Detection Of Colon Cancer Research Articles

An Optimized Deep Learning Based Optimization Algorithm for the Detection of Colon Cancer Using Deep Recurrent Neural Networks

Colon cancer is the second leading dreadful disease-causing death. The challenge in the colon cancer detection is the accurate identification of the lesion at the early stage such that mortality and morbidity can be reduced. In this work, a colon cancer classification method is identified out using Dragonfly-based water wave optimization (DWWO) based deep recurrent neural network. Initially, the input cancer images subjected to carry a pre-processing, in which outer artifacts are removed. The pre-processed image is forwarded for segmentation then the images are converted into segments using Generative adversarial networks (GAN). The obtained segments are forwarded for attribute selection module, where the statistical features like mean, variance, kurtosis, entropy, and textual features, like LOOP features are effectively extracted. Finally, the colon cancer classification is solved by using the deep RNN, which is trained by the proposed Dragonfly-based water wave optimization algorithm. The proposed DWWO algorithm is developed by integrating the Dragonfly algorithm and water wave optimization.

Awareness of the Adult Population Toward Colorectal Cancer in Qassim Region, Saudi Arabia.

Introduction Colon and rectal cancer are usually grouped together as they share similar features. They are the third most common cancer in the world and the second most common cancer in the Kingdom of Saudi Arabia (KSA). Awareness and preventive screening programs play a vital role in early diagnosis and improving the survival rate of such patients.This study aimed to assess the awareness and knowledge of colorectal cancer(CRC) among the adult population in the Qassim region of Saudi Arabia. Methods This is a cross-sectional study conducted among the adult population living in the Qassim region of Saudi Arabia. A self-administered questionnaire was disseminated among the adult population using WhatsApp and Twitter. The questionnaire contained demographic data and questions to assess the awareness, knowledge, and attitude regarding CRC risk factors and complications and the importance of screening. Results A total of 431 respondents were involved. The most common symptoms of colon cancer were abdominal painand change in bowel habits. Risk factors associated with colon cancer were inflammatory bowel diseaseand fatty food. Colonoscopy was the most dominant choice for the detection of colon cancer. Conclusion The awareness of the adult population toward CRCwas deficient. Better awareness can be predicted among individuals with better education. Addressing the gaps in awareness is vital to alleviate fears and misconceptions surrounding this disease.

Retracted: Value of MSCT plus MRI in the Detection of Colon Cancer.

[This retracts the article DOI: 10.1155/2022/6507865.].

Biomedical Image Analysis for Colon and Lung Cancer Detection Using Tuna Swarm Algorithm With Deep Learning Model

Biomedical Image Analysis for Colon and Lung Cancer Detection Using Tuna Swarm Algorithm With Deep Learning Model

The Expression of LINC00673 in Colon Cancer and the Melatonin's Effect on Its Expression

Objective: In this study, the cell lines SW620 and DLD-1 which belongs to colon cancer were employed with the aim of observing melatonin's effect on the growth of colon cancer cells and exploring the influence of melatonin about the autophagy of colon cancer. Besides. It uses qRT-PCR to identify LINCRNA00673 (LINC00673) expression in colon cancer and the influence of melatonin on LINC00673 expression.For the purpose of providing a fundamental theory for melatonin in anti-cancer therapy and also presenting as a biomarker for the early detection of colon cancer. Method: 1) Cultivating the colon cancer cell lines SW620 and DLD-1, and subdividing them in accordance with distinct concentrations of melatonin: 0 mM, 1mM, and 2mM, to detect the impact of melatonin for migration and growth of colon cancer cells by Cell Scratch Test and Colony Formation Assay.2) Detecting the expression level of autophagy-related molecules in colon cancer cells in response to melatonin employing Western Blot Assay. 3) Utilizing QT-PCR, extract RNA from two groups of cells and detect the expression of LINC00673 and LINC00673 following melatonin treatment. Result: 1) The Cell Scratch test and Colony Formation Assay indicated that compared with the normal group, the scratch healing of SW620 cells and DLD-1 cells decreased. Furthermore, following melatonin administration, the number of cell colonies decreased. And the results are statistically significant. 2) Western blotting results illustrated that after the interferon of melatonin, the expression level of the autophagy-related molecule Beclin-1 increased significantly, whereas the expression of P62 decreased. 3) QT-PCR results demonstrated that the expression of LINC00673 in colon cancer cells SW620 and DLD-1 was significantly increased. Moreover, melatonin had an inhibiting effect on it. Conclusion: These in vitro experiments managed to prove that melatonin possesses a powerful anti-colon cancer effect and can inhibit the growth and spread of colon cancer. Western blotting indicated that melatonin can inhibit colon cancer through autophagy. Besides which, it is discovered that melatonin may exert its anticancer effect by influencing LINC00673.

