-
Gastrointestinal Endoscopy Dec 2021Colorectal cancer is a leading cause of death. Colonoscopy is the criterion standard for detection and removal of precancerous lesions and has been shown to reduce...
BACKGROUND AND AIMS
Colorectal cancer is a leading cause of death. Colonoscopy is the criterion standard for detection and removal of precancerous lesions and has been shown to reduce mortality. The polyp miss rate during colonoscopies is 22% to 28%. DEEP DEtection of Elusive Polyps (DEEP) is a new polyp detection system based on deep learning that alerts the operator in real time to the presence and location of polyps. The primary outcome was the performance of DEEP on the detection of elusive polyps.
METHODS
The DEEP system was trained on 3611 hours of colonoscopy videos derived from 2 sources and was validated on a set comprising 1393 hours from a third unrelated source. Ground truth labeling was provided by offline gastroenterologist annotators who were able to watch the video in slow motion and pause and rewind as required. To assess applicability, stability, and user experience and to obtain some preliminary data on performance in a real-life scenario, a preliminary prospective clinical validation study was performed comprising 100 procedures.
RESULTS
DEEP achieved a sensitivity of 97.1% at 4.6 false alarms per video for all polyps and of 88.5% and 84.9% for polyps in the field of view for less than 5 and 2 seconds, respectively. DEEP was able to detect polyps not seen by live real-time endoscopists or offline annotators in an average of .22 polyps per sequence. In the clinical validation study, the system detected an average of .89 additional polyps per procedure. No adverse events occurred.
CONCLUSIONS
DEEP has a high sensitivity for polyp detection and was effective in increasing the detection of polyps both in colonoscopy videos and in real procedures with a low number of false alarms. (Clinical trial registration number: NCT04693078.).
Topics: Adenomatous Polyps; Artificial Intelligence; Colonic Polyps; Colonoscopy; Colorectal Neoplasms; Humans; Prospective Studies
PubMed: 34216598
DOI: 10.1016/j.gie.2021.06.021 -
Asian Pacific Journal of Cancer... Jul 2023The current gold standard non-invasive test for detecting pre-cancerous changes is the faecal immunochemical test (FIT). However, this test can lack sensitivity and...
BACKGROUND
The current gold standard non-invasive test for detecting pre-cancerous changes is the faecal immunochemical test (FIT). However, this test can lack sensitivity and specificity and testing for another biomarker may address these limitations. Chitinase 3-like 1 (CHI3L1) is emerging as a potential biomarker of inflammation-associated carcinogenic changes in epithelial cells. In this study CHI3L1 levels were analysed in patients and controls to determine their ability to improve detection of early CRC either alone or in combination with a FIT.
METHODS
CHI3L1 levels were measured by ELISA in serum and stool samples from cohorts of CRC and healthy donors as well as stool samples from a cohort of symptomatic primary care patients. Faecal haemoglobin was also analysed in the same primary care samples using FIT.
RESULTS
CHI3L1 levels were a good discriminatory marker of CRC, with no significant difference between levels detected in the stool and serum samples. ROC curves that determined the optimal cut-point however identified that stool samples gave higher sensitivity (83% versus 69%) and specificity (89% versus 74%) than matched serum samples. Faecal CHI3L1 levels in the primary care patients were not significantly different (p=0.193) from those detected in the healthy controls. ROC curve analysis confirmed that faecal CHI3L1 levels had limited ability to discriminate between patients who did or didn't have evidence of lesions (AUC=0.52, p=0.74). Similarly, CHI3L1 levels did not reliably identify those symptomatic primary care patients who subsequently presented with early-stage disease (polyps and adenomas) or CRC. The discriminatory power of FIT was not increased by incorporating the CHI3L1 results in this setting.
CONCLUSION
There was no evidence that measurement of faecal CHI3L1 has the potential to increase diagnostic accuracy, either alone or in combination with a FIT, in symptomatic primary care patients.
Topics: Humans; Biomarkers; Colorectal Neoplasms; Early Detection of Cancer; Feces; Hemoglobins; Occult Blood; Primary Health Care; Sensitivity and Specificity; Chitinase-3-Like Protein 1
PubMed: 37505758
DOI: 10.31557/APJCP.2023.24.7.2289 -
Journal of Cancer Research and Clinical... Dec 2023Ultrasound imaging is the preferred method for the early diagnosis of endometrial diseases because of its non-invasive nature, low cost, and real-time imaging features....
PURPOSE
Ultrasound imaging is the preferred method for the early diagnosis of endometrial diseases because of its non-invasive nature, low cost, and real-time imaging features. However, the accurate evaluation of ultrasound images relies heavily on the experience of radiologist. Therefore, a stable and objective computer-aided diagnostic model is crucial to assist radiologists in diagnosing endometrial lesions.
