Search

Wireless capsule endoscopy (WCE) is an efficient tool to investigate gastrointestinal tract disorders and perform painless imaging of the intestine. Despite that, several concerns make its wide applicability and adaptation challenging like efficacy, tolerance, safety, and performance. Besides, automatic analysis of the WCE provided dataset is of great importance for detecting abnormalities. Imaging of the patient’s digestive tract through WCE produces a large dataset that requires a substantial amount of time and a special skill set from a medical practitioner for analysis. Several computer-aided and vision-based solutions have been proposed to resolve these issues, yet, they do not provide the desired level of accuracy and further improvements are still needed. The current study aims to devise a system that can perform the task of automatic analysis of WCE images to identify abnormalities and assist practitioners for robust diagnosis. This study adopts a deep neural network approach and proposes a model name BIR (bleedy image recognizer) that combines the MobileNet with a custom-built convolutional neural network (CNN) model to classify WCE bleedy images. BIR uses the MobileNet model for initial-level computation for its lower computation power requirement and subsequently the output is fed to the CNN for further processing. A dataset of 1650 WCE images is used to train and test the model using the measures of accuracy, precision, recall, F1 score, and Cohen’s kappa to evaluate the performance of the BIR. Results indicate the promising outcomes with achieved accuracy, precision, recall, F1 score, and Cohen’s kappa of 0.993, 1.000, 0.994, 0.997, and 0.995 respectively. The accuracy of the BIR model is 0.978 with the Google collected WCE image dataset which is better than the state-of-art approaches.

Background: Computer-aided disease detection schemes from wireless capsule endoscopy (WCE) videos have received great attention by the researchers for reducing physicians’ burden due to the time-consuming and risky manual review process. While single disease classification schemes are greatly dealt by the researchers in the past, developing a unified scheme which is capable of detecting multiple gastrointestinal (GI) diseases is very challenging due to the highly irregular behavior of diseased images in terms of color patterns. Method: In this paper, a computer-aided method is developed to detect multiple GI diseases from WCE videos utilizing linear discriminant analysis (LDA) based region of interest (ROI) separation scheme followed by a probabilistic model fitting approach. Commonly in training phase, as pixel-labeled images are available in small number, only the image-level annotations are used for detecting diseases in WCE images, whereas pixel-level knowledge, although a major source for learning the disease characteristics, is left unused. In view of learning the characteristic disease patterns from pixel-labeled images, a set of LDA models are trained which are later used to extract the salient ROI from WCE images both in training and testing stages. The intensity patterns of ROI are then modeled by a suitable probability distribution and the fitted parameters of the distribution are utilized as features in a supervised cascaded classification scheme. Results: For the purpose of validation of the proposed multi-disease detection scheme, a set of pixel-labeled images of bleeding, ulcer and tumor are used to extract the LDA models and then, a large WCE dataset is used for training and testing. A high level of accuracy is achieved even with a small number of pixel-labeled images. Conclusion: Therefore, the proposed scheme is expected to help physicians in reviewing a large number of WCE images to diagnose different GI diseases., An automatic multiple GI disease detection scheme is proposed in this paper based on PDF modeling of multiple linear discriminant analysis-based regions of interest from from wireless capsule endoscopy videos.

Endoscopy is an effective tool for early diagnosis and treatment of gastrointestinal diseases. An increasing number of medical institutions prefer wireless capsule endoscopy (WCE) rather than traditional endoscopies. However, the capsule endoscopy that launched in the current commercial market have disadvantages of inability of active locomotion and low data transmission rate. This paper presents a 6 square coils electromagnetic control system with Wi-Fi based video transmission module for WCE. By adjusting the driven current of these coils, the electromagnetic device is able to generate uniform magnetic field to control the rotation as well as a gradient magnetic field to control the propulsion of the capsule endoscopy. In addition, the proposed WCE improves the transmission rate via high frequency communication technology. This WCE is Wi-Fi enabled in the frequency of 2.4 GHz which achieves a data rate up to 150 Mbps. The designed WCE employs an in-body transceiver receiving control signals and transferring image data outside the body. Compare to current commercial capsule endoscopy, the experiment results demonstrate that the proposed WCE achieves high transmission rate up to 30 frame per second (fps). Furthermore, it realizes the five degree of freedom motion and improves the control precision.

