Distinguishing patients with celiac disease by quantitative analysis of videocapsule endoscopy images

Computer vision-based solutions to overcome the limitations of wireless capsule endoscopy

Endoscopic investigation plays a critical role in the diagnosis of gastrointestinal (GI) diseases. Since 2001, Wireless Capsule Endoscopy (WCE) has been available for small bowel exploration and is in continuous development. Over the last decade, WCE has achieved impressive improvements in areas such as miniaturisation, image quality and battery life. As a result, WCE is currently a very useful alternative to wired enteroscopy in the investigation of various small bowel abnormalities and has the potential to become the leading screening technique for the entire gastrointestinal tract. However, commercial solutions still have several limitations, namely incomplete examination and limited diagnostic capacity. These deficiencies are related to technical issues, such as image quality, motion estimation and power consumption management. Computational methods, based on image processing and analysis, can help to overcome these challenges and reduce both the time required by reviewers and human interpretation errors. Research groups have proposed a series of methods including algorithms for locating the capsule or lesion, assessing intestinal motility and improving image quality.In this work, we provide a critical review of computational vision-based methods for WCE image analysis aimed at overcoming the technological challenges of capsules. This article also reviews several representative public datasets used to evaluate the performance of WCE techniques and methods. Finally, some promising solutions of computational methods based on the analysis of multiple-camera endoscopic images are presented.

Artificial intelligence within the small bowel: are we lagging behind?

PURPOSE OF REVIEW

The use of artificial intelligence in small bowel capsule endoscopy is expanding. This review focusses on the use of artificial intelligence for small bowel pathology compared with human data and developments to date.

RECENT FINDINGS

The diagnosis and management of small bowel disease has been revolutionized with the advent of capsule endoscopy. Reading of capsule endoscopy videos however is time consuming with an average reading time of 40 min. Furthermore, the fatigued human eye may miss subtle lesions including indiscreet mucosal bulges. In recent years, artificial intelligence has made significant progress in the field of medicine including gastroenterology. Machine learning has enabled feature extraction and in combination with deep neural networks, image classification has now materialized for routine endoscopy for the clinician.

SUMMARY

Artificial intelligence is in built within the Navicam-Ankon capsule endoscopy reading system. This development will no doubt expand to other capsule endoscopy platforms and capsule endoscopies that are used to visualize other parts of the gastrointestinal tract as a standard. This wireless and patient friendly technique combined with rapid reading platforms with the help of artificial intelligence will become an attractive and viable choice to alter how patients are investigated in the future.

Understanding the role of the gut in undernutrition: what can technology tell us?

Gut function remains largely underinvestigated in undernutrition, despite its critical role in essential nutrient digestion, absorption and assimilation. In areas of high enteropathogen burden, alterations in gut barrier function and subsequent inflammatory effects are observable but remain poorly characterised. Environmental enteropathy (EE)-a condition that affects both gut morphology and function and is characterised by blunted villi, inflammation and increased permeability-is thought to play a role in impaired linear growth (stunting) and severe acute malnutrition. However, the lack of tools to quantitatively characterise gut functional capacity has hampered both our understanding of gut pathogenesis in undernutrition and evaluation of gut-targeted therapies to accelerate nutritional recovery. Here we survey the technology landscape for potential solutions to improve assessment of gut function, focussing on devices that could be deployed at point-of-care in low-income and middle-income countries (LMICs). We assess the potential for technological innovation to assess gut morphology, function, barrier integrity and immune response in undernutrition, and highlight the approaches that are currently most suitable for deployment and development. This article focuses on EE and undernutrition in LMICs, but many of these technologies may also become useful in monitoring of other gut pathologies.

WCE polyp detection with triplet based embeddings

Wireless capsule endoscopy is a medical procedure used to visualize the entire gastrointestinal tract and to diagnose intestinal conditions, such as polyps or bleeding. Current analyses are performed by manually inspecting nearly each one of the frames of the video, a tedious and error-prone task. Automatic image analysis methods can be used to reduce the time needed for physicians to evaluate a capsule endoscopy video. However these methods are still in a research phase. In this paper we focus on computer-aided polyp detection in capsule endoscopy images. This is a challenging problem because of the diversity of polyp appearance, the imbalanced dataset structure and the scarcity of data. We have developed a new polyp computer-aided decision system that combines a deep convolutional neural network and metric learning. The key point of the method is the use of the Triplet Loss function with the aim of improving feature extraction from the images when having small dataset. The Triplet Loss function allows to train robust detectors by forcing images from the same category to be represented by similar embedding vectors while ensuring that images from different categories are represented by dissimilar vectors. Empirical results show a meaningful increase of AUC values compared to state-of-the-art methods. A good performance is not the only requirement when considering the adoption of this technology to clinical practice. Trust and explainability of decisions are as important as performance. With this purpose, we also provide a method to generate visual explanations of the outcome of our polyp detector. These explanations can be used to build a physician's trust in the system and also to convey information about the inner working of the method to the designer for debugging purposes.

Celiac disease diagnosis from videocapsule endoscopy images with residual learning and deep feature extraction

BACKGROUND AND OBJECTIVE

Videocapsule endoscopy (VCE) is a relatively new technique for evaluating the presence of villous atrophy in celiac disease patients. The diagnostic analysis of video frames is currently time-consuming and tedious. Recently, computer-aided diagnosis (CAD) systems have become an attractive research area for diagnosing celiac disease. However, the images captured from VCE are susceptible to alterations in light illumination, rotation direction, and intestinal secretions. Moreover, textural features of the mucosal villi obtained by VCE are difficult to characterize and extract. This work aims to find a novel deep learning feature learning module to assist in the diagnosis of celiac disease.

METHODS

In this manuscript, we propose a novel deep learning recalibration module which shows significant gain in diagnosing celiac disease. In this recalibration module, the block-wise recalibration component is newly employed to capture the most salient feature in the local channel feature map. This learning module was embedded into ResNet50, Inception-v3 to diagnose celiac disease using a 10-time 10-fold cross-validation based upon analysis of VCE images. In addition, we employed model weights to extract feature points from training and test samples before the last fully connected layer, and then input to a support vector machine (SVM), k-nearest neighbor (KNN), and linear discriminant analysis (LDA) for differentiating celiac disease images from heathy controls.