METHODS
Transvaginal ultrasound images were collected from multiple hospitals in Quzhou city, Zhejiang province. The dataset comprised 1875 images from 734 patients, including cases of endometrial polyps, hyperplasia, and cancer. Here, we proposed a based self-supervised endometrial disease classification model (BSEM) that learns a joint unified task (raw and self-supervised tasks) and applies self-distillation techniques and ensemble strategies to aid doctors in diagnosing endometrial diseases.
RESULTS
The performance of BSEM was evaluated using fivefold cross-validation. The experimental results indicated that the BSEM model achieved satisfactory performance across indicators, with scores of 75.1%, 87.3%, 76.5%, 73.4%, and 74.1% for accuracy, area under the curve, precision, recall, and F1 score, respectively. Furthermore, compared to the baseline models ResNet, DenseNet, VGGNet, ConvNeXt, VIT, and CMT, the BSEM model enhanced accuracy, area under the curve, precision, recall, and F1 score in 3.3-7.9%, 3.2-7.3%, 3.9-8.5%, 3.1-8.5%, and 3.3-9.0%, respectively.
CONCLUSION
The BSEM model is an auxiliary diagnostic tool for the early detection of endometrial diseases revealed by ultrasound and helps radiologists to be accurate and efficient while screening for precancerous endometrial lesions.
Topics: Humans; Female; Uterine Diseases; Computer Simulation; Hospitals; Hyperplasia; Physicians; Precancerous Conditions
PubMed: 37947870
DOI: 10.1007/s00432-023-05467-7 -
Journal of Healthcare Engineering 2022This study aimed to investigate the relevance of the study with the neutrophil count and lymphocyte count ratio (NLR), platelet count and lymphocyte count ratio (PLR),...
OBJECTIVE
This study aimed to investigate the relevance of the study with the neutrophil count and lymphocyte count ratio (NLR), platelet count and lymphocyte count ratio (PLR), and red blood cell distribution width (RDW) in the prognostic evaluation of colorectal cancer patients.
METHODS
143 patients with colorectal cancer from January 2016 to January 2019 were selected by our hospital, and then, other 143 cases of physical examiners as normal groups were selecting to proceed colonoscopic biopsy to diagnose 106 cases of precancerous diseases related to colorectal cancer. Among them were the inflammatory bowel group ( = 56) and the colorectal polyp group ( = 50). Analysis of the survival impact factors of patients with carcinoma of the rectum, preoperative NLR, ROW, PLR, and prognostic relationship, and comparison of NLR, PLR, and RDW diagnostic rate and expression were performed.
RESULTS
Tissue type, TNM stage, lymph node metastasis, NLR, RDW, and PLR had a predictive influence on patients with colorectal cancer (0.05). There was no link between gender, age, aetiology, pathological type, and prognosis in patients with colorectal cancer ( > 0.05). Multiple variables in patients with colorectal cancer are affected by tissue categorization (poor differentiation), TNM stages (III, IV), lymph node metastases, NLR, ROW, and PLR (0.05). When compared to solo NLR, Row, and PLR diagnostics, the combination diagnosis and malignancy rates were greater, and the differences were statistically significant (0.05). Diagnostic sensitivity, specificity, and accuracy were greater when compared to single NLR, ROW, and PLR. When compared to the normal control group, NLR, ROW, and PLR have greater levels, and the differences are statistically significant (0.05). The patient survival declines more slowly as PLR, NLR, and the severity of the condition rises.
CONCLUSION
NLR, ROW, and PLR combined diagnosis has high accuracy in colorectal cancer diagnosis, and the prognosis of patients with NLR, ROW, and PLR levels has a tight association; so, clinically, the above signs should be identified, and the optimal treatment time is grasped.
Topics: Colorectal Neoplasms; Humans; Lymphocyte Count; Lymphocytes; Platelet Count; Prognosis
PubMed: 35469225
DOI: 10.1155/2022/7971415 -
Cancers Mar 2023Alterations in gut microbiota play a pivotal role in the adenoma-carcinoma sequence. However, there is still a notable lack of the correct implementation of tissue and... (Review)
Review
Alterations in gut microbiota play a pivotal role in the adenoma-carcinoma sequence. However, there is still a notable lack of the correct implementation of tissue and fecal sampling in the setting of human gut microbiota examination. This study aimed to review the literature and to consolidate the current evidence on the use of mucosa and a stool-based matrix investigating human gut microbiota changes in precancerous colorectal lesions. A systematic review of papers from 2012 until November 2022 published on the PubMed and Web of Science databases was conducted. The majority of the included studies have significantly associated gut microbial dysbiosis with premalignant polyps in the colorectum. Although methodological differences hampered the precise fecal and tissue-derived dysbiosis comparison, the analysis revealed several common characteristics in stool-based and fecal-derived gut microbiota structures in patients with colorectal polyps: simple or advanced adenomas, serrated lesions, and carcinomas in situ. The mucosal samples considered were more relevant for the evaluation of microbiota's pathophysiological involvement in CR carcinogenesis, while non-invasive stool sampling could be beneficial for early CRC detection strategies in the future. Further studies are required to identify and validate mucosa-associated and luminal colorectal microbial patterns and their role in CRC carcinogenesis, as well as in the clinical setting of human microbiota studies.