The Importance of Evaluating Serum Levels of Tumor Markers M2-PK and Inhibin A in Patients Undergoing Colonoscopy.

Despite the use of colonoscopy to detect colon cancer due to its aggressiveness, high cost, and lack of patient compliance, the use of laboratory tests with high accuracy and sensitivity, such as tumor marker M2-PK and Inhibin A is recommended and can be effective for early diagnosis and screening of patients in the early stages. We studied 46 patients admitted it the gastrointestinal ward of Amir al Momenin Hospital and 45 normal (age and sex-matched) subjects as a control group (case-control and retrospective studies). Before the colonoscopy, the level of tumor marker M2-PK in the stool sample and the serum level of Inhibin A were evaluated in patients and the control group. The level of tumor marker M2-PK was significantly higher in the group with hyperplastic polyps and colon cancer (P < .001) than in the control group. At the same time, there was no significant difference in Inhibin A level (P = .054). In the hyperplastic polyps group 73% and in the colorectal cancer group 27% had a positive immunochemical fecal occult blood (IFOBT) result, significantly higher than the control group (P < .001). Evaluation of the level of tumor marker M2-PK in the stool sample in association with the three-time iFOBT test method may be suggested as a quick and noninvasive method for screening and diagnosis of polyps and early stages of colon cancer.

Lack of information on colorectal cancer in general public might be the reason for low response rates to screening progam

Introduction: In Croatia, about 3,500 patients are diagnosed with a colorectal carcinoma annually. The National Program for Early Detection of Colon Cancer has been implemented in Croatia since 2007. The program’s primary goal is todetect colon cancer at an early stage of the disease, which should reduce the incidence and mortality from colon cancer. Aim: To determine the interest of the target group in participating in the program, reasons for not responding to the program, and their knowledge about it altogether. Materials and methods: Surveying 50 patients within the program’s target group about their response to and understanding of reasons for screening. Results: Although 84% of the participants know about the program, the response rate was unsatisfactory. Conclusion: Most of the respondents who suffered from a colorectal carcinoma received a call but did not respond due to the absence of symptoms, which correlates with a low level of knowledge on colorectal carcinoma.