RESULTS

Overall, the accuracy, sensitivity and specificity of the 10-time 10-fold cross-validation were 95.94%, 97.20% and 95.63%, respectively.

CONCLUSIONS

A novel deep learning recalibration module, with global response and local salient factors is proposed, and it has a high potential for utilizing deep learning networks to diagnose celiac disease using VCE images.

Current Evidence on Computer-Aided Diagnosis of Celiac Disease: Systematic Review

Celiac disease (CD) is a chronic autoimmune disease that occurs in genetically predisposed individuals in whom the ingestion of gluten leads to damage of the small bowel. It is estimated to affect 1 in 100 people worldwide, but is severely underdiagnosed. Currently available guidelines require CD-specific serology and atrophic histology in duodenal biopsy samples for the diagnosis of adult CD. In pediatric CD, but in recent years in adults also, nonbioptic diagnostic strategies have become increasingly popular. In this setting, in order to increase the diagnostic rate of this pathology, endoscopy itself has been thought of as a case finding strategy by use of digital image processing techniques. Research focused on computer aided decision support used as database video capsule, endoscopy and even biopsy duodenal images. Early automated methods for diagnosis of celiac disease used feature extraction methods like spatial domain features, transform domain features, scale-invariant features and spatio-temporal features. Recent artificial intelligence (AI) techniques using deep learning (DL) methods such as convolutional neural network (CNN), support vector machines (SVM) or Bayesian inference have emerged as a breakthrough computer technology which can be used for computer aided diagnosis of celiac disease. In the current review we summarize methods used in clinical studies for classification of CD from feature extraction methods to AI techniques.

Automated diagnosis of celiac disease by video capsule endoscopy using DAISY Descriptors

Celiac disease is a genetically determined disorder of the small intestine, occurring due to an immune response to ingested gluten-containing food. The resulting damage to the small intestinal mucosa hampers nutrient absorption, and is characterized by diarrhea, abdominal pain, and a variety of extra-intestinal manifestations. Invasive and costly methods such as endoscopic biopsy are currently used to diagnose celiac disease. Detection of the disease by histopathologic analysis of biopsies can be challenging due to suboptimal sampling. Video capsule images were obtained from celiac patients and controls for comparison and classification. This study exploits the use of DAISY descriptors to project two-dimensional images onto one-dimensional vectors. Shannon entropy is then used to extract features, after which a particle swarm optimization algorithm coupled with normalization is employed to select the 30 best features for classification. Statistical measures of this paradigm were tabulated. The accuracy, positive predictive value, sensitivity and specificity obtained in distinguishing celiac versus control video capsule images were 89.82%, 89.17%, 94.35% and 83.20% respectively, using the 10-fold cross-validation technique. When employing manual methods rather than the automated means described in this study, technical limitations and inconclusive results may hamper diagnosis. Our findings suggest that the computer-aided detection system presented herein can render diagnostic information, and thus may provide clinicians with an important tool to validate a diagnosis of celiac disease.

Quantitative histology-based classification system for assessment of the intestinal mucosal histological changes in patients with celiac disease

Background/Aims

The existing histological classifications for the interpretation of small intestinal biopsies are based on qualitative parameters with high intraobserver and interobserver variations. We have developed and propose a quantitative histological classification system for the assessment of intestinal mucosal biopsies.

Methods

We performed a computer-assisted quantitative histological assessment of digital images of duodenal biopsies from 137 controls and 124 patients with celiac disease (CeD) (derivation cohort). From the receiver-operating curve analysis, followed by multivariate and logistic regression analyses, we identified parameters for differentiating control biopsies from those of the patients with CeD. We repeated the quantitative histological analysis in a validation cohort (105 controls and 120 patients with CeD). On the basis of the results, we propose a quantitative histological classification system. The new classification was compared with the existing histological classifications for interobserver and intraobserver agreements by a group of qualified pathologists.

Results

Among the histological parameters, intraepithelial lymphocyte count of ≥25/100 epithelial cells, adjusted villous height fold change of ≤0.7, and crypt depth-to-villous height ratio of ≥0.5 showed good discriminative power between the mucosal biopsies from the patients with CeD and those from the controls, with 90.3% sensitivity, 93.5% specificity, and 96.2% area under the curve. Among the existing histological classifications, our quantitative histological classification showed the highest intraobserver (69.7%–85.03%) and interobserver (24.6%–71.5%) agreements.

Conclusions

Quantitative assessment increases the reliability of the histological assessment of mucosal biopsies in patients with CeD. Such a classification system may be used for clinical trials in patients with CeD.

Celiac disease gene expression data can be used to classify biopsies along the Marsh score severity scale

BACKGROUND AND AIM

The diagnosis of celiac disease autoimmune pathology relies on the subjective histological assignment of biopsies into Marsh score categories. It is hypothesized that Marsh score categories have unique gene expression signatures. The aims were as follows: first, to develop a celiac disease quantitative reverse transcription-polymerase chain reaction (RT-PCR) array; second, define gene expression signatures associated with Marsh score categories; and third, develop equations that classify biopsies into Marsh score categories and to monitor the efficacy of patient treatment.

METHODS

Gene targets for inclusion in the celiac RT-PCR (qRT-PCR) array were identified using systematic analysis of published celiac transcriptomic data. The array was used to assess the gene expression associated with histological changes in duodenal biopsies obtained from adult patients. Finally, Marsh score classification equations were defined using discriminant analysis.

RESULTS

The array contained 87 genes. The expression of 26 genes were significantly (p < 0.06) associated with the discrete Marsh score categories. As the Marsh score pathology of biopsies increased, there was a progression of innate immune gene expression through adaptive Th1-specific gene expression with a concurrent decrease in intestinal structural gene expression in high Marsh score samples. These 26 genes were used to define classification equations that accounted for 99% of the observed experimental variation and which could classify biopsies into Marsh score categories and monitor patient treatment progression.

CONCLUSIONS

This proof-of-concept study successfully developed a celiac RT-PCR array and has provided evidence that discriminant equations defined using gene expression data can objectively and accurately classify duodenal biopsies into Marsh score categories.

