A convolutional neural network approach for abnormality detection in Wireless Capsule Endoscopy

With the development of wireless capsule endoscopy (WCE) and its extensive applications in clinic, doctors need to spend more time on reviewing the WCE images for lesions diagnosis. Therefore, automatic lesion detection for WCE has gradually become a research hotspot, which aims to reduce the pressure of doctors and improves the diagnosis efficiency. Many researchers adopted the traditional machine learning method to realize polyp detections, however, these methods need to extract the features manually, which were unable to find higher features of WCE images. So, in this study, we proposed a novel method that based on convolution neural network (CNN) to automatically recognize polyp in small bowel WCE image. We utilized the Alexnet architecture, one of the classical CNN, to extract the features of WCE images and classify polyp images from normal ones. We selected 14408 images from different patients, including 408 polyp images and 14000 normal images. Since the amount of initial polyp images is small, then, we did the data augmentation, including rotation, luminance change, blurring, and noise. At last the experimental results demonstrated that the method we proposed had a promising performance in polyp detection, whose accuracy, sensitivity and specificity can reach at 99.88%, 99.40% and 99.93%, respectively. Additionally, we evaluated ROC curve and its AUC value, which further confirmed that our model has a high accuracy and reliability in polyp detection. This proposed method has great potential to be used in the clinical examination to help doctors from the tedious image reviewing work.

Diseases of the digestive tract, such as ulcers, pose a serious threat to human health. In fact, many types of endoscopy are employed to examine the patient’s gastrointestinal tract. Recently, wireless capsule endoscopy (WCE) is presented as an excellent diagnostic tool for evaluation of gastrointestinal diseases compared to traditional endoscopies. This diagnosis usually takes a long time, which is tiring, and so the doctors may miss parts where abnormalities of the gastrointestinal tract may present. Therefore, automated diagnostic technics to detect symptoms of gastrointestinal illness in WCE images is adopted as an excellent enhancement tool for these doctors. In this work, a new computer-aided diagnosis method for ulcer detection in WCE images is proposed. After a preprocessing step, fine-tuned convolutional neural network (CNN) is used to extract deep features from these images. Since the number of ulcer images in the available data sets is limited, the CNN networks used in this work were pre-trained on millions of labeled natural images (ImageNet). After the deep features extraction, a random forest classifier is employed to detect ulcer from WCE images. The proposed approach demonstrates promising results (96.73 % and 95.34 % in terms of precision and recall respectively). Those results are satisfactory when compared to recent state-of-the-art methods.

As one of the most common human helminths, hookworm is a leading cause of maternal and child morbidity, which seriously threatens human health. Recently, wireless capsule endoscopy (WCE) has been applied to automatic hookworm detection. Unfortunately, it remains a challenging task. In recent years, deep convolutional neural network (CNN) has demonstrated impressive performance in various image and video analysis tasks. In this paper, a novel deep hookworm detection framework is proposed for WCE images, which simultaneously models visual appearances and tubular patterns of hookworms. This is the first deep learning framework specifically designed for hookworm detection in WCE images. Two CNN networks, namely edge extraction network and hookworm classification network, are seamlessly integrated in the proposed framework, which avoid the edge feature caching and speed up the classification. Two edge pooling layers are introduced to integrate the tubular regions induced from edge extraction network and the feature maps from hookworm classification network, leading to enhanced feature maps emphasizing the tubular regions. Experiments have been conducted on one of the largest WCE datasets with WCE images, which demonstrate the effectiveness of the proposed hookworm detection framework. It significantly outperforms the state-of-the-art approaches. The high sensitivity and accuracy of the proposed method in detecting hookworms shows its potential for clinical application.

Self-supervised out-of-distribution detection in wireless capsule endoscopy images

This paper presents a methodology for detecting abnormal patterns in wireless capsule endoscopy (WCE) images. In particular, an average of 50,000 images are obtained during an WCE exam. Usually, these images are reviewed in a form of a video at speeds between 5 to 40 image-frames/sec. The time spent by a physician reading the results of WCE images varies between 45 to 180 minutes. This presents a major problem which is the reading process that consumes a significant amount of time and the results take several days before they become available since the physician has to find the time to study each video uninterrupted for up to 3 hours. The methodology presented here is based on the automatic detection of abnormal WCE patterns in an effort for reducing the reading time of the WCE images and the cost of the procedure as well. The methodology consists of a synergistic integration of image processing, analysis and recognition techniques for achieving the automatic detection of the WCE abnormal patterns.

This paper presents a novel multi-level approach for bleeding detection in Wireless Capsule Endoscopy (WCE) images. In the low-level processing, each cell of K×K pixels is characterized by an adaptive color histogram which optimizes the information representation for WCE images. A Neural Network (NN) cell-classifier is trained to classify cells in an image as bleeding or non-bleeding patches. In the intermediate-level processing, a block which covers 3×3 cells is formed. The intermediate-level representation of the block is generated from the low-level classifications of the cells, which captures the spatial local correlations of the cell classifications. Again, a NN block-classifier is trained to classify the blocks as bleeding or non-bleeding ones. In the high-level processing, the low-level cell-based and intermediate-level block-based classifications are fused for final detection. In this way, our approach can combine the low-level features from pixels and intermediate-level features from local regions to achieve robust bleeding detection. Experiments on real WCE videos have shown that the proposed method of multi-level classification is not only accurate in both detection and localization of potential bleedings in WCE images but also robust to complex local noisy features.