Abstract IA018: Microbiome and intestinal barriers in colon cancer

Abstract Multiple factors (e.g., risk genes, environmental triggers, immunity, and microbiome) contribute to the pathogenesis of colorectal cancer (CRC). The intestinal dysbiosis and disruption of barrier function are common manifestations of CRC. Vitamin D/Vitamin D receptor (VDR) is known to protects against dysbiosis and tumorigenesis. Considering the multiple functional roles of VDR in the development of cancer, it is important to dissect the mechanisms by which host factors contribute to the intestinal barrier function in protecting cells from tumorigenesis. Here, we reported that in human tissue, reduced VDR in the colon of CRC patients is associated with the reduced intestinal barrier and enhanced bacterial translocation to tumors. VDR deficiency in the intestinal epithelial cells led to the reduced thickness of mucus layer and enhanced penetration of microbiota into the epithelial cells. VDR was a transcriptional factor regulating the genes of some tight junction (TJ) Claudins. The positively correlated status of VDR and Claudins (e.g., Claudin- 5 and10) contribute to the altered intestinal permeability in tumorigenesis. In the intestinal epithelial VDR conditionally knockout VDR∆IEC mice and myeloid VDR deletion mice, dysbiosis and inflammatory cytokines in the VDR deficient context further altered TJ proteins and manipulated the barrier functions in the development of colon cancer. Furthermore, we found that mice with colonic adenomatous polyposis coli (APC) mutation reduced thickness of mucus layer and enhanced penetration of microbiota into the epithelial cells, thus enhancing tumorigenesis in the colon. Studies on host factor regulation of microbiome and barriers could be potentially applied to risk assessment, early detection, and prevention of colon cancer. These studies will provide an additional avenue to restore microbiome and barrier functions for risk assessment and prevention of CRC and other cancers. Citation Format: Jun Sun, Jilei Zhang, Yinglin Xia, Yong-Guo Zhang. Microbiome and intestinal barriers in colon cancer [abstract]. In: Proceedings of the AACR Special Conference on Colorectal Cancer; 2022 Oct 1-4; Portland, OR. Philadelphia (PA): AACR; Cancer Res 2022;82(23 Suppl_1):Abstract nr IA018.

EPH144 The Impact of COVID-19 Pandemic on the Detection and Management of Cancer: A German Claims Data Analysis

The COVID-19 pandemic impacted the German healthcare system comprehensively. It is being widely discussed whether there was also a negative effect on cancer care, but evidence to quantify such an effect is lacking. This analysis aims to investigate the pandemic impact on the detection and management of breast cancer (BC), colon cancer (CC), lung cancer (LC), and prostate cancer (PC) in Germany. The analysis was based on anonymized claims data from a German statutory health insurance (AOK PLUS). Incidence was defined as at least one inpatient or two outpatient diagnoses of one of the four cancer types after a 24-month pre-index period without any cancer diagnoses. Standardized cumulative incidences and the proportions of patients with a metastatic diagnosis within the first three months after incident diagnosis were compared between 2019 and 2020. Differences in the time to treatment initiation (surgery, radiation, or systemic anticancer treatment) between 2019 and 2020 were investigated using Kaplan-Meier estimation. The standardized cumulative incidence of BC, CC, and LC in 2020 were significantly lower compared to 2019 (BC: 0.040%/0.038%; CC: 0.045%/0.037%; LC: 0.028%/0.025%; all p<0.001), while similar incidence of PC was observed (0.036%/0.036%; p>0.001). In all groups, the proportion of patients with metastases at the time of incident diagnosis did not differ considerably between the years. The time to treatment initiation after incident diagnosis did not differ significantly between 2019 and 2020 in the investigated populations. In incident BC/CC/LC/PC patients, the median time to therapy start was 42/47/59/90 days in 2019 versus 40/47/56/94 days in 2020 (HR [95%-CI]: 1.01 [0.92-1.10]/1.02 [0.92-1.3]/1.06 [0.94-1.20]/0.97 [0.88-1.07]). Significantly fewer patients were diagnosed with BC, CC, and LC in 2020 compared to 2019, possibly due to lower cancer screening rates. However, cancer therapy initiation and timing in newly diagnosed patients in 2020 were not substantially different from the pre-pandemic year.