In this study, a square spiral conformal antenna design operating in the 433 MHz (ISM) band to be used in wireless capsule endoscopy applications has been designed. Antenna characteristics have been studied for different environments of the body and have been shown to antenna be workable in all environments. In order to test the data transmission between the capsule and the receiver unit located outside the human body, the S21 parameter simulations were carried out using the dipole antenna as the receiving antenna. It has been shown that the designed antenna is suitable for capsule endoscopy applications with the obtained data transmission and antenna parameters.

Wireless capsule endoscopy (WCE) is a modality used for the non-invasive examination of the gastrointestinal (GI) tract. Physicians diagnose pathologies in images derived from Capsule Endoscopy (CE) using specific gaze patterns to observe pathologically related visual cues. Lately, deep learning has advanced in the domain of human eye-fixation estimation in natural images. However, the potentials of predicting the eye related patterns, such as eye fixations, in medical images has not been thoroughly investigated. In this work, we propose a CNN auto-encoder model, that is capable of predicting saliency maps estimating the gaze-patterns, in terms of eye-fixations, of physicians in CE images. The proposed model outperforms other approaches for visual saliency estimation based on physicians' eye fixation by providing an AUC-J of 0.726 among CE images depicting various pathological and normal cases.

Active wireless capsule endoscopy (WCE) under magnetic actuation is a promising technology to reduce the inspection time and relieve the burden of physicians. In this paper, we propose a reciprocally rotating magnetic actuation method for trajectory following of a capsule and develop its dynamic model. For the trajectory following task, we investigate the closed-loop tracking control strategies based on different controllers to actuate the capsule in the complex environments. The effectiveness of our method is validated in extensive experiments in a simulation environment as well as in an ex-vivo pig colon. The results demonstrate that the proposed method can accurately and efficiently actuate the capsule to follow the desired trajectory in the complex environments, achieving tracking errors on the order of millimeter. Moreover, the experiments on the ex-vivo pig colon show that the proposed reciprocally rotating magnetic actuation method has the potential to reduce the clinical risks and improve the safety and clinical acceptability of this technology.

Bleeding is a common symptom of many gastrointestinal (GI) diseases. Hence, early detection of bleeding is a very important thing in the medical field. Wireless capsule endoscopy (WCE) is a major breakthrough in the medical field. This makes it very easy to diagnose various gastrointestinal (GI) related ailments. The primary benefit of WCE is painless diagnostic. As a result, it has received favorable reviews from physicians and patients. The WCE generates several images of the small intestine during diagnosis. However, only a highly trained doctor can make an effective diagnosis of bleeding images from this. Accuracy is an unavoidable factor in medical imaging. Even a small flaw inaccuracy can have huge repercussions. Therefore developing a high accurate WCE bleeding detection system is a very urgent research problem. This article proposes an improved AlexNet architecture to address the limitation of bleeding detection in the WCE procedure. The average accuracy of the proposed AlexNet is 94.5 % and its sensitivity and specificity are 95.24 % and 96.72 % respectively.

In the modern era, humans are suffering from life-threatening diseases that are curable if detected in the early stages. Wireless capsule endoscopy is a modern, non-invasive technique to diagnose the gastrointestinal tract for cancer, Crohn's disease, bleeding, etc. Hence this diagnostic technique has the utmost importance which attracted attention in recent times. To achieve higher data rates and frame rates with higher image resolution, an inner wall dual loop antenna with a u-slotted patch at the UWB band (3.1 GHz to 10.6 GHz) is proposed in this paper. The simulated results show the antenna's excellent performance in terms of return loss with a bandwidth of 7.5 GHz ranging from 3.1 GHz to 10.6 GHz. The antenna is quasi omnidirectional with maximum gains of -36.78 dBi, -38 dBi, -34.36 dBi, and -34.68 dBi at 3 GHz, 5 GHz, 8 GHz, and 10 GHz respectively which is common in antennas for wireless capsule endoscopy because of the highly lossy environment.

