scholarly journals On Structural Entropy and Spatial Filling Factor Analysis of Colonoscopy Pictures

Entropy ◽  
2019 ◽  
Vol 21 (3) ◽  
pp. 256 ◽  
Author(s):  
Szilvia Nagy ◽  
Brigita Sziová ◽  
János Pipek
Keyword(s):  
Filling Factor ◽  
Characteristic Curve ◽  
Colorectal Polyps ◽  
Image Resolution ◽  
Characteristic Curves ◽  
Structural Entropy ◽  
Factor Map ◽  
Definition Of ◽  
Model Structures ◽  
Aided Diagnosis

Colonoscopy is the standard device for diagnosing colorectal cancer, which develops from little lesions on the bowel wall called polyps. The Rényi entropies-based structural entropy and spatial filling factor are two scale- and resolution-independent quantities that characterize the shape of a probability distribution with the help of characteristic curves of the structural entropy–spatial filling factor map. This alternative definition of structural entropy is easy to calculate, independent of the image resolution, and does not require the calculation of neighbor statistics, unlike the other graph-based structural entropies.The distant goal of this study was to help computer aided diagnosis in finding colorectal polyps by making the Rényi entropy based structural entropy more understood. The direct goal was to determine characteristic curves that can differentiate between polyps and other structure on the picture. After analyzing the distribution of colonoscopy picture color channels, the typical structures were modeled with simple geometrical functions and the structural entropy–spatial filling factor characteristic curves were determined for these model structures for various parameter sets. A colonoscopy image analying method, i.e., the line- or column-wise scanning of the picture, was also tested, with satisfactory matching of the characteristic curve and the image.

scholarly journals Application of Structural Entropy and Spatial Filling Factor in Colonoscopy Image Classification

Entropy ◽  
2021 ◽  
Vol 23 (8) ◽  
pp. 936
Author(s):  
Brigita Sziová ◽  
Szilvia Nagy ◽  
Zoltán Fazekas
Keyword(s):  
Filling Factor ◽  
Fuzzy Inference ◽  
System Development ◽  
Automatic Segmentation ◽  
Fuzzy Rule ◽  
Colorectal Polyps ◽  
True Negative ◽  
Metaheuristic Methods ◽  
Structural Entropy ◽  
Input Variables

For finding colorectal polyps the standard method relies on the techniques and devices of colonoscopy and the medical expertise of the gastroenterologist. In case of images acquired through colonoscopes the automatic segmentation of the polyps from their environment (i.e., from the bowel wall) is an essential task within computer aided diagnosis system development. As the number of the publicly available polyp images in various databases is still rather limited, it is important to develop metaheuristic methods, such as fuzzy inference methods, along with the deep learning algorithms to improve and validate detection and classification techniques. In the present manuscript firstly a fuzzy rule set is generated and validated. The former process is based on a statistical approach and makes use of histograms of the antecedents. Secondly, a method for selecting relevant antecedent variables is presented. The selection is based on the comparision of the histograms computed from the measured values for the training set. Then the inclusion of the Rényi-entropy-based structural entropy and the spatial filling factor into the set of input variables is proposed and assessed. The beneficial effect of including the mentioned structural entropy of the entropies from the hue and saturation (H and S) colour channels resulted in 65% true positive and 60% true negative rate of the classification for an advantageously selected set of antecedents when working with HSV images.

scholarly journals TOWARDS AN UNIFIED FRAMEWORK FOR VALIDATING DEEP LEARNING METHODS FOR COLORECTAL POLYPS: FIRST STEPS

2021 ◽  
Vol 108 (Supplement_3) ◽  
Author(s):  
L F Sánchez Peralta ◽  
J F Ortega Morán ◽  
Cr L Saratxaga ◽  
J B Pagador ◽  
A Picón ◽  
...  
Keyword(s):  
Deep Learning ◽  
Colorectal Polyps ◽  
Polyp Detection ◽  
Unified Framework ◽  
Endoscopic Image ◽  
Learning Techniques ◽  
Great Utility ◽  
Adenoma Detection ◽  
Definition Of ◽  
Basic Features