CJT-DEO: Condorcet’s Jury Theorem and Differential Evolution Optimization based ensemble of deep neural networks for pulmonary and Colorectal cancer classification

Cancer is one of the most dangerous diseases globally, causing adverse effects on human life, with early detection and treatment planning being crucial for patients. Amongst different malignancies, Lung and Colorectal cancer cause the first and second most cancer deaths in the world, respectively. In this study, the authors aim to analyze LC25000 histopathological image dataset for lung and colon cancer detection. The fundamental goal of the proposed research is to leverage the ensemble learning approach to improve the classification performance of deep learning models. Many previous studies have proposed several ensemble methods and weighting schemes. However, none of them optimized the assigned weights using a meta-heuristic-based approach as per our best knowledge. The authors have applied Differential Evolution optimization to optimize and find the optimal assigned weights to the classifiers while training the ensemble model. In addition, a novel approach to ensemble base learners with majority voting based on Condorcet’s Jury Theorem has also been proposed. This proposed method has been shown to save a lot of computational efforts by eliminating the training procedure of meta-learners. Besides this, the authors also demonstrated that Condorcet’s Jury Theorem holds while ensembling the N number of classifiers in Neural Networks. Our proposed method and experimental results outperformed compared to the state-of-the-art with the optimized ensemble model showing an accuracy of 99.78% and Condorcet’s Jury Theorem-based ensemble model 99.88% on 5-class classification.

A Framework for Lung and Colon Cancer Diagnosis via Lightweight Deep Learning Models and Transformation Methods.

Among the leading causes of mortality and morbidity in people are lung and colon cancers. They may develop concurrently in organs and negatively impact human life. If cancer is not diagnosed in its early stages, there is a great likelihood that it will spread to the two organs. The histopathological detection of such malignancies is one of the most crucial components of effective treatment. Although the process is lengthy and complex, deep learning (DL) techniques have made it feasible to complete it more quickly and accurately, enabling researchers to study a lot more patients in a short time period and for a lot less cost. Earlier studies relied on DL models that require great computational ability and resources. Most of them depended on individual DL models to extract features of high dimension or to perform diagnoses. However, in this study, a framework based on multiple lightweight DL models is proposed for the early detection of lung and colon cancers. The framework utilizes several transformation methods that perform feature reduction and provide a better representation of the data. In this context, histopathology scans are fed into the ShuffleNet, MobileNet, and SqueezeNet models. The number of deep features acquired from these models is subsequently reduced using principal component analysis (PCA) and fast Walsh-Hadamard transform (FHWT) techniques. Following that, discrete wavelet transform (DWT) is used to fuse the FWHT's reduced features obtained from the three DL models. Additionally, the three DL models' PCA features are concatenated. Finally, the diminished features as a result of PCA and FHWT-DWT reduction and fusion processes are fed to four distinct machine learning algorithms, reaching the highest accuracy of 99.6%. The results obtained using the proposed framework based on lightweight DL models show that it can distinguish lung and colon cancer variants with a lower number of features and less computational complexity compared to existing methods. They also prove that utilizing transformation methods to reduce features can offer a superior interpretation of the data, thus improving the diagnosis procedure.

High incidence combination of multiple primary malignant tumors of the digestive system.