Equations defined using gene expression and histological data resolve coeliac disease biopsies within the Marsh score continuum

BACKGROUND/AIM

The gold standard diagnostic for coeliac disease (CD) is subjective histological assignment of biopsies into the Marsh score categories. It is hypothesized that discrete Marsh score categories can be quantitatively resolved into a continuum using discriminant equations defined using histological and gene expression data. Therefore, the aim of this study was to use a combination of histological and gene expression data to develop equations that classify CD patient biopsies into a quantitative Marsh score continuum which could be used by clinicians to monitor CD treatment efficacy.

METHODS

Both empirical and simulated gene expression and histological data were used to define predictive Marsh score equations. The distances of treated sample biopsies from the Marsh score standards were determined using the Mahalanobis distance calculation.

RESULTS

Three function, high resolution discriminant equations derived from simulated data were used to accurately classify 99.6% of simulated and empirically derived biopsy data. The first function resolved active (Marsh type 3) CD from mild (Marsh type 1) CD. The second function resolved normal (no specific pathology) biopsies from mild CD. The third function resolved active Marsh score 3 into a and b subcategories. Finally, measuring the Mahalanobis distance enabled the conversion of discrete Marsh score categories into a continuum.

CONCLUSIONS

This proof-of-concept study successfully demonstrated that the discrete Marsh score scale can be converted into a quantitative continuum capable of high resolution monitoring of patient treatment efficacy using equations defined by gene expression and histology data.

A comprehensive review on the utility of capsule endoscopy in coeliac disease: From computational analysis to the bedside

Small bowel capsule endoscopy (SBCE) can identify macroscopic changes of coeliac disease and assess the extent of disease in the small bowel beyond the duodenum. SBCE has a good sensitivity for the detection of coeliac disease in comparison to histology owing to several ideal features such as a high magnification. It also plays a useful role in detecting complications in patients with refractory coeliac disease. Several studies have been carried out on transforming images obtained from small bowel capsule endoscopy to enable the automated detection of features related to coeliac disease. This review discusses the current roles played by small bowel capsule endoscopy in coeliac disease. It identifies future potential roles of this technique and describes in great detail the role of computational analysis in the detection of coeliac disease and how it can be adapted to current available technology.

Capsule endoscopy for patients with coeliac disease

Coeliac disease is an autoimmune mediated condition in response to gluten. A combination of innate and adaptive immune responses results in villous shortening in the small bowel (SB) that can be morphologically picked up on capsule endoscopy. It is the only imaging modality that can provide mucosal views of the entire SB, while histology is generally limited to the proximal SB. Radiological modalities are not designed to pick up changes in villous morphology. Areas covered: In this review, we provide a comprehensive analysis on the justified use of small bowel capsule endoscopy (SBCE) in the assessment of patients with coeliac disease; compare SBCE to histology, serology, and symptomatology; and provide an overview on automated quantitative analysis for the detection of coeliac disease. We also provide insight into future work on SBCE in relation to coeliac disease. Expert commentary: SBCE has opened up new avenues for the diagnosis and monitoring of patients with coeliac disease. However, larger studies with new and established coeliac disease patients and with greater emphasis on morphological features on SBCE are required to better define the role of SBCE in the setting of coeliac disease.

Automated classification of celiac disease during upper endoscopy: Status quo and quo vadis

A large amount of digital image material is routinely captured during esophagogastroduodenoscopies but, for the most part, is not used for confirming the diagnosis process of celiac disease which is primarily based on histological examination of biopsies. Recently, considerable effort has been undertaken to make use of image material by developing semi- or fully-automated systems to improve the diagnostic workup. Recently, focus was especially laid on developing state-of-the-art deep learning architectures, exploiting the endoscopist's expert knowledge and on making systems fully automated and thereby completely observer independent. In this work, we summarize recent trends in the field of computer-aided celiac disease diagnosis based on upper endoscopy and discuss about recent progress, remaining challenges, limitations currently prohibiting a deployment in clinical practice and future efforts to tackle them.

Coeliac disease and the videocapsule: what have we learned till now

Celiac disease is diagnosed in part by finding areas of pathology in the small bowel (SB) mucosa. This can often be difficult because the pathologic alterations, including atrophy of the small intestinal villi, can often be sparse and subtle. Some of the quantitative methods for detecting and measuring the presence of villous atrophy from videocapsule endoscopy (VCE) images are presented and discussed. These methods consist of static features of measurement including texture, gray level, and presence and abundance of fissures contained within each acquired image. The methods also consist of dynamic measurements including spectral analysis, and determining motion from a sequence of endoscopic images as obtained from a VCE clip. Thus far, several methods have been found useful to characterize the SB mucosa of untreated celiac disease patients versus control patients lacking villous atrophy, which have revealed significant differences in texture, frequency, and motion on analysis of VCE. In untreated celiac patients undergoing endoscopy, there tends to be greater magnitude of changes and spatial differences in textural descriptors, longer periodic components, indicating slower periodic activity, and differences in feature location, suggesting alterations in motility at areas of pathology as compared to patients without villous atrophy. Improvements in the quantitative analysis of VCE imaging in celiac patients is important to detect pathology in suspected patients, so that biopsies can be obtained from pertinent regions of the small intestinal mucosa. Improvements are also necessary so that patients with celiac disease can be monitored to evaluate the progress of mucosal healing after onset of treatment.

Recommendations to quantify villous atrophy in video capsule endoscopy images of celiac disease patients

AIM

To quantify the presence of villous atrophy in endoscopic images for improved automation.

METHODS

There are two main categories of quantitative descriptors helpful to detect villous atrophy: (1) Statistical and (2) Syntactic. Statistical descriptors measure the small intestinal substrate in endoscope-acquired images based on mathematical methods. Texture is the most commonly used statistical descriptor to quantify villous atrophy. Syntactic descriptors comprise a syntax, or set of rules, for analyzing and parsing the substrate into a set of objects with boundaries. The syntax is designed to identify and distinguish three-dimensional structures based on their shape.

RESULTS

The variance texture statistical descriptor is useful to describe the average variability in image gray level representing villous atrophy, but does not determine the range in variability and the spatial relationships between regions. Improved textural descriptors will incorporate these factors, so that areas with variability gradients and regions that are orientation dependent can be distinguished. The protrusion syntactic descriptor is useful to detect three-dimensional architectural components, but is limited to identifying objects of a certain shape. Improvement in this descriptor will require incorporating flexibility to the prototypical template, so that protrusions of any shape can be detected, measured, and distinguished.