Background: Direct detection of polyps from colon capsule endoscopy (CCE) videos is an ultimate goal not only for physicians but also for biomedical engineers who are working on automatic internal lesions like polyps. There is also a great enthusiasm among biomedical professionals to make advanced systems for aiding doctors to have a faster and accurate diagnosis by the means of polyp detection from CCE acquired video streams. Such systems must be able to localize polyps correctly and extract the whole lesions from the video frames completely.Material and Methods: In this paper, a new approach toward object-wise polyp detection from CCE frames in a video stream is proposed. The proposed method employs modified region proposal CNNs to localize the existing polyps from CCE acquired video frames and after that a level-set method known as Distance Regularized Level Set Evolution (DRLSE) is employed for automatic model-based segmentation of localized polyps. The pixel-wise detection of polyps is necessary for polyp classification and will help gastroenterologists to determine appropriate prognosis and treatment for the patients.Results and conclusion: The proposed method is trained by the means of an CCE still image dataset which includes manually annotated polyps. The trained network is then applied to CCE video images. The results demonstrate that the proposed method is able to localize and detect polyps both region-wise and pixel-wise with a good rate of accuracy.

The antenna is an important component to realize the wireless communication between the wireless capsule endoscope system inside the human body and the outside world. This paper designs an ultra-wideband conformal capsule antenna with the volume of 18mm3, which uses in wireless capsule endoscopy systems. The antenna is a coplanar waveguide-fed compact planar slotted microstrip antenna, which uses flexible materials to achieve conformal antennas, and is used in the industrial, scientific, and medical frequency bands of 2.45-2.48 GHz (ISM). Under the premise of satisfying radiation safety, while ensuring the performance of the antenna, the ultrawideband operation of the antenna is realized with a bandwidth of 108.9%, and the gain of the antenna at 2.45GHz is −25.4dBi. The research results provide an important reference for the development and application of conformal capsule antennas.

We demonstrate a proof of concept highly miniaturised fluorescence imager and its application to detecting cancer in resected human colon cancer tissues. Fluorescence imaging modalities have already been successfully implemented in traditional endoscopy. However, the procedure still causes discomfort and requires sedation. Wireless fluorescence capsule endoscopy has the potential to improve diagnostic accuracy with less inconvenience for patients. In this paper we present a 5 mm $\times6$ mm $\times5$ mm optical block that is small enough to integrate into a capsule endoscope. The block integrates ultrathin filters for optical isolation and was successfully integrated with a sensitive CMOS SPAD array to detect green fluorescence from Flavin Adenine Dinucleotide (FAD), which is an endogenous fluorophore responsible for autofluorescence in human tissues, and fluorescence from the cancer selective molecular probe ProteoGREEN™-gGlu used to label colorectal cancer cells. In vitro studies were validated using a commercial Modulus™ Microplate reader. The potential use of the device in capsule endoscopy was further validated by imaging healthy and malignant resected human tissues from the colon to detect changes in autofluorescence signal that are crucial for cancer diagnosis.

This paper proposes a position and posture estimation method for capsule endoscopy. The proposed method aims to enable daily life diagnosis with a wearable device, which is a neck corset that includes a coil. The geomagnetic and acceleration sensors in the capsule endoscopy and the corset are used for localization. The accuracy of the estimated position and posture is evaluated with hardware measurements. Experimental results show that the proposed method estimates the posture within five-degree error and the position with the maximum error of 21.4 mm.

Conventional radiological and endoscopic techniques utilizing long tubes were ineffective in visualizing the small bowel mucosa until the development of wireless capsule endoscopy (WCE). WCE is a revolutionary endoscopic technology that can diagnose the complete gastrointestinal tract. However, the existing capsule technologies are passive, and thus they cannot be navigated to or held in a specific location. The design of an active capsule will present the opportunity to move and stop a device at any targeted locations leading to numerous medical applications such as drug delivery or collecting tissue samples for examinations in the laboratory. This paper implements a new locomotion methodology for WCE systems using an electromagnetic platform. The platform produces a dynamic electromagnetic field to control the motion of the capsule. The strength and the direction of the electromagnetic field that is generated by the platform are continuously adjusted in order to maintain the equilibrium state during the capsule movement. We present the detailed design of the proposed platform with an experimental setup with polyvinyl chloride tubes and ex vivo to demonstrate the performance of the capsule motion.

