Classification Of Polyps Research Articles

A generative flow-based model for volumetric data augmentation in 3D deep learning for computed tomographic colonography

PurposeDeep learning can be used for improving the performance of computer-aided detection (CADe) in various medical imaging tasks. However, in computed tomographic (CT) colonography, the performance is limited by the relatively small size and the variety of the available training datasets. Our purpose in this study was to develop and evaluate a flow-based generative model for performing 3D data augmentation of colorectal polyps for effective training of deep learning in CADe for CT colonography.MethodsWe developed a 3D-convolutional neural network (3D CNN) based on a flow-based generative model (3D Glow) for generating synthetic volumes of interest (VOIs) that has characteristics similar to those of the VOIs of its training dataset. The 3D Glow was trained to generate synthetic VOIs of polyps by use of our clinical CT colonography case collection. The evaluation was performed by use of a human observer study with three observers and by use of a CADe-based polyp classification study with a 3D DenseNet.ResultsThe area-under-the-curve values of the receiver operating characteristic analysis of the three observers were not statistically significantly different in distinguishing between real polyps and synthetic polyps. When trained with data augmentation by 3D Glow, the 3D DenseNet yielded a statistically significantly higher polyp classification performance than when it was trained with alternative augmentation methods.ConclusionThe 3D Glow-generated synthetic polyps are visually indistinguishable from real colorectal polyps. Their application to data augmentation can substantially improve the performance of 3D CNNs in CADe for CT colonography. Thus, 3D Glow is a promising method for improving the performance of deep learning in CADe for CT colonography.

Endoscopic Recognition and Management Strategies for Malignant Colorectal Polyps: Recommendations of the US Multi-Society Task Force on Colorectal Cancer.

Endoscopic Recognition and Management Strategies for Malignant Colorectal Polyps: Recommendations of the US Multi-Society Task Force on Colorectal Cancer.

Duodenal Adenomas and Cancer in MUTYH-associated Polyposis: An International Cohort Study

Duodenal Adenomas and Cancer in MUTYH-associated Polyposis: An International Cohort Study

Real-time artificial intelligence–based histologic classification of colorectal polyps with augmented visualization

Artificial intelligence (AI)-based computer-aided diagnostic (CADx) algorithms are a promising approach for real-time histology (RTH) of colonic polyps. Our aim is to present a novel in situ CADx approach that seeks to increase transparency and interpretability of results by generating an intuitive augmented visualization of the model's predicted histology over the polyp surface. We developed a deep learning model using semantic segmentation to delineate polyp boundaries and a deep learning model to classify subregions within the segmented polyp. These subregions were classified independently and were subsequently aggregated to generate a histology map of the polyp's surface. We used 740 high-magnification narrow-band images from 607 polyps in 286 patients and over 65,000 subregions to train and validate the model. The model achieved a sensitivity of .96, specificity of .84, negative predictive value (NPV) of .91, and high-confidence rate (HCR) of .88, distinguishing 171 neoplastic polyps from 83 non-neoplastic polyps of all sizes. Among 93 neoplastic and 75 non-neoplastic polyps≤5mm, the model achieved a sensitivity of .95, specificity of .84, NPV of .91, and HCR of .86. The CADx model is capable of accurately distinguishing neoplastic from non-neoplastic polyps and provides a histology map of the spatial distribution of localized histologic predictions along the delineated polyp surface. This capability may improve interpretability and transparency of AI-based RTH and offer intuitive, accurate, and user-friendly guidance in real time for the clinical management and documentation of optical histology results.