Abstract INTRODUCTION Deep learning techniques have significantly contributed to the field of medical imaging analysis. In case of colorectal cancer, they have shown a great utility for increasing the adenoma detection rate at colonoscopy, but a common validation methodology is still missing. In this study, we present preliminary efforts towards the definition of a validation framework. MATERIAL AND METHODS Different models based on different backbones and encoder-decoder architectures have been trained with a publicly available dataset that contains white light and NBI colonoscopy videos, with 76 different lesions from colonoscopy procedures in 48 human patients. A computer aided detection (CADe) demonstrator has been implemented to show the performance of the models. RESULTS This CADe demonstrator shows the areas detected as polyp by overlapping the predicted mask on the endoscopic image. It allows selecting the video to be used, among those from the test set. Although it only present basic features such as play, pause and moving to the next video, it easily loads the model and allows for visualization of results. The demonstrator is accompanied by a set of metrics to be used depending on the aimed task: polyp detection, localization and segmentation. CONCLUSIONS The use of this CADe demonstrator, together with a publicly available dataset and predefined metrics will allow for an easier and more fair comparison of methods. Further work is still required to validate the proposed framework.

scholarly journals Automatic detection and segmentation of adenomatous colorectal polyps during colonoscopy using Mask R-CNN

2020 ◽  
Vol 15 (1) ◽  
pp. 588-596 ◽  
Author(s):  
Jie Meng ◽  
Linyan Xue ◽  
Ying Chang ◽  
Jianguang Zhang ◽  
Shilong Chang ◽  
...  
Keyword(s):  
Deep Neural Networks ◽  
Alimentary Tract ◽  
Multiple Scale ◽  
Colorectal Polyps ◽  
Training Dataset ◽  
Cad System ◽  
Multicenter Trials ◽  
Testing Dataset ◽  
Adenoma Detection ◽  
Aided Diagnosis

AbstractColorectal cancer (CRC) is one of the main alimentary tract system malignancies affecting people worldwide. Adenomatous polyps are precursors of CRC, and therefore, preventing the development of these lesions may also prevent subsequent malignancy. However, the adenoma detection rate (ADR), a measure of the ability of a colonoscopist to identify and remove precancerous colorectal polyps, varies significantly among endoscopists. Here, we attempt to use a convolutional neural network (CNN) to generate a unique computer-aided diagnosis (CAD) system by exploring in detail the multiple-scale performance of deep neural networks. We applied this system to 3,375 hand-labeled images from the screening colonoscopies of 1,197 patients; of whom, 3,045 were assigned to the training dataset and 330 to the testing dataset. The images were diagnosed simply as either an adenomatous or non-adenomatous polyp. When applied to the testing dataset, our CNN-CAD system achieved a mean average precision of 89.5%. We conclude that the proposed framework could increase the ADR and decrease the incidence of interval CRCs, although further validation through large multicenter trials is required.

A conditional approach for the receiver operating characteristic curve construction to evaluate diagnostic test performance in a family-matched case–control design

2021 ◽  
pp. 096228022199595
Author(s):  
Yalda Zarnegarnia ◽  
Shari Messinger
Keyword(s):  
Receiver Operating Characteristic Curve ◽  
Receiver Operating Characteristic ◽  
Operating Characteristic ◽  
Characteristic Curve ◽  
Case Control ◽  
Paired Data ◽  
Characteristic Curves ◽  
Receiver Operating Characteristic Curves ◽  
Operating Characteristic Curve ◽  
Receiver Operating

Receiver operating characteristic curves are widely used in medical research to illustrate biomarker performance in binary classification, particularly with respect to disease or health status. Study designs that include related subjects, such as siblings, usually have common environmental or genetic factors giving rise to correlated biomarker data. The design could be used to improve detection of biomarkers informative of increased risk, allowing initiation of treatment to stop or slow disease progression. Available methods for receiver operating characteristic construction do not take advantage of correlation inherent in this design to improve biomarker performance. This paper will briefly review some developed methods for receiver operating characteristic curve estimation in settings with correlated data from case–control designs and will discuss the limitations of current methods for analyzing correlated familial paired data. An alternative approach using conditional receiver operating characteristic curves will be demonstrated. The proposed approach will use information about correlation among biomarker values, producing conditional receiver operating characteristic curves that evaluate the ability of a biomarker to discriminate between affected and unaffected subjects in a familial paired design.