Wireless capsule endoscopy (WCE) can painlessly capture a large number of images inside the intestine. However, only a small portion of these WCE images contain hemorrhage. It is thus critical to develop automated hemorrhage detection method to facilitate the diagnosis of intestinal diseases. However, automated hemorrhage detection is complicated by 1) the extreme imbalance between the amount of hemorrhage images and that of normal images; and 2) the variety of the appearance, texture, and luminance inside the intestine. In this paper, we proposed to learn a robust intestinal hemorrhage detection model via Convolutional Neural Networks (CNNs), because of CNNs' extraordinary performance in solving various image understanding tasks. Specially, we explored different CNN architectures and data augmentation methods. Besides, we investigated the correlation between hemorrhage detection accuracy and image quality. Across about 1.3k hemorrhage images and 40k normal images, the learned CNN model achieves an F-measure of 98.87%.

High pooled performance of convolutional neural networks in computer-aided diagnosis of GI ulcers and/or hemorrhage on wireless capsule endoscopy images: a systematic review and meta-analysis

Wireless capsule endoscopy (WCE) is a device that can move through human body and capture the small bowel entirely. Thus, it is presented as an excellent diagnostic tool for evaluation of gastrointestinal diseases compared with traditional endoscopies. However, the diagnosis by the physicians is tedious since it requires reviewing the video extracted from the capsule and analysing all of its frames. This tedious task has fuelled the efforts of researchers to provide automated diagnostic techniques for WCE frameworks to detect symptoms of gastrointestinal illness. In this paper, a new computer-aided diagnosis method for abnormalities detection in WCE images is proposed. After a preprocessing step, we extract from these images the descriptor we feed to a kernel extreme learning machine to perform the classification process. The descriptor used in this work is a combination between the histogram of oriented gradients (HOG) that were extracted using the hue component of the HSV colour space, and a modified rotation-invariant local binary pattern. The proposed approach has been tested on different datasets, and the results obtained are satisfactory when compared to the state-of-the-art works.

Ulcer detection in Wireless Capsule Endoscopy images using deep CNN

A new feature extraction technique for the detection of lesions created from mucosal inflammations in Crohn’s disease, based on wireless capsule endoscopy (WCE) images processing is presented here. More specifically, a novel filtering process, namely Hybrid Adaptive Filtering (HAF), was developed for efficient extraction of lesion-related structural/textural characteristics from WCE images, by employing Genetic Algorithms to the Curvelet-based representation of images. Additionally, Differential Lacunarity (DLac) analysis was applied for feature extraction from the HAF-filtered images. The resulted scheme, namely HAF-DLac, incorporates support vector machines for robust lesion recognition performance. For the training and testing of HAF-DLac, an 800-image database was used, acquired from 13 patients who undertook WCE examinations, where the abnormal cases were grouped into mild and severe, according to the severity of the depicted lesion, for a more extensive evaluation of the performance. Experimental results, along with comparison with other related efforts, have shown that the HAF-DLac approach evidently outperforms them in the field of WCE image analysis for automated lesion detection, providing higher classification results, up to 93.8% (accuracy), 95.2% (sensitivity), 92.4% (specificity) and 92.6% (precision). The promising performance of HAF-DLac paves the way for a complete computer-aided diagnosis system that could support physicians’ clinical practice.

Weakly supervised methods, such as class activation maps (CAM) based, have been applied to achieve bleeding segmentation with low annotation efforts in Wireless Capsule Endoscopy (WCE) images. However, the CAM labels tend to be extremely noisy, and there is an irreparable gap between CAM labels and ground truths for medical images. This paper proposes a new Discrepancy-basEd Active Learning (DEAL) approach to bridge the gap between CAMs and ground truths with a few annotations. Specifically, to liberate labor, we design a novel discrepancy decoder model and a CAMPUS (CAM, Pseudo-label and groUnd-truth Selection) criterion to replace the noisy CAMs with accurate model predictions and a few human labels. The discrepancy decoder model is trained with a unique scheme to generate standard, coarse and fine predictions. And the CAMPUS criterion is proposed to predict the gaps between CAMs and ground truths based on model divergence and CAM divergence. We evaluate our method on the WCE dataset and results show that our method outperforms the state-of-the-art active learning methods and reaches comparable performance to those trained with full annotated datasets with only 10% of the training data labeled. The source code is available at https://github.com/baifanxxx/DEAL.KeywordsActive learningSegmentationWCE images

Deep convolutional neural networks with ensemble learning and transfer learning for automated detection of gastrointestinal diseases

Wireless capsule endoscopy (WCE) is an innovative solution for gastrointestinal disease detection. Limited by WCE hardware and cost of manufacture, WCE image resolution is commonly low, which creates problems for attention to image details and visual perception in medical diagnosis. Under the sparse representation framework, we propose an adaptive dictionary pair learning method to obtain more appropriate representation of each patch with more relevant atoms according to patch content. Specifically, the dictionary pair is learned from high-low resolution cluster patches based on sparse constraint of input patches. Careful examination of the WCE images show there exist unnatural block areas. In order to further improve performance, the autoregressive model is applied to enhance local structure. Intensive experiments have been conducted on WCE image dataset and natural image dataset, including comparison test between the state-of-art methods and ours, and the results validate the effectiveness of the proposed method both on visual perception effect and objective indices.

Wireless capsule endoscopy (WCE) is a device used to inspect the gastrointestinal (GI) track. This technology is noninvasive compared to other methods that are traditionally adopted in the examination of GI track. From the physicians' point of view, the WCE is a favorable approach because of its efficiency and accuracy. In this paper, a new discriminant mechanism of bleeding regions is proposed based on the use of Support Vector Machine (SVM) classifier. Different from the traditional SVM approach, in the proposed method single pixels are not used for training and testing data, however a cluster of pixels are used. This approach aims to eliminate some very small judged bleeding areas which in fact are not. The reported results demonstrate that the accuracy of the proposed method is significantly increased in comparison with traditional approaches. In addition, another improved method based on the square comparison is also proposed, and this has increased the gain.