Automated Activity Estimation of the Cold-Water Coral Lophelia pertusa by Multispectral Imaging and Computational Pixel Classification

AbstractThe cold-water coral Lophelia pertusa builds up bioherms that sustain high biodiversity in the deep ocean worldwide. Photographic monitoring of the polyp activity represents a helpful tool to characterize the health status of the corals and to assess anthropogenic impacts on the microhabitat. Discriminating active polyps from skeletons of white Lophelia pertusa is usually time consuming and error prone due to their similarity in color in common red–green–blue (RGB) camera footage. Acquisition of finer-resolved spectral information might increase the contrast between the segments of polyps and skeletons, and therefore could support automated classification and accurate activity estimation of polyps. For recording the needed footage, underwater multispectral imaging systems can be used, but they are often expensive and bulky. Here we present results of a new, lightweight, compact, and low-cost deep-sea tunable LED-based underwater multispectral imaging system (TuLUMIS) with eight spectral channels. A branch of healthy white Lophelia pertusa was observed under controlled conditions in a laboratory tank. Spectral reflectance signatures were extracted from pixels of polyps and skeletons of the observed coral. Results showed that the polyps can be better distinguished from the skeleton by analysis of the eight-dimensional spectral reflectance signatures compared to three-channel RGB data. During a 72-h monitoring of the coral with a half-hour temporal resolution in the laboratory, the polyp activity was estimated based on the results of the multispectral pixel classification using a support vector machine (SVM) approach. The computational estimated polyp activity was consistent with that of the manual annotation, which yielded a correlation coefficient of 0.957.

Potential applications of artificial intelligence in colorectal polyps and cancer: Recent advances and prospects.

Since the advent of artificial intelligence (AI) technology, it has been constantly studied and has achieved rapid development. The AI assistant system is expected to improve the quality of automatic polyp detection and classification. It could also help prevent endoscopists from missing polyps and make an accurate optical diagnosis. These functions provided by AI could result in a higher adenoma detection rate and decrease the cost of polypectomy for hyperplastic polyps. In addition, AI has good performance in the staging, diagnosis, and segmentation of colorectal cancer. This article provides an overview of recent research focusing on the application of AI in colorectal polyps and cancer and highlights the advances achieved.

An improved deep learning approach and its applications on colonic polyp images detection

BackgroundColonic polyps are more likely to be cancerous, especially those with large diameter, large number and atypical hyperplasia. If colonic polyps cannot be treated in early stage, they are likely to develop into colon cancer. Colonoscopy is easily limited by the operator’s experience, and factors such as inexperience and visual fatigue will directly affect the accuracy of diagnosis. Cooperating with Hunan children’s hospital, we proposed and improved a deep learning approach with global average pooling (GAP) in colonoscopy for assisted diagnosis. Our approach for assisted diagnosis in colonoscopy can prompt endoscopists to pay attention to polyps that may be ignored in real time, improve the detection rate, reduce missed diagnosis, and improve the efficiency of medical diagnosis.MethodsWe selected colonoscopy images from the gastrointestinal endoscopy room of Hunan children’s hospital to form the colonic polyp datasets. And we applied the image classification method based on Deep Learning to the classification of Colonic Polyps. The classic networks we used are VGGNets and ResNets. By using global average pooling, we proposed the improved approaches: VGGNets-GAP and ResNets-GAP.ResultsThe accuracies of all models in datasets exceed 98%. The TPR and TNR are above 96 and 98% respectively. In addition, VGGNets-GAP networks not only have high classification accuracies, but also have much fewer parameters than those of VGGNets.ConclusionsThe experimental results show that the proposed approach has good effect on the automatic detection of colonic polyps. The innovations of our method are in two aspects: (1) the detection accuracy of colonic polyps has been improved. (2) our approach reduces the memory consumption and makes the model lightweight. Compared with the original VGG networks, the parameters of our VGG19-GAP networks are greatly reduced.