CONCLUSION

Improved quantitative descriptors of villous atrophy are being developed, which will be useful in detecting subtle, varying patterns of villous atrophy in the small intestinal mucosa of suspected and known celiac disease patients.

Extraction and processing of videocapsule data to detect and measure the presence of villous atrophy in celiac disease patients

BACKGROUND

Videocapsule endoscopy is a relative new method to analyze the gastrointestinal tract for the presence of pathologic features. It is of relevance to detect villous atrophy in the small bowel, which is a defining symptom of celiac disease.

METHOD

In this tutorial, methods to extract and process videocapsule endoscopy data are elucidated. The algorithms, computer code, and paradigms to analyze image series are described in detail. The topics covered include extraction of data, analysis of texture, eigenanalysis, spectral analysis, three-dimensional projection, and estimation of motility. The basic paradigms to implement these processes are provided.

RESULTS

Examples of successful quantitative analysis implementations for selected untreated celiac disease patients with villous atrophy versus control patients with normal villi were illustrated. Based on the implementations, it was evident that celiac patients tended to have a rougher small intestinal texture as compared with control patients. From three-dimensional projection, celiac patients exhibited larger surface protrusions emanating from the small intestinal mucosa, which may represent clumps of atrophied villi. The periodicity of small intestinal contractions tends to be slower when villous atrophy is present, and the estimated degree of motility is reduced as compared with control image series. Basis image construction suggested that fissuring and mottling of the mucosal surface is predominant in untreated celiac patients, and mostly absent in controls.

CONCLUSIONS

Implementation of computerized methods, as described in this tutorial, will likely be useful for the automated detection and measurement of villous atrophy, and to map its extent along the small intestine of celiac patients.

Making texture descriptors invariant to blur

Besides a high distinctiveness, robustness (or invariance) to image degradations is very desirable for texture feature extraction methods in real-world applications. In this paper, focus is on making arbitrary texture descriptors invariant to blur which is often prevalent in real image data. From previous work, we know that most state-of-the-art texture feature extraction methods are unable to cope even with minor blur degradations if the classifier's training stage is based on idealistic data. However, if the training set suffers similarly from the degradations, the obtained accuracies are significantly higher. Exploiting that knowledge, in this approach the level of blur of each image is increased to a certain threshold, based on the estimation of a blur measure. Experiments with synthetically degraded data show that the method is able to generate a high degree of blur invariance without loosing too much distinctiveness. Finally, we show that our method is not limited to ideal Gaussian blur.

Quantitative measurements in capsule endoscopy

This review summarizes several approaches for quantitative measurement in capsule endoscopy. Video capsule endoscopy (VCE) typically provides wireless imaging of small bowel. Currently, a variety of quantitative measurements are implemented in commercially available hardware/software. The majority is proprietary and hence undisclosed algorithms. Measurement of amount of luminal contamination allows calculating scores from whole VCE studies. Other scores express the severity of small bowel lesions in Crohn׳s disease or the degree of villous atrophy in celiac disease. Image processing with numerous algorithms of textural and color feature extraction is further in the research focuses for automated image analysis. These tools aim to select single images with relevant lesions as blood, ulcers, polyps and tumors or to omit images showing only luminal contamination. Analysis of motility pattern, size measurement and determination of capsule localization are additional topics. Non-visual wireless capsules transmitting data acquired with specific sensors from the gastrointestinal (GI) tract are available for clinical routine. This includes pH measurement in the esophagus for the diagnosis of acid gastro-esophageal reflux. A wireless motility capsule provides GI motility analysis on the basis of pH, pressure, and temperature measurement. Electromagnetically tracking of another motility capsule allows visualization of motility. However, measurement of substances by GI capsules is of great interest but still at an early stage of development.

Pondering the potential of quantitative analysis of video capsule endoscopy images in the management of children with celiac disease

Celiac disease is common throughout the world, affecting approximately 1% of patients of all age groups. In this review, the role of video capsule endoscopy in characterizing the small intestinal lumen of pediatric patients with celiac disease is discussed in detail. Quantitative aspects of video capsule endoscopy which may be helpful in diagnosing pediatric patients are highlighted.

Videocapsule endoscopy in celiac disease: indications and timing

BACKGROUND

Because of its technical characteristics (i.e. 8-fold magnification, capability to inspect the entire small bowel) and minimal invasiveness, videocapsule endoscopy (VCE) has been proposed as a useful tool for managing patients with celiac disease (CD).

KEY MESSAGES

Although VCE has been found to be highly sensitive and specific in identifying CD endoscopic markers, it is still inadequate to replace esophagogastroduodenoscopy (EGD) with biopsies in the diagnosis of CD. Nevertheless, it represents a reliable alternative in patients unable or unwilling to undergo EGD. Up to now, available studies have failed to identify any correlation between the length of small bowel involvement and the severity of symptoms. The available evidence on the use of VCE in diagnosing CD in equivocal cases (patients with positive serology and negative or nonspecific histology or those with negative serology and histologically proven villous atrophy) is limited, and its role is still under discussion. In CD patients not improving on gluten-free diet, a complete workup is necessary. In patients with nonresponsive (NRCD) or refractory CD (RCD), VCE has been shown to be able not only to detect significant findings, driving further management, but also to rule out major complications. Nevertheless, in this setting, the inability of VCE to take tissue samples and the risk of capsule retention can represent major limitations.

CONCLUSIONS

At the present time, for diagnostic purposes, VCE can be proposed only in patients unable or unwilling to undergo EGD, whereas it could be useful in some equivocal cases. Conversely, there is no room for VCE either to estimate the length of the small bowel affected by villous atrophy or to follow up patients improving on gluten-free diet. In patients with NRCD or RCD, VCE can play a role, but it should be combined with other diagnostic techniques.