scholarly journals Derivation of water flooding characteristic curve for offshore low-amplitude structural reservoir with strong bottom water

2021 ◽  
Author(s):  
Ying-xian Liu ◽  
Jie Tan ◽  
Hui Cai ◽  
Yan-lai Li ◽  
Chun-yan Liu
Keyword(s):  
Bottom Water ◽  
Characteristic Curve ◽  
Calculated Data ◽  
Oil Production ◽  
Water Flooding ◽  
Characteristic Curves ◽  
Recoverable Reserves ◽  
Curve Method ◽  
Water Cut ◽  
Low Amplitude

AbstractThe water flooding characteristic curve method is one of the essential techniques to predict recoverable reserves. However, the recoverable reserves indicated by the existing water flooding characteristic curves of low-amplitude reservoirs with strong bottom water increase gradually, and the current local recovery degree of some areas has exceeded the predicted recovery rate. The applicability of the existing water flooding characteristic curves in low-amplitude reservoirs with strong bottom water is lacking, which affects the accurate prediction of development performance. By analyzing the derivation process of the conventional water flooding characteristic curve method, this manuscript finds out the reasons for the poor applicability of the existing water flooding characteristic curve in low-amplitude reservoir with strong bottom water and corrects the existing water flooding characteristic curve according to the actual situation of the oilfield and obtains the improvement method of water flooding characteristic curve in low-amplitude reservoir with strong bottom water. After correction, the correlation coefficient between $$\frac{{k_{ro} }}{{k_{rw} }}$$ k ro k rw and $$S_{w}$$ S w is 95.92%. According to the comparison between the actual data and the calculated data, in 2021/3, the actual water cut is 97.29%, the water cut predicted by the formula is 97.27%, the actual cumulative oil production is 31.19 × 104t, and the predicted cumulative oil production is 31.31 × 104t. The predicted value is consistent with the actual value. It provides a more reliable method for predicting low-amplitude reservoirs' recoverable ability with strong bottom water and guides the oilfield's subsequent decision-making.

Fr612 INFLUENCE OF COMPUTER AIDED DIAGNOSIS WITH AUGMENTED VISUALIZATION ON ENDOSCOPIC CLASSIFICATION OF COLORECTAL POLYPS

2021 ◽  
Vol 160 (6) ◽  
pp. S-376
Author(s):  
Eladio Rodriguez-Diaz ◽  
Gyorgy Baffy Wai-Kit Lo ◽  
Hiroshi Mashimo ◽  
Aparna Repaka ◽  
Alexander Goldowsky ◽  
...  
Keyword(s):  
Computer Aided Diagnosis ◽  
Colorectal Polyps ◽  
Computer Aided ◽  
Endoscopic Classification ◽  
Aided Diagnosis

Convolutional Neural Networks in Computer-Aided Diagnosis of Colorectal Polyps and Cancer: A Review

2021 ◽  
Author(s):  
Kamyab Keshtkar
Keyword(s):  
Colorectal Cancer ◽  
Neural Networks ◽  
Deep Learning ◽  
Computer Aided Diagnosis ◽  
Colorectal Polyps ◽  
Support Vector ◽  
Training Process ◽  
Computer Aided ◽  
Conventional Models ◽  
Aided Diagnosis

As a relatively high percentage of adenoma polyps are missed, a computer-aided diagnosis (CAD) tool based on deep learning can aid the endoscopist in diagnosing colorectal polyps or colorectal cancer in order to decrease polyps missing rate and prevent colorectal cancer mortality. Convolutional Neural Network (CNN) is a deep learning method and has achieved better results in detecting and segmenting specific objects in images in the last decade than conventional models such as regression, support vector machines or artificial neural networks. In recent years, based on the studies in medical imaging criteria, CNN models have acquired promising results in detecting masses and lesions in various body organs, including colorectal polyps. In this review, the structure and architecture of CNN models and how colonoscopy images are processed as input and converted to the output are explained in detail. In most primary studies conducted in the colorectal polyp detection and classification field, the CNN model has been regarded as a black box since the calculations performed at different layers in the model training process have not been clarified precisely. Furthermore, I discuss the differences between the CNN and conventional models, inspect how to train the CNN model for diagnosing colorectal polyps or cancer, and evaluate model performance after the training process.