A CNN CADx System for Multimodal Classification of Colorectal Polyps Combining WL, BLI, and LCI Modalities

Colorectal polyps are critical indicators of colorectal cancer (CRC). Blue Laser Imaging and Linked Color Imaging are two modalities that allow improved visualization of the colon. In conjunction with the Blue Laser Imaging (BLI) Adenoma Serrated International Classification (BASIC) classification, endoscopists are capable of distinguishing benign and pre-malignant polyps. Despite these advancements, this classification still prevails a high misclassification rate for pre-malignant colorectal polyps. This work proposes a computer aided diagnosis (CADx) system that exploits the additional information contained in two novel imaging modalities, enabling more informative decision-making during colonoscopy. We train and benchmark six commonly used CNN architectures and compare the results with 19 endoscopists that employed the standard clinical classification model (BASIC). The proposed CADx system for classifying colorectal polyps achieves an area under the curve (AUC) of 0.97. Furthermore, we incorporate visual explanatory information together with a probability score, jointly computed from White Light, Blue Laser Imaging, and Linked Color Imaging. Our CADx system for automatic polyp malignancy classification facilitates future advances towards patient safety and may reduce time-consuming and costly histology assessment.

Effect of interpolation on specular reflections in texture‐based automatic colonic polyp detection

AbstractReflections of LED light cause unwanted noise effects called specular reflection (SR) on colonoscopic images. The aim of this study was to seek answers to the following two questions. (a) How are the texture features used in automatic detection of polyps affected by the interpolation on specular reflections? (b) If they are affected does it really affect the classification performance? In order to answer these questions, we used 610 colonoscopy images, and divided each image into tiles whose sizes were 32‐by‐32 pixels. From these tiles, we selected the ones without any specular reflection. We added different shape and size specular reflections cropped from real images onto the reflection‐free tiles. We then used the nearest neighbors, bilinear and bicubic interpolation techniques on the tiles on which SRs were added. On these tiles we extracted 116 texture features using 3 second‐order approaches, and 4 first‐order statistics. First, we used paired sample t test. Second, we performed automatic classification of polyps and background using random forest and k nearest neighbors (k‐NN) approaches using the texture features for different combinations of specular reflections added on the tiles from the polyp or background. The results showed that depending on the size of specular reflection, interpolation can cause a significant difference between the texture features that were coming from reflection‐free tiles and the same tiles on which interpolation was performed. In addition, we note that bicubic interpolation may be preferred to eliminate specular reflection when texture features are used for background and polyp discrimination.

Decomposition of color wavelet with higher order statistical texture and convolutional neural network features set based classification of colorectal polyps from video endoscopy

Gastrointestinal cancer is one of the leading causes of death across the world. The gastrointestinal polyps are considered as the precursors of developing this malignant cancer. In order to condense the probability of cancer, early detection and removal of colorectal polyps can be cogitated. The most used diagnostic modality for colorectal polyps is video endoscopy. But the accuracy of diagnosis mostly depends on doctors' experience that is crucial to detect polyps in many cases. Computer-aided polyp detection is promising to reduce the miss detection rate of the polyp and thus improve the accuracy of diagnosis results. The proposed method first detects polyp and non-polyp then illustrates an automatic polyp classification technique from endoscopic video through color wavelet with higher-order statistical texture feature and Convolutional Neural Network (CNN). Gray Level Run Length Matrix (GLRLM) is used for higher-order statistical texture features of different directions (Ɵ = 0o, 45o, 90o, 135o). The features are fed into a linear support vector machine (SVM) to train the classifier. The experimental result demonstrates that the proposed approach is auspicious and operative with residual network architecture, which triumphs the best performance of accuracy, sensitivity, and specificity of 98.83%, 97.87%, and 99.13% respectively for classification of colorectal polyps on standard public endoscopic video databases.