Clinical reports of multiple primary malignant tumors (MPMTs) in the digestive system are increasing. In China, although the survival rate of patients with MPMTs is increasing, the quality of life is very low. Many patients have reached the advanced stage when the second primary tumor is found, resulting in no early intervention and treatment. This is due to the misunderstanding of MPMTs by clinicians, who treat such tumors as metastases. Therefore, before a patient has a second primary tumor, doctors should understand some common combinations of digestive system MPMTs to provide clinical guidance to the patient. To explore the high incidence combination of digestive system MPMTs under heterochronism and synchronization. A total of 1902 patients with MPMTs at Peking Union Medical College Hospital were analyzed retrospectively. They were divided into metachronous MPMT and synchronous MPMT groups, and then the high incidence combinations of the first primary cancer and the second primary cancer in metachronous cancer and synchronous cancer were sorted. Sex and age differences between metachronous and synchronous tumors were tested by the chi square test and t test, respectively. A P value < 0.05 was considered as statistically significant, and SPSS version 26.0 (SPSS Inc., Chicago, Illinois, United States) was used for statistical analysis. Among the 1902 patients with MPMTs confirmed by pathology, 1811 (95.2%) cases were secondary primary cancers, 89 (4.7%) cases were tertiary primary cancers, and 2 (0.1%) cases were quaternary primary cancers. Most (88.2%) of the secondary primary cancers were identified as metachronous multiple primary cancers six months after diagnosis of the first primary cancer. The top ten most common MPMTs in the first primary cancer group ranged from high to low as follows: Breast cancer, thyroid cancer, nonuterine cancer, lung cancer, colon cancer, kidney cancer, uterine cancer, bladder cancer, rectal cancer, and gastric cancer. The highest incidence rate of the first primary cancer in male metachronous cancer was lung cancer (11.6%), the highest incidence rate of the second primary cancer was still lung cancer (24.9%), the highest incidence rate of the first primary cancer in female metachronous cancer was breast cancer (32.7%), and the highest incidence rate of the second primary cancer was lung cancer (20.8%). Among them, breast cancer, nonuterine cancer and uterine cancer were female-specific malignant tumor types, and thyroid cancer also accounted for 79.6% of female patients. The top five metachronous cancer combinations, independent of female-specific malignant tumor types and thyroid cancer, were colon cancer and lung cancer (26 cases), kidney cancer and lung cancer (25 cases), rectal cancer and lung cancer (20 cases), gastric cancer and lung cancer (17 cases), and bladder cancer and lung cancer (17 cases). The most common synchronous cancer combination was colon cancer and rectal cancer (15 cases). Screening for lung cancer should be performed six months after the detection of colon cancer while rectal cancer screening should be performed within six months.

Bacterial translocation and barrier dysfunction enhance colonic tumorigenesis

In the development of colon cancer, the intestinal dysbiosis and disruption of barrier function are common manifestations. In the current study, we hypothesized that host factors, e.g., vitamin D receptor deficiency or adenomatous polyposis coli (APC) mutation, contribute to the enhanced dysbiosis and disrupted barrier in the pathogenesis of colorectal cancer (CRC). Using the human CRC database, we found enhanced tumor-invading bacteria and reduced colonic VDR expression, which was correlated with a reduction of Claudin-10 mRNA and protein. In the colon of VDRΔIEC mice, deletion of intestinal epithelial VDR led to lower protein of tight junction protein Claudin-10. Lacking VDR and a reduction of Claudin-10 are associated with an increased number of tumors in the mice without myeloid VDR. Intestinal permeability was significantly increased in the mice with myeloid VDR conditional deletion. Further, mice with conditional colonic APC mutation showed reduced mucus layer, enhanced bacteria in tumors, and loss of Claudin-10. Our data from human samples and colon cancer models provided solid evidence- on the host factor regulation of bacterial translocation and dysfunction on barriers in colonic tumorigenesis. Studies on the host factor regulation of microbiome and barriers could be potentially applied to risk assessment, early detection, and prevention of colon cancer.

S228 Multimodal Analysis of cfDNA Methylation Sequencing Improves Early Colon Cancer Detection

S228 Multimodal Analysis of cfDNA Methylation Sequencing Improves Early Colon Cancer Detection

Transfer Learning Approach and Nucleus Segmentation with MedCLNet Colon Cancer Database.