Motility alterations in celiac disease and non-celiac gluten sensitivity

Regulation of gut motility is complex and involves neuromuscular, immune and environmental mechanisms. It is well established that patients with celiac disease (CD) often display gut dysmotility. Studies have shown the presence of disturbed esophageal motility, altered gastric emptying, and dysmotility of the small intestine, gallbladder and colon in untreated CD. Most of these motor abnormalities resolve after a strict gluten-free diet, suggesting that mechanisms related to the inflammatory condition and disease process are responsible for the motor dysfunction. Motility abnormalities are also a hallmark of functional bowel disorders such as irritable bowel syndrome (IBS), where it has been proposed as underlying mechanism for symptom generation (diarrhea, constipation, bloating). Non-celiac gluten sensitivity (NCGS) is a poorly defined entity, mostly self-diagnosed, that presents clinically with IBS symptoms in the absence of specific celiac markers. Patients with NCGS are believed to react symptomatically to wheat components, and some studies have proposed the presence of low-grade inflammation in these patients. There is little information regarding the functional characterization of these patients before and after a gluten-free diet. A study suggested the presence of altered gastrointestinal transit in NCGS patients who also have a high prevalence of nonspecific anti-gliadin antibodies. Results of an ongoing clinical study in NCGS patients with positive anti-gliadin antibodies before and after a gluten-free diet will be discussed. Elucidating the mechanisms for symptom generation in NCGS patients is important to find new therapeutic alternatives to the burden of imposing a strict gluten-free diet in patients who do not have CD.

Quantitative image analysis of celiac disease

We outline the use of quantitative techniques that are currently used for analysis of celiac disease. Image processing techniques can be useful to statistically analyze the pixular data of endoscopic images that is acquired with standard or videocapsule endoscopy. It is shown how current techniques have evolved to become more useful for gastroenterologists who seek to understand celiac disease and to screen for it in suspected patients. New directions for focus in the development of methodology for diagnosis and treatment of this disease are suggested. It is evident that there are yet broad areas where there is potential to expand the use of quantitative techniques for improved analysis in suspected or known celiac disease patients.

Survey on computer aided decision support for diagnosis of celiac disease

Celiac disease (CD) is a complex autoimmune disorder in genetically predisposed individuals of all age groups triggered by the ingestion of food containing gluten. A reliable diagnosis is of high interest in view of embarking on a strict gluten-free diet, which is the CD treatment modality of first choice. The gold standard for diagnosis of CD is currently based on a histological confirmation of serology, using biopsies performed during upper endoscopy. Computer aided decision support is an emerging option in medicine and endoscopy in particular. Such systems could potentially save costs and manpower while simultaneously increasing the safety of the procedure. Research focused on computer-assisted systems in the context of automated diagnosis of CD has started in 2008. Since then, over 40 publications on the topic have appeared. In this context, data from classical flexible endoscopy as well as wireless capsule endoscopy (WCE) and confocal laser endomicrosopy (CLE) has been used. In this survey paper, we try to give a comprehensive overview of the research focused on computer-assisted diagnosis of CD.

•

The state-of-the-art research in automated diagnosis of celiac disease is presented.

•

A systematic review of methods and techniques used in this field is given.

•

Specific issues and challenges in the field are identified and discussed.

Narrow band imaging combined with water immersion technique in the diagnosis of celiac disease

BACKGROUND

The "multiple-biopsy" approach both in duodenum and bulb is the best strategy to confirm the diagnosis of celiac disease; however, this increases the invasiveness of the procedure itself and is time-consuming.

AIM

To evaluate the diagnostic yield of a single biopsy guided by narrow-band imaging combined with water immersion technique in paediatric patients.

METHODS

Prospective assessment of the diagnostic accuracy of narrow-band imaging/water immersion technique-driven biopsy approach versus standard protocol in suspected celiac disease.

RESULTS

The experimental approach correctly diagnosed 35/40 children with celiac disease, with an overall diagnostic sensitivity of 87.5% (95% CI: 77.3-97.7). An altered pattern of narrow-band imaging/water immersion technique endoscopic visualization was significantly associated with villous atrophy at guided biopsy (Spearman Rho 0.637, p<0.001). Concordance of narrow-band imaging/water immersion technique endoscopic assessments was high between two operators (K: 0.884). The experimental protocol was highly timesaving compared to the standard protocol.

CONCLUSIONS

An altered narrow-band imaging/water immersion technique pattern coupled with high anti-transglutaminase antibodies could allow a single guided biopsy to diagnose celiac disease. When no altered mucosal pattern is visible even by narrow-band imaging/water immersion technique, multiple bulbar and duodenal biopsies should be obtained.

Identification of intestinal wall abnormalities and ischemia by modeling spatial uncertainty in computed tomography imaging findings

Intestinal abnormalities and ischemia are medical conditions in which inflammation and injury of the intestine are caused by inadequate blood supply. Acute ischemia of the small bowel can be life-threatening. Computed tomography (CT) is currently a gold standard for the diagnosis of acute intestinal ischemia in the emergency department. However, the assessment of the diagnostic performance of CT findings in the detection of intestinal abnormalities and ischemia has been a difficult task for both radiologists and surgeons. Little effort has been found in developing computerized systems for the automated identification of these types of complex gastrointestinal disorders. In this paper, a geostatistical mapping of spatial uncertainty in CT scans is introduced for medical image feature extraction, which can be effectively applied for diagnostic detection of intestinal abnormalities and ischemia from control patterns. Experimental results obtained from the analysis of clinical data suggest the usefulness of the proposed uncertainty mapping model.

Capsule endoscopy: Present status and future expectation

Video capsule endoscopy (CE) since its introduction 13 years back, has revolutionized our approach to small intestinal diseases. Obscure gastrointestinal bleed (OGIB) continues to be the most important indication for CE with a high sensitivity, specificity as well as positive and negative predictive values. It is best performed during ongoing bleed or immediately thereafter. Overt OGIB has a higher diagnostic yield than occult OGIB. However, even in iron deficiency anemia, CE is emerging as important investigation after initial negative work up. In suspected Crohn’s disease (CD), CE has been shown superior to traditional imaging and endoscopic technique and should be considered after a negative ileocolonoscopy. Although CE has also been used for evaluating established CD, a high capsule retention rate precludes its use ahead of cross-sectional imaging. Celiac disease, particularly where gastro-duodenoscopy cannot be performed or is normal, can also be investigated by CE. Small bowel tumor, hereditary polyposis syndrome, and non-steroidal anti-inflammatory drugs induced intestinal damage are other indications for CE. Capsule retention is the only significant adverse outcome of CE and occurs mostly in presence of intestinal obstruction. This can be prevented by use of Patency capsule prior to CE examination. Presence of cardiac pacemaker and intracardiac devices continue to be relative contraindications for CE, though data do not suggest interference of CE with these devices. Major limitations of CE today include failure to control its movement from outside, inability of CE to acquire tissue for diagnosis, and lack of therapeutic help. With ongoing interesting and exciting developments taking place in these areas, these issues would be solved in all probability in near future. CE has the potential to become one of the most important tools in diagnostic and possibly in the therapeutic field of gastrointestinal disorder.