Mo1678 EVALUATING RECOMMENDATIONS FOR SURVEILLANCE FOR PATIENTS WITH LOW RISK ADENOMAS

Significance: Current practice guidelines recommend surveillance colonoscopies among patients with adenomas removed and intervals vary based on polyp classification. High risk adenomas (HRAs) (≥1 cm, villous features, or high-grade dysplasia) are most likely to progress to cancer and have a recommended 3-year follow-up whereas low risk adenomas (LRAs) (1-2 tubular adenomas ≤1 cm), have a recommended surveillance interval of 5-10 years. Over-performance of surveillance colonoscopies in patients with LRAs increases medical risk and carries significant societal cost. This study examined surveillance recommendations in practice for low risk adenomas. A retrospective chart review in an academic health system of 284 participants [median age: 57.4 (41.6-75.2); 54.6% female] with 1-2 tubular adenomas ≤1 centimeter on index colonoscopy, and a completed surveillance colonoscopy, was performed. Data were collected from exams between January 2011-June 2019. Exclusion criteria included family history of colon cancer, or personal history of polyps, inflammatory bowel disease, adenomas, or colon cancer. The recommended follow-up interval at index colonoscopy and surveillance colonoscopy findings were abstracted. Almost half of participants were recommended to follow up in 5 years (43.3%, n=123). Other recommendations were: 3 years (18.7%), “await pathology results” (19.0%), 5-10 years (8.5%), 3-5 years depending on pathology (3.5%), <3 years (3.2%), 3-5 years (1.4%), 10 years (0.7%), and other (1.8%). The median duration (months) between colonoscopies was 60.07 (12.19 – 88.07). On surveillance, number of LRAs discovered varied: zero (52.1%), one (25.0%), two (12.3%), three (4.6%), four (3.2%), five (0.7%), six or more (1.4%), and uncertain (0.7%). Fifteen participants (5.3%) had one HRA and one (0.4%) had two. No colorectal cancer was detected. Variables significantly associated (p<0.05) with adenomas on surveillance exam included older age, recommendation for follow up after index exam, and worse surveillance exam Boston Bowel Preparation Score (BBPS). Variables with no significant association included sex, race, ethnicity, index exam BBPS, quantity of LRAs and hyperplastic polyps on index exam, LRA size, duration of time between index and surveillance exams, diabetes, smoking history, and body mass index. The majority of participants with LRAs at baseline did not develop HRAs between colonoscopies, supporting a longer surveillance interval. No participants developed colon cancer. Although the current recommended interval is 5-10 years, <10% had a surveillance recommendation of 10 years. Many were told to follow up before 5 years. The findings support the current guidelines of performing a surveillance exam after 5-10 years and suggest that there is low risk in waiting to perform a surveillance exam until closer to the 10 year mark.

Mo1648 PRACTICE PATTERNS OF POLYPECTOMY TECHNIQUES FOR THE RESECTION OF POLYPS BASED ON SIZE, MORPHOLOGY AND LOCATION - RESULTS FROM A MULTICENTER TRIAL.

Conventionally, resection with cold forceps and cold snare have been preferred for sub-centimeter polyps, whereas hot snare has been used for larger polyps. We examined the practice patterns and trends in methods of polyp resection as part of a large multi-center randomized control trial (RCT) focused on adenoma detection rates. Average risk patients presenting for screening and surveillance colonoscopies were prospectively enrolled in this trial evaluating ADR. The morphology, size and location of colon polyps were collected. Polyps were resected using cold forceps, cold snare (CSP), hot snare (HSP) or EMR. Based on the size, the polyps were divided into 1-9mm, 10-19mm and ≥ 20 mm. We compared the use of cold snare and hot snare use based on polyp location (right versus left colon), size (1-9mm, 10-19mm, >20mm) and polyp morphology based on Paris classification. Continuous variables were compared using one way analysis of variance, categorical variables compared using chi-square test. Of 810 patients enrolled thus far, (mean age 61.50 ± 9.33 years, 74% males) a total of 1808 polyps were resected during either screening (40%) or surveillance (60%) colonoscopy. A total of 1650 (91%) polyps found were 1-9mm, 126(7%) were 10-19mm and 25(1.3%) were >20mm. Overall, cold snare polypectomy (CSP) was used in 68% of polypectomies (1215/1808). Based on the polyp size, CSP was used in 70% of 1-9mm polyps, and 41% in polyps sized 10-19mm. Hot snare polypectomy (HSP) was used in 7% of total polypectomies (127/1808), including all (100%) resected polyps >20mm and 64 (51%) polyps 10-19mm. CSP and HSP usage was not statistically significant based on polyp location (right colon: 66.9% vs 70.3%) (left colon: 5.8% vs 10.1%) respectively. HSP was used for all > 20 mm polyps, and for 61.4% of all pedunculated polyps (Paris type 1p). For 10-19 mm polyps, there was no significant difference between CSP and HSP (overall 40.8 % vs 51.2%), or based on location (42.5% vs 46% for right colon and 36.8% vs 63.2% for left colon, p-value=0.106) or based on Paris classification for sessile polyps (Paris type Is and IIa) – 43% vs 48% (p=0.182) The majority (90%) of polyps detected during average risk screening/surveillance colonoscopy are 1-9 mm in size. Cold snare polypectomy is the resection method of choice for polyps < 10 mm, whereas hot snare is preferred for the resection of all large (≥20mm) and majority of the pedunculated polyps. For 10-19 mm polyps, there was no difference between the use of cold snare or hot snare polypectomy based either on polyp location or morphology (Paris Is and IIa).