Can early neurological improvement after mechanical thrombectomy be used as a surrogate for final stroke outcome?

2018 ◽  
Vol 11 (5) ◽  
pp. 450-454 ◽  
Author(s):  
Sebastien Soize ◽  
Guillaume Fabre ◽  
Matthias Gawlitza ◽  
Isabelle Serre ◽  
Serge Bakchine ◽  
...  
Keyword(s):  
Sensitivity And Specificity ◽  
Mechanical Thrombectomy ◽  
Characteristic Curve ◽  
Optimal Threshold ◽  
Future Research ◽  
Nihss Score ◽  
Neurological Improvement ◽  
Good Outcome ◽  
Optimal Thresholds ◽  
Definition Of

Background and purposeWe aimed to identify the best definition of early neurological improvement (ENI) at 2 and 24 hours after mechanical thrombectomy (MT) and determine its ability to predict a good functional outcome at 3 months.MethodsThis retrospective analysis was based on a prospectively collected registry of patients treated by MT for ischemic stroke from May 2010 to March 2017. We included patients treated with stent-retrievers with National Institute of Health Stroke Scale (NIHSS) score before treatment and at 2 and/or 24 hours after treatment and modified Rankin Score (mRS) at 3 months. Receiver operating characteristic curve analysis was performed to estimate optimal thresholds for ENI at 2 and 24 hours. The relationship between optimal ENI definitions and good outcome at 3 months (mRS 0–2) was assessed by logistic regression.ResultsThe analysis included 246 patients. At 2 hours, the optimal threshold to predict a good outcome at 3 months was improvementin the NIHSS score of >1 point (AUC 0.83,95% CI 0.77 to 0.87), with sensitivity and specificity 78.3% (62.2–85.7%) and 84.6% (77.2–90.3%), respectively, and OR 12.67 (95% CI 4.69 to 31.10, p<0.0001). At 24 hours, the optimal threshold was an improvementin the NIHSS score of >4 points (AUC 0.93, 95% CI 0.89 to 0.96), with sensitivity and specificity 93.8% (87.7–97.5%) and 83.2% (75.7–89.2%), respectively, and OR 391.32 (95% CI 44.43 to 3448.35, p<0.0001).ConclusionsENI 24 hours after thrombectomy appears to be a straightforward surrogate of long-term endpoints and may have value in future research.

scholarly journals Centered Kernel Alignment Enhancing Neural Network Pretraining for MRI-Based Dementia Diagnosis

2016 ◽  
Vol 2016 ◽  
pp. 1-10 ◽  
Author(s):  
David Cárdenas-Peña ◽  
Diego Collazos-Huertas ◽  
German Castellanos-Dominguez
Keyword(s):  
Neural Network ◽  
Characteristic Curve ◽  
Principal Component ◽  
Imaging Data ◽  
Resonance Imaging ◽  
Dementia Diagnosis ◽  
Target Matrix ◽  
Computer Aided ◽  
Artificial Neural Network Ann ◽  
Aided Diagnosis

Dementia is a growing problem that affects elderly people worldwide. More accurate evaluation of dementia diagnosis can help during the medical examination. Several methods for computer-aided dementia diagnosis have been proposed using resonance imaging scans to discriminate between patients with Alzheimer’s disease (AD) or mild cognitive impairment (MCI) and healthy controls (NC). Nonetheless, the computer-aided diagnosis is especially challenging because of the heterogeneous and intermediate nature of MCI. We address the automated dementia diagnosis by introducing a novel supervised pretraining approach that takes advantage of the artificial neural network (ANN) for complex classification tasks. The proposal initializes an ANN based on linear projections to achieve more discriminating spaces. Such projections are estimated by maximizing the centered kernel alignment criterion that assesses the affinity between the resonance imaging data kernel matrix and the label target matrix. As a result, the performed linear embedding allows accounting for features that contribute the most to the MCI class discrimination. We compare the supervised pretraining approach to two unsupervised initialization methods (autoencoders and Principal Component Analysis) and against the best four performing classification methods of the 2014CADDementiachallenge. As a result, our proposal outperforms all the baselines (7% of classification accuracy and area under the receiver-operating-characteristic curve) at the time it reduces the class biasing.

Sign in / Sign up

Export Citation Format

Share Document