Machine learning has been recently used especially in the medical field. In the diagnosis of serious diseases such as cancer, deep learning techniques can be used to reduce the workload of experts and to produce quick solutions. The nuclei found in the histopathology dataset are an essential parameter in disease detection. The nucleus segmentation was performed using the colorectal histology MNIST dataset for nucleus detection in this study. The graph theory, PSO, watershed, and random walker algorithms were used for the segmentation process. In addition, we present the 10-class MedCLNet visual dataset consisting of the NCT-CRC-HE-100K dataset, LC25000 dataset, and GlaS dataset that can be used in transfer learning studies from deep learning techniques. The study proposes a transfer learning technique using the MedCLNet database. Deep neural networks pre-trained with the proposed transfer learning method were used in the classification with the colorectal histology MNIST dataset in the experimental process. DenseNet201, DenseNet169, InceptionResNetV2, InceptionV3, ResNet152V2, ResNet101V2, and Xception deep learning algorithms were used in transfer learning and the classification studies. The proposed approach was analyzed before and after transfer learning with different methods (DenseNet169 + SVM, DenseNet169 + GRU). In the performance measurement, using the colorectal histology MNIST dataset, 94.29% accuracy was obtained in the DenseNet169 model, which was initiated with random weights in the multi-classification study, and 95.00% accuracy after transfer learning was applied. In comparison with the results obtained from empirical studies, it was demonstrated that the proposed method produced satisfactory outcomes. The application is expected to provide a secondary evaluation for physicians in colon cancer detection and the segmentation.

Potential diagnostic value of the hematological parameters lymphocyte-monocyte ratio and hemoglobin-platelet ratio for detecting colon cancer.

ObjectiveTo evaluate the efficacy of using the lymphocyte–monocyte ratio (LMR), hemoglobin–platelet ratio (HPR), and carcinoembryonic antigen (CEA) levels alone or in combination for diagnosing colon cancer.MethodsWe assessed 124 consecutive patients who were pathologically diagnosed with colon cancer and 131 patients who were diagnosed with benign colon tumors in this retrospective study. We then analyzed correlations between LMR, HPR, and clinicopathological findings. The diagnostic values of LMR, HPR, and CEA alone or in combination in colon cancer patients were evaluated by receiver operating characteristic curves.ResultsThe median LMR, HPR, and CEA values in colon cancer patients showed significant correlation with the depth of tumor invasion, lymph node metastasis, and TNM stage. Moreover, there was a significant difference in HPR between patients with tumor size ≥5 cm and those with tumor size <5 cm. Compared with LMR, HPR, or CEA alone, combinations of CEA with LMR, CEA with HPR, and HPR with LMR all had higher area under the curve values, among which the combination of all three (LMR, HPR, and CEA) had the highest area under the curve.ConclusionThe combination of LMR, HPR, and CEA may be a valuable indicator for monitoring colon cancer.

An Efficient Deep Learning Approach for Colon Cancer Detection

Colon cancer is the second most common cause of cancer death in women and the third most common cause of cancer death in men. Therefore, early detection of this cancer can lead to lower infection and death rates. In this research, we propose a new lightweight deep learning approach based on a Convolutional Neural Network (CNN) for efficient colon cancer detection. In our method, the input histopathological images are normalized before feeding them into our CNN model, and then colon cancer detection is performed. The efficiency of the proposed system is analyzed with publicly available histopathological images database and compared with the state-of-the-art existing methods for colon cancer detection. The result analysis demonstrates that the proposed deep model for colon cancer detection provides a higher accuracy of 99.50%, which is considered the best accuracy compared with the majority of other deep learning approaches. Because of this high result, the proposed approach is computationally efficient.

Automated Detection and Characterization of Colon Cancer with Deep Convolutional Neural Networks.

Colon cancer is a momentous reason for illness and death in people. The conclusive diagnosis of colon cancer is made through histological examination. Convolutional neural networks are being used to analyze colon cancer via digital image processing with the introduction of whole-slide imaging. Accurate categorization of colon cancers is necessary for capable analysis. Our objective is to promote a system for detecting and classifying colon adenocarcinomas by applying a deep convolutional neural network (DCNN) model with some preprocessing techniques on digital histopathology images. It is a leading cause of cancer-related death, despite the fact that both traditional and modern methods are capable of comparing images that may encompass cancer regions of various sorts after looking at a significant number of colon cancer images. The fundamental problem for colon histopathologists is differentiating benign from malignant illnesses to having some complicated factors. A cancer diagnosis can be automated through artificial intelligence (AI), enabling us to appraise more patients in less time and at a decreased cost. Modern deep learning (MDL) and digital image processing (DIP) approaches are used to accomplish this. The results indicate that the proposed structure can accurately analyze cancer tissues to a maximum of 99.80%. By implementing this approach, medical practitioners will establish an automated and reliable system for detecting various forms of colon cancer. Moreover, CAD systems will be built in the near future to extract numerous aspects from colonoscopic images for use as a preprocessing module for colon cancer diagnosis.