Gastrointestinal polyp classification through empirical mode decomposition and neural features

This study describes an automated detection of polyp type as it is very important to determine the existence of dysplasia—a stage leading to the development of gastrointestinal cancer. The polyp-type classification is performed by a multiclass support vector machine from feature-fusion of bi-dimensional empirical mode decomposition (BEMD) and convolutional neural network (CNN). An extensive experiment is performed using standard datasets by extracted features from the individual technique as well as a fusion of features from BEMD and CNN. The fusion technique confirms satisfactory performance compared to other techniques with an accuracy of 98.94%. Moreover, it shows potentiality in precisely classifying some challenging polyps even though these are somehow confusing for human experts.

Optical imaging technology in colonoscopy: Is there a role for photometric stereo?

Colonoscopy screening for the detection and removal of colonic adenomas is central to efforts to reduce the morbidity and mortality of colorectal cancer. However, up to a third of adenomas may be missed at colonoscopy, and the majority of post-colonoscopy colorectal cancers are thought to arise from these. Adenomas have three-dimensional surface topographic features that differentiate them from adjacent normal mucosa. However, these topographic features are not enhanced by white light colonoscopy, and the endoscopist must infer these from two-dimensional cues. This may contribute to the number of missed lesions. A variety of optical imaging technologies have been developed commercially to enhance surface topography. However, existing techniques enhance surface topography indirectly, and in two dimensions, and the evidence does not wholly support their use in routine clinical practice. In this narrative review, co-authored by gastroenterologists and engineers, we summarise the evidence for the impact of established optical imaging technologies on adenoma detection rate, and review the development of photometric stereo (PS) for colonoscopy. PS is a machine vision technique able to capture a dense array of surface normals to render three-dimensional reconstructions of surface topography. This imaging technique has several potential clinical applications in colonoscopy, including adenoma detection, polyp classification, and facilitating polypectomy, an inherently three-dimensional task. However, the development of PS for colonoscopy is at an early stage. We consider the progress that has been made with PS to date and identify the obstacles that need to be overcome prior to clinical application.

SAGES TAVAC safety and efficacy analysis confocal laser endomicroscopy.