The active endoscopy system has a locomotion system that allows the endoscopist to control the capsule's motion in the GI tract. The endoscopist can move, or stop the capsule at any targeted locations, which provides a better and more reliable diagnosis. We propose a novel design of a capsule's locomotor with shape memory alloy (SMA) actuators. The hinge mechanism to translates the linear movement of the SMA into a translational motion to the capsule's links. The proposed design has the added advantage of steering the capsule's camera system by activating one or more SMA wires. To the best of our knowledge, this is the first proposed active capsule that has a steering mechanism that controls the camera view angle at the top of the capsule. This camera's steering mechanism leads to a wider angle range for the system without adding to the capsule's weight and size. Also, this will help to better capture pathological areas without increasing power consumption. The proposed capsule prototype is manufactured using 3D printing technology and powered by a battery installed in the capsule.

In this study, a conformal loop antenna at 400 MHz frequency was designed and implemented for capsule endoscopy applications. The performance of the designed antenna has been further tested for the digestive system organs through which the capsule will move. In order to realize the designed antenna, the capsule is pressed with a three-dimensional printer, and the loop antenna printed on the copper plate is wrapped conformally on the outer surface of the capsule with the kapton tape. The results obtained from the measurements carried out in the laboratory were consistent with the simulation results.

In wireless capsule endoscopy (WCE), transmitting chroma along with luma components through radio frequency transmission consumes large amount of power. Conventional methods transmit both luma along with sub-sampled chroma components. In the proposed method only luma component is encoded and transmitted. Chroma component is predicted from reconstructed luma at the decoder using generative adversarial network. To improve the quality of the reconstructed WCE images, chroma components are Wyner-Ziv encoded. Predicted chroma at the decoder serves as side information for decoding chroma components. Transmission of chroma components using proposed method needs very few bits compare to other WCE-JPEG based image compression methods. The proposed compression method achieves better compression ratio of 94.73 % at PSNR 43.59dB compared to other WCE image compression methods.

Wireless capsule endoscopy (WCE) is a medical device for the inspection of some parts of the gastrointestinal (GI) tract. Currently, WCE transmits to the outside low resolution video, although high quality images would be necessary for improving the detection of diseases. Ultra wideband (UWB) frequency band is a good candidate to achieve such requirement due to its large available bandwidth. Besides, the knowledge of the exact location of WCE would also help physicians to locate the suspicious tissue so diagnosis time could be shortened. Received Signal Strength (RSS) based localization is one of the simplest localization techniques. However, RSS localization requires a good knowledge of the path loss model, what is not easy for in-body communications. The objective of this paper is to provide a review of the problem of modelling the path loss for WCE in UWB as well as give implementation guidelines for WCE localization.

Wireless capsule endoscopy (WCE) is a minimally invasive procedure performed with a tiny swallowable optical endoscope that allows exploration of the human digestive tract. The medical device transmits tens of thousands of colour images, which are manually reviewed by a medical expert. This paper highlights the significance of using inputs from multiple colour spaces to train a classical U-Net model for automated semantic segmentation of eight WCE abnormalities. We also present a novel approach of grouping similar abnormalities during the training phase. Experimental results on the KID datasets demonstrate that a U-Net with 4-channel inputs outperforms the single-channel U-Net providing state-of-the-art semantic segmentation of WCE abnormalities.

The ulcer is one of the most common and dangerous among the effect of many deadly diseases in the Gastrointestinal tract. It is complicated to diagnose and detect the tiny intestine ulcers by applying other alternative methods of endoscopy. Wireless Capsule Endoscopy (WCE) technique is rapidly using more conveniently to visualize these ulcers. However, it is challenging and time-consuming for the clinicians to check the vast amount of images captured from the WCE. So, it has become the most crucial concern to provide an automated system for detecting the ulcer to help the clinicians. In this research paper, a unique automatic ulcer diagnosis model is introduced to detect ulcers from images that have been converted from the captured WCE video. In the proposed method, Some consecutive approaches, like pre-processing and fuzzy logic framework, have been applied for extracting the ulcer portion on L*a*b colour model. The proposed method has obtained a tremendous result of sensitivity 95%, accuracy 95.5%, specificity 97%, F1 score 96.48%, precision 98%, and negative predicted value 91% by utilizing the statistical feature and KNN classifier. Therefore, from the analysis of the analytical results and comparison studies, it is highly optimistic about having a positive impact on this research arena.