PubMed: 36900392
DOI: 10.3390/cancers15051602 -
Frontiers in Pharmacology 2023Long-term maintenance therapy with proton pump inhibitors (PPIs) is a common treatment strategy for acid-related gastrointestinal diseases. However, concerns have been...
Long-term maintenance therapy with proton pump inhibitors (PPIs) is a common treatment strategy for acid-related gastrointestinal diseases. However, concerns have been raised about the potential increased risk of gastric cancer and related precancerous lesions with long-term PPI use. This systematic review and meta-analysis aimed to evaluate this potential risk. We searched PubMed, Embase, and the Cochrane Central Register of Controlled Trials for randomised controlled trials published before 1 March 2023, with no language restrictions. The primary endpoint was the occurrence and progression of gastric mucosal atrophy, intestinal metaplasia, Enterochromaffin-like (ECL) cell hyperplasia, gastric polyps, and gastric cancer during the trial and follow-up. Data were analysed using a random effects model. Of the 4,868 identified studies, 10 met the inclusion criteria and were included in our analysis, comprising 27,283 participants. Compared with other treatments, PPI maintenance therapy for more than 6 months was associated with an increased risk of ECL cell hyperplasia (OR 3.01; 95% CI 1.29 to 7.04; = 0.01). However, no significant increase was found in the risk of gastric mucosal atrophy (OR 1.01; 95% CI 0.55 to 1.85; = 0.97), intestinal metaplasia (OR 1.14; 95% CI 0.49 to 2.68; = 0.76), gastric polyps (OR 1.13; 95% CI 0.68 to 1.89; = 0.64), or gastric cancer (OR 1.06; 95% CI 0.79 to 1.43; = 0.71). This systematic review and meta-analysis does not support an increased risk of gastric cancer or related precancerous lesions with long-term PPI maintenance therapy. However, long-term PPI use should be monitored for potential complications such as ECL cell hyperplasia. Further studies are needed to confirm these findings and evaluate the safety of PPI maintenance therapy for acid-related gastrointestinal diseases. https://www.crd.york.ac.uk/prospero/, Identifier: PROSPERO (CRD42022379692).
PubMed: 37693896
DOI: 10.3389/fphar.2023.1244400 -
Frontiers in Genetics 2022Colorectal cancer (CRC) is the third leading cause of cancer death globally. Early detection and removal of precancerous polyps can significantly reduce the chance of...
Colorectal cancer (CRC) is the third leading cause of cancer death globally. Early detection and removal of precancerous polyps can significantly reduce the chance of CRC patient death. Currently, the polyp detection rate mainly depends on the skill and expertise of gastroenterologists. Over time, unidentified polyps can develop into cancer. Machine learning has recently emerged as a powerful method in assisting clinical diagnosis. Several classification models have been proposed to identify polyps, but their performance has not been comparable to an expert endoscopist yet. Here, we propose a multiple classifier consultation strategy to create an effective and powerful classifier for polyp identification. This strategy benefits from recent findings that different classification models can better learn and extract various information within the image. Therefore, our Ensemble classifier can derive a more consequential decision than each individual classifier. The extracted combined information inherits the ResNet's advantage of residual connection, while it also extracts objects when covered by occlusions through depth-wise separable convolution layer of the Xception model. Here, we applied our strategy to still frames extracted from a colonoscopy video. It outperformed other state-of-the-art techniques with a performance measure greater than 95% in each of the algorithm parameters. Our method will help researchers and gastroenterologists develop clinically applicable, computational-guided tools for colonoscopy screening. It may be extended to other clinical diagnoses that rely on image.
PubMed: 35559018
DOI: 10.3389/fgene.2022.844391 -
Life (Basel, Switzerland) Mar 2023Colorectal cancer is one of the most common malignancies and the leading cause of cancer death worldwide. Wireless capsule endoscopy is currently the most frequent...