Confocal laser endomicroscopy (CLE) is a novel endoscopic adjunct that allows real-time in vivo histological examination of mucosal surfaces. By using intravenous or topical fluorescent agents, CLE highlights certain mucosal elements that facilitate an optical biopsy in real time. CLE technology has been used in different organ systems including the gastrointestinal tract. There has been numerous studies evaluating this technology in gastrointestinal endoscopy, our aim was to evaluate the safety, value, and efficacy of this technology in the gastrointestinal tract. The Society of American Gastrointestinal and Endoscopic Surgeons (SAGES) Technology and Value Assessment Committee (TAVAC) performed a PubMed/Medline database search of clinical studies involving CLE in May of 2018. The literature search used combinations of the keywords: confocal laser endomicroscopy, pCLE, Cellvizio, in vivo microscopy, optical histology, advanced endoscopic imaging, and optical diagnosis. Bibliographies of key references were searched for relevant studies not covered by the PubMed search. Case reports and small case series were excluded. The manufacturer's website was also used to identify key references. The United States Food and Drug Administration (U.S. FDA) Manufacturer And User facility and Device Experience (MAUDE) database was searched for reports regarding the device malfunction or injuries. The technology offers an excellent safety profile with rare adverse events related to the use of fluorescent agents. It has been shown to increase the detection of dysplastic Barrett's esophagus, gastric intraepithelial neoplasia/early gastric cancer, and dysplasia associated with inflammatory bowel disease when compared to standard screening protocols. It also aids in the differentiation and classification of colorectal polyps, indeterminate biliary strictures, and pancreatic cystic lesions. CLE has an excellent safety profile. CLE can increase the diagnostic accuracy in a number of gastrointestinal pathologies.

Deep Neural Networks approaches for detecting and classifying colorectal polyps

Deep Learning (DL) has attracted a lot of attention in the field of medical image analysis because of its higher performance in image classification when compared to previous state-of-the-art techniques. In addition, a recent meta-analysis found that the diagnostic performance of DL models is equivalent to that of health-care professionals. In this scenario, a lot of research using DL for polyp detection and classification have been published showing promising results in the last five years. Our work aims to review the most relevant studies from a technical point of view, focusing on the low-level details for the implementation of the DL models. To do so, this review analyzes the published research covering aspects like DL architectures, training strategies, data augmentation, transfer learning, or the features of the datasets used and their impact on the performance of the models. Additionally, comparative tables summarizing the main aspects analyzed in this review are publicly available at https://github.com/sing-group/deep-learning-colonoscopy.

Evaluation of a Deep Neural Network for Automated Classification of Colorectal Polyps on Histopathologic Slides

Histologic classification of colorectal polyps plays a critical role in screening for colorectal cancer and care of affected patients. An accurate and automated algorithm for the classification of colorectal polyps on digitized histopathologic slides could benefit practitioners and patients. To evaluate the performance and generalizability of a deep neural network for colorectal polyp classification on histopathologic slide images using a multi-institutional data set. This prognostic study used histopathologic slides collected from January 1, 2016, to June 31, 2016, from Dartmouth-Hitchcock Medical Center, Lebanon, New Hampshire, with 326 slides used for training, 157 slides for an internal data set, and 25 for a validation set. For the external data set, 238 slides for 179 distinct patients were obtained from 24 institutions across 13 US states. Data analysis was performed from April 9 to November 23, 2019. Accuracy, sensitivity, and specificity of the model to classify 4 major colorectal polyp types: tubular adenoma, tubulovillous or villous adenoma, hyperplastic polyp, and sessile serrated adenoma. Performance was compared with that of local pathologists' at the point of care identified from corresponding pathology laboratories. For the internal evaluation on the 157 slides with ground truth labels from 5 pathologists, the deep neural network had a mean accuracy of 93.5% (95% CI, 89.6%-97.4%) compared with local pathologists' accuracy of 91.4% (95% CI, 87.0%-95.8%). On the external test set of 238 slides with ground truth labels from 5 pathologists, the deep neural network achieved an accuracy of 87.0% (95% CI, 82.7%-91.3%), which was comparable with local pathologists' accuracy of 86.6% (95% CI, 82.3%-90.9%). The findings suggest that this model may assist pathologists by improving the diagnostic efficiency, reproducibility, and accuracy of colorectal cancer screenings.