Use of shape-from-shading to estimate three-dimensional architecture in the small intestinal lumen of celiac and control patients

BACKGROUND

As measured from videocapsule endoscopy images, the small intestinal mucosa of untreated celiac patients has significantly greater and more varied texture compared to normal patients. Three-dimensional modeling using shape-from-shading principles may further increase classification accuracy.

METHODS

A sequence of 200 consecutive videocapsule images acquired at a 2s(-1) frame rate and 576×576 pixel dimension, were obtained at four locations in the small intestinal lumen of ten patients with biopsy-proven celiac disease and ten control patients. Each two-dimensional image was converted to a three-dimensional architectural approximation by considering the 256 grayscale level to be linearly representative of image depth. From the resulting three-dimensional architecture, distinct luminal protrusions, representative of the macro-architecture, were automatically identified by computer algorithm. The range and number of protrusions per image, and their width and height, were determined for celiacs versus controls and tabulated as mean±SD.

RESULTS

The mean number of villous protrusions per image was 402.2±15.0 in celiacs versus 420.8±24.0 in controls (p<0.001). The average protrusion width was 14.7 pixels in celiacs versus 13.9 pixels in controls (p=0.01). The mean protrusion height was 3.10±2.34 grayscale levels for celiacs versus 2.70±0.43 grayscale levels for controls (p<0.001). Thus celiac patients had significantly fewer protrusions on the luminal surface of the small intestine as compared with controls, and these protrusions had greater dimensions, suggesting they are indicative of a mosaic (cobblestone) macro-architectural pattern which is common in celiacs.

CONCLUSIONS

Shape-from-shading modeling is useful to explore luminal macro-architecture and to detect significant differences in luminal morphology in celiac versus normal patients, which can increase the usefulness of videocapsule studies.

Computational method for high resolution spectral analysis of fractionated atrial electrograms

BACKGROUND

The discrete Fourier transform (DFT) is often used as a spectral estimator for analysis of complex fractionated atrial electrograms (CFAE) acquired during atrial fibrillation (AF). However, time resolution can be unsatisfactory, as the frequency resolution is proportional to rate/time interval. In this study we compared the DFT to a new spectral estimator with improved time-frequency resolution.

METHOD

Recently, a novel spectral estimator (NSE) based upon signal averaging was derived and implemented computationally. The NSE is similar to the DFT in that both estimators model the autocorrelation function to form the power spectrum. However, as derived in this study, NSE frequency resolution is proportional to rate/period(2) and thus unlike the DFT, is not directly dependent on the window length. We hypothesized that the NSE would provide improved time resolution while maintaining satisfactory frequency resolution for computation of CFAE spectral parameters. Window lengths of 8s, 4s, 2s, 1s, and 0.5s were used for analysis. Two criteria gauged estimator performance. Firstly, a periodic electrogram pattern with phase jitter was embedded in interference. The error in detecting the frequency of the periodic pattern was determined. Secondly, significant differences in spectral parameters for paroxysmal versus persistent AF data, which have known dissimilarities, were determined using the DFT versus NSE methods. The parameters measured were the dominant amplitude, dominant frequency, and mean spectral profile.

RESULTS

At all time resolutions, the error in detecting the frequency of the repeating electrogram pattern was less for NSE than for DFT (p<0.001). The DFT was accurate to 2s time resolution/0.5 Hz frequency resolution, while the NSE was accurate to 0.5s time resolution/0.05 Hz frequency resolution. At all time resolutions, significant differences in the dominant amplitude spectral parameter for paroxysmal versus persistent CFAE were greater using NSE than DFT (p<0.0001). For three of five time resolutions, the NSE had greater significant differences than DFT for discriminating the dominant frequency and mean spectral profile parameters between AF types.

CONCLUSIONS

The results suggest that the NSE has improved performance versus DFT for measurement of CFAE spectral properties.

Clinical applications of small bowel capsule endoscopy

Video capsule endoscopy has revolutionized our ability to visualize the entire small bowel mucosa. This modality is established as a valuable tool for the diagnosis of obscure gastrointestinal bleeding, Crohn's disease, small bowel tumors, and other conditions involving the small bowel mucosa. This review includes an overview of the current and potential future clinical applications of small bowel video endoscopy.

Clinical applications of small bowel capsule endoscopy

Video capsule endoscopy has revolutionized our ability to visualize the entire small bowel mucosa. This modality is established as a valuable tool for the diagnosis of obscure gastrointestinal bleeding, Crohn’s disease, small bowel tumors, and other conditions involving the small bowel mucosa. This review includes an overview of the current and potential future clinical applications of small bowel video endoscopy.

Scale invariant texture descriptors for classifying celiac disease

Scale invariant texture recognition methods are applied for the computer assisted diagnosis of celiac disease. In particular, emphasis is given to techniques enhancing the scale invariance of multi-scale and multi-orientation wavelet transforms and methods based on fractal analysis. After fine-tuning to specific properties of our celiac disease imagery database, which consists of endoscopic images of the duodenum, some scale invariant (and often even viewpoint invariant) methods provide classification results improving the current state of the art. However, not each of the investigated scale invariant methods is applicable successfully to our dataset. Therefore, the scale invariance of the employed approaches is explicitly assessed and it is found that many of the analyzed methods are not as scale invariant as they theoretically should be. Results imply that scale invariance is not a key-feature required for successful classification of our celiac disease dataset.

Categorization and segmentation of intestinal content frames for wireless capsule endoscopy

Wireless capsule endoscopy (WCE) is a device that allows the direct visualization of gastrointestinal tract with minimal discomfort for the patient, but at the price of a large amount of time for screening. In order to reduce this time, several works have proposed to automatically remove all the frames showing intestinal content. These methods label frames as {intestinal content- clear} without discriminating between types of content (with different physiological meaning) or the portion of image covered. In addition, since the presence of intestinal content has been identified as an indicator of intestinal motility, its accurate quantification can show a potential clinical relevance. In this paper, we present a method for the robust detection and segmentation of intestinal content in WCE images, together with its further discrimination between turbid liquid and bubbles. Our proposal is based on a twofold system. First, frames presenting intestinal content are detected by a support vector machine classifier using color and textural information. Second, intestinal content frames are segmented into {turbid, bubbles, and clear} regions. We show a detailed validation using a large dataset. Our system outperforms previous methods and, for the first time, discriminates between turbid from bubbles media.