Colorectal cancer is one of the most common malignancies and the leading cause of cancer death worldwide. Wireless capsule endoscopy is currently the most frequent method for detecting precancerous digestive diseases. Thus, precise and early polyps segmentation has significant clinical value in reducing the probability of cancer development. However, the manual examination is a time-consuming and tedious task for doctors. Therefore, scientists have proposed many computational techniques to automatically segment the anomalies from endoscopic images. In this paper, we present an end-to-end 2D attention residual U-Net architecture (AttResU-Net), which concurrently integrates the attention mechanism and residual units into U-Net for further polyp and bleeding segmentation performance enhancement. To reduce outside areas in an input image while emphasizing salient features, AttResU-Net inserts a sequence of attention units among related downsampling and upsampling steps. On the other hand, the residual block propagates information across layers, allowing for the construction of a deeper neural network capable of solving the vanishing gradient issue in each encoder. This improves the channel interdependencies while lowering the computational cost. Multiple publicly available datasets were employed in this work, to evaluate and verify the proposed method. Our highest-performing model was AttResU-Net, on the MICCAI 2017 WCE dataset, which achieved an accuracy of 99.16%, a Dice coefficient of 94.91%, and a Jaccard index of 90.32%. The experiment findings show that the proposed AttResU-Net overcomes its baselines and provides performance comparable to existing polyp segmentation approaches.
PubMed: 36983874
DOI: 10.3390/life13030719 -
Genes & Genetic Systems Apr 2022Long non-coding RNAs have been proposed as biomarkers for the detection, prevention and screening of various malignancies. In this study, two lncRNAs (ANRIL and BANCR)...
Long non-coding RNAs have been proposed as biomarkers for the detection, prevention and screening of various malignancies. In this study, two lncRNAs (ANRIL and BANCR) were assessed for biomarker application in the early detection of colorectal cancer (CRC) through stool specimen testing, as a non-invasive and cost-effective methodology. A total of 40 stool samples were collected from patients referred to the hospital with colorectal cancer or adenomatous polyps as pre-cancerous lesions; patients were diagnosed using colonoscopy and pathology reports were available. Twenty control samples were also obtained from healthy subjects for comparison. RNA extraction and cDNA synthesis were followed by real-time PCR to evaluate lncRNA expression. The up-regulation of ANRIL in 20% of samples taken from polyp patients, combined with up-regulation in 65% of patients with CRC, confirmed the potential usefulness of ANRIL as a prognostic biomarker (AUC 0.95; P < 0.0001). BANCR relative expression analysis illustrated significant up-regulation in polyp (P < 0.04) and tumoural participants (P < 0.03) compared with normal control individuals. The expression patterns of ANRIL and BANCR in polyp cases were significantly correlated according to correlation analysis (r = 0.45, P < 0.045). ANRIL expression patterns in stool specimens of polyp and tumour cases supported the use of ANRIL as a prognostic biomarker for screening patients in the early stages of CRC. Up-regulation of BANCR in pre-cancerous lesions as well as down-regulation of ANRIL may also be a specific marker pair for easy, convenient and fast CRC prognosis.
Topics: Biomarkers; Colorectal Neoplasms; Humans; Prognosis; RNA, Long Noncoding
PubMed: 35296566
DOI: 10.1266/ggs.21-00102 -
Cellular and Molecular Life Sciences :... Nov 2021Exposures from the external and internal environments lead to the modification of genomic DNA, which is implicated in the cause of numerous diseases, including cancer,... (Review)
Review
Exposures from the external and internal environments lead to the modification of genomic DNA, which is implicated in the cause of numerous diseases, including cancer, cardiovascular, pulmonary and neurodegenerative diseases, together with ageing. However, the precise mechanism(s) linking the presence of damage, to impact upon cellular function and pathogenesis, is far from clear. Genomic location of specific forms of damage is likely to be highly informative in understanding this process, as the impact of downstream events (e.g. mutation, microsatellite instability, altered methylation and gene expression) on cellular function will be positional-events at key locations will have the greatest impact. However, until recently, methods for assessing DNA damage determined the totality of damage in the genomic location, with no positional information. The technique of "mapping DNA adductomics" describes the molecular approaches that map a variety of forms of DNA damage, to specific locations across the nuclear and mitochondrial genomes. We propose that integrated comparison of this information with other genome-wide data, such as mutational hotspots for specific genotoxins, tumour-specific mutation patterns and chromatin organisation and transcriptional activity in non-cancerous lesions (such as nevi), pre-cancerous conditions (such as polyps) and tumours, will improve our understanding of how environmental toxins lead to cancer. Adopting an analogous approach for non-cancer diseases, including the development of genome-wide assays for other cellular outcomes of DNA damage, will improve our understanding of the role of DNA damage in pathogenesis more generally.
Topics: Animals; Chromosome Mapping; DNA; DNA Damage; Genome; Genome-Wide Association Study; Genomics; Humans; Mutation; Neoplasms
PubMed: 34463773
DOI: 10.1007/s00018-021-03923-6