Design of a Split Ring Resonator Integrated with On‐Chip Terahertz Waveguides for Colon Cancer Detection

AbstractFinite element method (FEM) simulations (employing ANSYS High Frequency Structure Simulator, HFSS) are used to investigate the response of terahertz (THz) frequency range split‐ring resonators (SRRs) integrated with on‐chip THz waveguides to cancerous tissues. Two‐port S‐parameter simulations are performed to obtain the transmission spectra (S21) of a planar Goubau line (PGL) integrated with an SRR. Permittivity and loss tangent of the colonic tissues are both taken into account in the numerical simulation. The transmission spectra of the SRR integrated PGL are obtained for cancerous and healthy tissues in close proximity to the SRR, and it is found that they can be distinguished by the resonant frequency shift of the SRR induced by dielectric loading. The electric field distribution and magnitude near the SRR for various capacitive gap widths of SRR are investigated to understand how the gap width affects the maximum electric field magnitude and the vertical extent of the electric field in the gap area. The simulated imaging of colonic tissue consisting of healthy and cancerous tissues using the SRR integrated PGL device with a protective layer on it is performed, showing how the technique could in principle be used to distinguish tumor margins with realistic THz dielectric parameters.

Novel multiplex stool-based assay for the detection of early-stage colon cancer in a Chinese population.

BACKGROUNDStool DNA (sDNA) methylation analysis is a promising, noninvasive approach for colorectal cancer screening; however, reliable biomarkers for detecting early-stage colon cancer (ECC) are lacking, particularly in the Chinese population.AIMTo identify a novel stool-based assay that can improve the effectiveness of ECC screening.METHODSA blinded case-control study was performed using archived stool samples from 125 ECC patients, and 125 control subjects with normal colonoscopy. The cohort was randomly divided into training and test sets at a 1.5:1 ratio. Targeted bisulfite sequencing (TBSeq) was conducted on five pairs of preoperative and postop-erative sDNA samples from ECC patients to identify DNA methylation biomarkers, which were validated using pyrosequencing. By logistic regression analysis, a multiplex stool-based assay was developed in the training set, and the detection performance was further assessed in the test set and combined set. The χ2 test was used to investigate the association of detection sensitivity with clinico-pathological features.RESULTSFollowing TBSeq, three hypermethylated cytosine-guanine sites were selected as biomarkers, including paired box 8, Ras-association domain family 1 and secreted frizzled-related protein 2, which differed between the groups and were involved in important cancer pathways. An sDNA panel containing the three biomarkers was constructed with a logistic model. Receiver operating characteristic (ROC) analysis revealed that this panel was superior to the fecal immunochemical test (FIT) or serum carcinoembryonic antigen for the detection of ECC. We further found that the combination of the sDNA panel with FIT could improve the screening effectiveness. In the combined set, the sensitivity, specificity and area under the ROC curve for this multiplex assay were 80.0%, 93.6% and 0.918, respectively, and the performance remained excellent in the subgroup analysis by tumor stage. In addition, the detection sensitivity did not differ with tumor site, tumor stage, histological differentiation, age or sex, but was significantly higher in T4 than in T1-3 stage tumors (P = 0.041).CONCLUSIONWe identified a novel multiplex stool-based assay combining sDNA methylation biomarkers and FIT, which could detect ECC with high sensitivity and specificity throughout the colon, showing a promising application perspective.