Quantitative estimates of motility from videocapsule endoscopy are useful to discern celiac patients from controls

BACKGROUND

Prior work has shown that videocapsule endoscopy image features are a useful tool for quantitatively distinguishing the intestinal mucosal surface of untreated celiac patients from that of controls. The use of dynamic estimates of wall motility may further help to improve classification.

METHODS

Videocapsule endoscopy clips (200 frames each, 2 frames/s, 576 × 576 pixels/frame) were acquired at five small intestinal locations in 11 untreated celiac patients (celiacs) and ten controls. Color images were converted to grayscale and analyzed frame-by-frame. Variations in the position and width of the center of the small intestinal lumen were quantitatively estimated. The darkest grayscale pixels were used as an estimate of the lumen center. Over 200 frames, the standard deviation of the lumen center xy position and the mean and standard deviation in lumen center width were used as dynamic estimates of wall motility. These parameters were plotted in three-dimensional space, and the best discriminant function was used to classify celiacs versus controls at each of the following five locations: (1) duodenal bulb, (2) distal duodenum, (3) jejunum, (4) ileum, and (5) distal ileum.

RESULTS

The overall sensitivity for the classification of celiacs versus controls at all five locations was 98.2 %, while the specificity was 96.0 %. From location 1 to 5, there was a tendency for the lumen center width to diminish in terms of frame-to-frame variability by 7.6 % in celiacs (r (2) = 0.4) and 9.7 % in controls (r (2) = 0.7).

CONCLUSIONS

In addition to examining the mucosal surface, videocapsule endoscopy can assess small bowel intestinal motility and aid in distinguishing celiac patients from controls.

Quantitative estimates of motility from videocapsule endoscopy are useful to discern celiac patients from controls

BACKGROUND

Prior work has shown that videocapsule endoscopy image features are a useful tool for quantitatively distinguishing the intestinal mucosal surface of untreated celiac patients from that of controls. The use of dynamic estimates of wall motility may further help to improve classification.

METHODS

Videocapsule endoscopy clips (200 frames each, 2 frames/s, 576 × 576 pixels/frame) were acquired at five small intestinal locations in 11 untreated celiac patients (celiacs) and ten controls. Color images were converted to grayscale and analyzed frame-by-frame. Variations in the position and width of the center of the small intestinal lumen were quantitatively estimated. The darkest grayscale pixels were used as an estimate of the lumen center. Over 200 frames, the standard deviation of the lumen center xy position and the mean and standard deviation in lumen center width were used as dynamic estimates of wall motility. These parameters were plotted in three-dimensional space, and the best discriminant function was used to classify celiacs versus controls at each of the following five locations: (1) duodenal bulb, (2) distal duodenum, (3) jejunum, (4) ileum, and (5) distal ileum.

RESULTS

The overall sensitivity for the classification of celiacs versus controls at all five locations was 98.2 %, while the specificity was 96.0 %. From location 1 to 5, there was a tendency for the lumen center width to diminish in terms of frame-to-frame variability by 7.6 % in celiacs (r (2) = 0.4) and 9.7 % in controls (r (2) = 0.7).

CONCLUSIONS

In addition to examining the mucosal surface, videocapsule endoscopy can assess small bowel intestinal motility and aid in distinguishing celiac patients from controls.

Adaptable image cuts for motility inspection using WCE

The Wireless Capsule Endoscopy (WCE) technology allows the visualization of the whole small intestine tract. Since the capsule is freely moving, mainly by the means of peristalsis, the data acquired during the study gives a lot of information about the intestinal motility. However, due to: (1) huge amount of frames, (2) complex intestinal scene appearance and (3) intestinal dynamics that make difficult the visualization of the small intestine physiological phenomena, the analysis of the WCE data requires computer-aided systems to speed up the analysis. In this paper, we propose an efficient algorithm for building a novel representation of the WCE video data, optimal for motility analysis and inspection. The algorithm transforms the 3D video data into 2D longitudinal view by choosing the most informative, from the intestinal motility point of view, part of each frame. This step maximizes the lumen visibility in its longitudinal extension. The task of finding "the best longitudinal view" has been defined as a cost function optimization problem which global minimum is obtained by using Dynamic Programming. Validation on both synthetic data and WCE data shows that the adaptive longitudinal view is a good alternative to the traditional motility analysis done by video analysis. The proposed novel data representation a new, holistic insight into the small intestine motility, allowing to easily define and analyze motility events that are difficult to spot by analyzing WCE video. Moreover, the visual inspection of small intestine motility is 4 times faster then by means of video skimming of the WCE.

Transformation of videocapsule images to detect small bowel mucosal differences in celiac versus control patients

BACKGROUND

Videocapsule endoscopy can be useful to detect small intestinal pathology in celiac disease patients. However, presence of extraneous features including air bubbles and opaque fluids can complicate the analysis. A technique for quantitative analysis of videocapsule images is presented that is robust to presence of extraneous features.

METHOD

Videocapsule clips were acquired from five small intestinal locations in 12 celiacs with villous atrophy and 11 control patients. Clips were 200 frames in length, their resolution was 576 × 576 pixels and 256 grayscale levels, with 2/s frame rate. The dominant period (DP), defined as the tallest peak in the ensemble average power spectrum, was computed over each clip without removal of extraneous features. Ensemble average basis images were constructed, and measurements were made of their frame-to-frame variation in brightness and texture.

RESULTS

From pooled basis images, celiac images had greater texture than controls and exhibited more brightness variation (p<0.05 in mean and p<0.01 in standard deviation). In celiacs, correlation existed between greater textural alterations versus longer DP (r²=0.47), and between greater brightness variation and longer DP (r²=0.33). There was no significant correlation between quantitative features and DP in controls (r²<0.25).

CONCLUSIONS

Using this new method, celiac videoclips were quantitatively distinguishable from control videoclips without manual or computer-assisted detection, masking, and removal of extraneous image features. Furthermore, in celiac but not control basis images, larger textural and brightness alterations were correlated to longer DP. Greater textural and brightness alterations, and thus longer periodicities, are likely related to presence of villous atrophy.