Wireless Capsule Endoscopy (WCE) is an advantageous tool for diagnosing gastrointestinal disorders. An automated analysis method is required to detect abnormality because a large number of images is produced along WCE journey through the patient‘s digestive tract. In this paper, a computer-aided method is proposed to detect bleeding frames. Bleeding is the most common abnormality in WCE images. This abnormality is distinguishable by using its color features. Hence, in this method, at first, different statistical features are extracted from color channels in different color spaces. Then, a feature selection method based on genetic algorithm is proposed to select a subset of the appropriate features. The experimental results show that the proposed feature selection method increases the detection accuracy and reduces the size of initial features set by almost half. The achieved accuracy, recall and precision of the proposed method are 97.58, 96.76 and 98.29, respectively.

This paper proposes a miniaturized quadrifilar helical antenna for wireless capsule endoscopy (WCE). The helix arms of the antenna are made with copper material. The helix height, radius, width and pitch angle is 8.85 mm, 3.3 mm, 0.4 mm and 13.7° respectively. The number of helix's turn is 0.5. A polyimide material is selected for modeling the capsule of WCE. The gastrointestinal tract (GI) with muscle phantom is also modeled using CST Microwave Studio (CST) software package. The performance of the antenna is analyzed in free space, capsule and muscle phantom. Resonant frequency of the antenna inside the body phantom is 3.36 GHz with S11 parameter of −10 dB and bandwidth 1 GHz ranging from 2.8 to 3.8 GHz. The antenna shows an omnidirectional radiation pattern and linear polarization when simulated inside the body phantom. The maximum radiation efficiency is −44.26 dB. Moreover, the gains of the antenna in free space, capsule, and the body model are obtained as 4.317 dB, 4.019 dB and −38.57 dB respectively.

This article demonstrates the design and development of a Minkowski fractal loop antenna for wireless capsule endoscopy (WCE) system. Targeted frequency of operation is 433.92 MHz (ISM band). The proposed structure occupies a layout area of around 8.715×7.79 mm2 over an FR4 substrate with a bandwidth of 170 MHz (350-520 MHz). Radiation pattern of the proposed antenna is studied for various locations of gastro-intestinal tract. For this study, two different inhomogeneous human phantoms (Male and Female) have been taken into consideration. Finally, link budget calculations were made to evaluate the signal to noise ratio power levels. Link margin finally indicates the suitability of the proposed antenna in biomedical application. There is a very close relevance between experimental results and simulated values.

In order to solve the problem of energy limitation in capsule endoscopy, the button battery can be replaced by an external wireless power. A small wireless power transfer system for delivering power to capsule endoscopy has been presented here. The wireless power transfer system is built by using a small cylindrical receiving resonator and two cylindrical source resonators to transfer power through coupled magnetic resonance. Experimentally, it is found that the wireless power transfer efficiency is stable within a certain range due to the stable magnetic field generated by two source resonators. The proposed design has the significant advantage of stable power transfer and small size for capsule endoscopy.

In this paper, a novel method is proposed to detect bleeding frames in capsule endoscopy images. Viewing directly the gastrointestinal tract is a general way for bleeding detection. Capsule endoscopy is a painless and non-invasive direct visual inspection way for diagnosing diseases in the gastrointestinal tract. This technology has many advantages over the other techniques. However, it is a tedious task for physicians to review the whole video frames. An automatic computer-aided method is then used in order to reduce the burden of physician. it is based on expectation maximization segmentation approach that decomposes the input image into potential regions of interest. Two sets of color and statistical features are then extracted from these regions. Finally, support vector machine is used to discriminate bleeding from normal frames. The obtained results on a large database demonstrate the efficacy of the proposed method in term of recall and precision measures

Wireless capsule endoscopy (WCE) is a swallowable device used for screening different parts of the human digestive system. Automatic WCE image analysis methods reduce the duration of the screening procedure and alleviate the burden of manual screening by medical experts. Recent studies widely employ convolutional neural networks (CNNs) for automatic analysis of WCE images; however, these studies do not consider CNN's structural and computational complexities. In this paper, we address the problem of simplifying the CNN's structure. A low complexity CNN structure for bleeding zone detection is proposed which takes a single patch as input and then outputs a segmented patch of the same size. The proposed network is inspired by the FCN paradigm with a simplified structure. Since it is based on image patches, the resulting network benefits from moderate-sized intermediate feature maps. Moreover, the problem of redundant computations in patch-based methods is circumvented by non-overlapping patch processing. The proposed method is evaluated using the publicly available KID dataset for WCE image analysis. Experimental results show that the proposed network has better accuracy and AUC than previous structures while requiring less computational operations.

