serial colonoscopy, mirror growth patterns seen in human colorectal polyps, and progress to colorectal cancer. Further characterization of cellular and molecular features are needed to determine which features can be used to risk-stratify polyps for progress to colorectal cancer and potentially guide prevention strategies.
2014-05-03 | GSE46626 | GEO
Project description:Interventions: None
Primary outcome(s): The primary outcome of the study is the accuracy of the CAD-CNN system for predicting histology of diminutive colorectal polyps (1-5mm) compared with the accuracy of the prediction of the endoscopist. Both the CAD-CNN system and the endoscopist will use NBI for their predictions.
Accuracy is defined as the percentage of correctly predicted optical diagnoses of the CAD-CNN system and/or endoscopist compared to the gold standard pathology. For the calculation of the accuracy, adenomas and SSLs will be dichotomised as neoplastic polyps, while HPs and other non-neoplastic histology are considered non-neoplastic.
Study Design: N/A: single arm study, Open (masking not used), N/A , unknown, Other
| 2443187 | ecrin-mdr-crc