Video capsule endoscopy in celiac disease

Video capsule endoscopy (VCE) provides a safe, non-invasive way to visualize the small intestine and is helpful in celiac disease patients in select situations. VCE can be performed in patients who are unable or unwilling to undergo conventional endoscopy, those with positive celiac serology with normal duodenal biopsies, and also in those who develop alarm symptoms. VCE has limitations including subjective interpretation. Techniques are being developed to standardize assessment of VCE images in patients with known or suspected celiac disease. Pilot studies using computer-based quantification methods have shown promise in examining the 3-dimensional mucosal structure and motility.

Is enteroscopy necessary for diagnosis of celiac disease?

Celiac disease (CD) is an autoimmune inflammatory disease of the small intestine as a result of reaction to wheat protein, gluten. Exclusion of dietary gluten is the mainstay of the treatment that necessitates a precise diagnosis of the disease. Serological screening may aid in identifying patients with suspected CD, which should be confirmed by intestinal biopsy. It has been shown that duodenal biopsies are good for detection of the disease in most patients. However, there is a group of patients with positive serology and inconclusive pathology. As a result of the widespread use of serology, many patients with equivocal findings grow quickly. Unfortunately current endoscopic methods can only diagnose villous atrophy, which can be present in the later grades of disease (i.e., Marsh III). To diagnose CD correctly, going deeper in the intestine may be necessary. Enteroscopy can reveal changes in CD in the intestinal mucosa in 10%-17% of cases that have negative histology at initial workup. Invasiveness of the method limits its use. Capsule endoscopy may be a good substitute for enteroscopy. However, both techniques should be reserved for patients with suspected diagnosis of complications. This paper reviews the current literature in terms of the value of enteroscopy for diagnosis of CD.

Computer-aided decision support systems for endoscopy in the gastrointestinal tract: a review

Today, medical endoscopy is a widely used procedure to inspect the inner cavities of the human body. The advent of endoscopic imaging techniques-allowing the acquisition of images or videos-created the possibility for the development of the whole new branch of computer-aided decision support systems. Such systems aim at helping physicians to identify possibly malignant abnormalities more accurately. At the beginning of this paper, we give a brief introduction to the history of endoscopy, followed by introducing the main types of endoscopes which emerged so far (flexible endoscope, wireless capsule endoscope, and confocal laser endomicroscope). We then give a brief introduction to computer-aided decision support systems specifically targeted at endoscopy in the gastrointestinal tract. Then we present general facts and figures concerning computer-aided decision support systems and summarize work specifically targeted at computer-aided decision support in the gastrointestinal tract. This summary is followed by a discussion of some common issues concerning the approaches reviewed and suggestions of possible ways to resolve them.

Diagnosing celiac disease by video capsule endoscopy (VCE) when esophagogastroduodenoscopy (EGD) and biopsy is unable to provide a diagnosis: a case series

Background

Video capsule endoscopy (VCE) is mainly used to evaluate patients with celiac disease in whom their course after diagnosis has been unfavorable and the diagnosis of adenocarcinoma, lymphoma or refractory celiac disease is entertained, but it has been suggested that VCE could replace esophagogastroduodenoscopy (EGD) and biopsy under certain circumstances.

Methods

We report a single center case series of 8 patients with suspected celiac disease who were diagnosed by VCE.

Results

EGD and biopsy had been performed in 4 patients resulting in a negative biopsy, declined by 2, and contraindicated in 2 due to hemophilia and von Willebrand disease. In all patients, mucosal changes of scalloping, mucosal mosaicism and reduced folds were seen in either the duodenum or jejunum on VCE. Follow-up in 7 patients demonstrated improvement in either their serological abnormalities or their presenting clinical features on a gluten-free diet.

Conclusions

Our case series demonstrates that VCE and the visualization of the characteristic mucosal changes of villous atrophy may replace biopsy as the mode of diagnosis when EGD is either declined or contraindicated, or when duodenal biopsies are negative and there remains a high index of suspicion. Further study is needed to clarify the role and cost of diagnosing celiac disease with VCE.

Advances in small bowel imaging

The small intestine has been difficult to examine by traditional endoscopic and radiologic techniques. Within the past 10 years, advances have led to an explosion of technologies that facilitate examination of the entire small intestine. Wireless video capsule endoscopy, deep enteroscopy using balloon-assisted or spiral techniques, computer tomography (CT) and magnetic resonance (MR) enterography have facilitated the diagnosis, monitoring, and management of patients with small intestinal diseases. These technologies are complementary, each with its advantages and limitations. Capsule endoscopy provides a detailed view of the mucosal surface and has excellent patient acceptance, but does not allow therapeutics. Deep enteroscopy allows careful inspection of the mucosa and therapeutics, but is time consuming and invasive. Enterography (CT or MR) allows examination of the small bowel wall and surrounding structures. The initial best test for detecting small intestinal disease depends on clinical presentation and an astute differential diagnosis.

Endoscope distortion correction does not (easily) improve mucosa-based classification of celiac disease

Distortion correction is applied to endoscopic duodenal imagery to improve automated classification of celiac disease affected mucosa patches. In a set of six edge- and shape-related feature extraction techniques, only a single one is able to consistently benefit from distortion correction, while for others, even a decrease of classification accuracy is observed. Different types of distortion correction do not lead to significantly different behaviour in the observed application scenario.

Update on celiac disease - etiology, differential diagnosis, drug targets, and management advances

Celiac disease (CD) is an immune-mediated enteropathy triggered by exposure to wheat gluten and similar proteins found in rye and barley that affects genetically susceptible persons. This immune-mediated enteropathy is characterized by villous atrophy, intraepithelial lymphocytosis, and crypt hyperplasia. Once thought a disease that largely presented with malnourished children, the wide spectrum of disease activity is now better recognized and this has resulted in a shift in the presenting symptoms of most patients with CD. New advances in testing, both serologic and endoscopic, have dramatically increased the detection and diagnosis of CD. While the gluten-free diet is still the only treatment for CD, recent investigations have explored alternative approaches, including the use of altered nonimmunogenic wheat variants, enzymatic degradation of gluten, tissue transglutaminase inhibitors, induction of tolerance, and peptides to restore integrity to intestinal tight junctions.