Review of computational methods for the detection and classification of polyps in colonoscopy imaging

Computer-aided diagnosis (CAD) is a tool with great potential to help endoscopists in the tasks of detecting and histologically classifying colorectal polyps. In recent years, different technologies have been described and their potential utility has been increasingly evidenced, which has generated great expectations among scientific societies. However, most of these works are retrospective and use images of different quality and characteristics which are analysed off line. This review aims to familiarise gastroenterologists with computational methods and the particularities of endoscopic imaging, which have an impact on image processing analysis. Finally, the publicly available image databases, needed to compare and confirm the results obtained with different methods, are presented.

Efficacy and safety of three different endoscopic methods in treatment of 6–20 mm colorectal polyps

Objective: Endoscopic resection of colorectal polyps is widely established as the optimal method to manage precancerous lesions. But the optimal technique for removal of the polyps is uncertain. The aim of this study was to compare the efficacy and safety of three methods for the management of 6–20mm colorectal polyps.Methods: A prospective, randomised controlled trial was conducted at the 900TH Hospital of Joint Logistics Support Force in Fujian, China. Endoscopically diagnosed colorectal polyps, 6–20 mm in size, were randomly assigned to the cold snare polypectomy (CSP), cold snare endoscopic mucosal resection (CS-EMR) or endoscopic mucosal resection (EMR) group. After polypectomy, additional 3–5 forceps biopsies by leading narrow-band imaging (NBI) were performed at the base and margins of polypectomy sites to assess the presence of residual polyp tissue and all samples were sent for histopathological analysis to assess completeness of resection. Polypectomy timing, tissue retrieval and complications were recorded at the time of the procedure.Results: A total of 781 polyps in 404 patients were assessed and randomly assigned to each group. Of these, 763 polyps were finally analyzed based on the pathology results. The complete resection rates with CSP, CS-EMR and EMR were 81.6%, 94.1% and 95.5%, respectively (p < .001). The intraprocedural bleeding rate, immediately after polypectomy, was significantly higher in the CSP group than in the CS-EMR and EMR group (9.4% vs. 4.4% vs. 1.9%; p < .001). However,delayed bleeding was higher in the EMR group than in the CSP and CS-EMR group (2.6% vs. 1.2% vs. 0.8%, respectively; p = .215). In the multivariate analysis showed that the operation method, lesion size, morphology and the number of resection were independent risk factors for complete resection rate (CRR) (p < .05), but the location and pathological classification of polyps had no significant influence on CRR.Conclusions: CS-EMR is safe and effective in the treatment of 6–20 mm colorectal polyps. Especially for 6–15 mm non-pedunculated polyps, CS-EMR has a high histological complete resection rate comparable to EMR, and retains the low risk of delayed complications after polypectomy with cold snare. CS-EMR is expected to become a more valuable new cold-cutting technique after cold snare polypectomy.

Deep learning uncertainty and confidence calibration for the five-class polyp classification from colonoscopy.

There are two challenges associated with the interpretability of deep learning models in medical image analysis applications that need to be addressed: confidence calibration and classification uncertainty. Confidence calibration associates the classification probability with the likelihood that it is actually correct - hence, a sample that is classified with confidence X% has a chance of X% of being correctly classified. Classification uncertainty estimates the noise present in the classification process, where such noise estimate can be used to assess the reliability of a particular classification result. Both confidence calibration and classification uncertainty are considered to be helpful in the interpretation of a classification result produced by a deep learning model, but it is unclear how much they affect classification accuracy and calibration, and how they interact. In this paper, we study the roles of confidence calibration (via post-process temperature scaling) and classification uncertainty (computed either from classification entropy or the predicted variance produced by Bayesian methods) in deep learning models. Results suggest that calibration and uncertainty improve classification interpretation and accuracy. This motivates us to propose a new Bayesian deep learning method that relies both on calibration and uncertainty to improve classification accuracy and model interpretability. Experiments are conducted on a recently proposed five-class polyp classification problem, using a data set containing 940 high-quality images of colorectal polyps, and results indicate that our proposed method holds the state-of-the-art results in terms of confidence calibration and classification accuracy.