Among the syndrome of many dangerous diseases in the human digestive tract ulcer is one of the most common. Diagnosis of small intestine ulcers cannot be detected by using other procedures of endoscopy. For visualizing these ulcers, wireless capsule endoscopy (WCE) technique is excessively being used. In WCE system a huge amount of images are achieved which is very much difficult and time erosive for physicians to check manually. So, for helping the physicians it is very much needed to provide an automated computer-aided system. In this paper, an automated computer-aided diagnosis system is mentioned for detecting ulcer from WCE images achieved from the capsule endoscopy video footage. In the first stage, we convert the source image into LAB color spaces for taking features. Then quadratic support vector machine (Q-SVM) classifier has been applied for separating the ulcer and non-ulcer images. Our achieved experiment results 90.39% accuracy and 99.00% sensitivity, validating the effectiveness of the proposed method. Therefore, results of the advanced methodology are given and statistical comparisons are also given related to the other works.

One-third of the world population suffers from diseases related to gastrointestinal (GI) tract. Capsule endoscopy is a non-sedative, non-invasive and patient-friendly technology to scan the entire GI tract. However, capsule endoscopy generates approximately 60000 images which make the diagnosis process time consuming and tiresome for physicians. Hence a computer-aided diagnosis system is a must. In this study, addresses a multi-class medical image analysis problem using image processing and machine learning techniques. This work proposes a system comprising preprocessing, feature extraction and classification of capsule endoscopy images for automatic detection of GI tract diseases. The system performs with an accuracy of 93% and precision of 91.9%.

Testing using the co-simulation method of an algorithm is important to visualize the hardware performance of the algorithm. This study aims to introduce a method for comparing the hardware performance of two different algorithms with a focus on the existing compression system for capsule endoscopy as a case study. The size limitation of capsule endoscopy which constrains the power source and the complexity of the hardware utilization are the main reasons for selecting the capsule endoscopy compression system for this case study. Undertaking a comparison between different algorithms from the existing image compression algorithms in capsule endoscopy can highlight the main testing steps of the co-simulation testing method. To allow a fair comparison, the hardware co-simulation is implemented using the same FPGA device for the methods compared in this study and the results acquired are discussed.

The passive movement of capsule endoscopy has geared research to investigate the possibility of developing a locomotion mechanism to allow the control of the capsule movement. The size and the power capacity of capsule endoscopy have been the main bottlenecks to achieve an efficient locomotion design. Running a locomotion device for a long time it seems to be impossible without the utilization of an external magnetic field that has been found to be unsafe for human risks. A short utilization of a locomotion mechanism, which is controlled from outside the capsule, could solve the problems of size and power capacity limitations. The short utilization of the simplified design could solve a major risk of capsule endoscopy which is capsule retention. This paper is set out to investigate the feasibility of developing an internal capsule retention release mechanism to solve the problem of capsule endoscopy retention. This includes the design and configuration of new capsule model to mount the proposed locomotion system. Two types of motors were tested in this paper; these are a two-phase stepper motor and a coin-type vibration motor. The design of the printed circuit board (PCB) of the nRF51×822 was reconfigured to fit within the small dimensions of the new model. The operation performance of both motors using a limited power source has been recorded which show that the proposed design is able to work efficiently for a limited time. However, a number of steps need to be undertaken to realize the feasibility of the new design for practice which includes testing in an animal or a fabricated model that allow for peristaltic movement similar to the human GI tract movement.

Wireless Capsule Endoscopy overcomes various limitations of ordinary endoscopy system including irritating wire and inability to diagnose complicated parts of intestine. Antenna is the main component that removes the hassle of wire. This paper represents the design and miniaturization of a patch antenna for capsule endoscopy with a resonance frequency in the MICS (402–405 MHz) band. The length and diameter of designed antenna are 28.47 mm and 12.45 mm respectively, resulting in a total volume of 4.493 cubic centimeters. The maximum specific absorption rate of this antenna is measured to be 0.0082 WIKg inside human phantom stomach.

Bleeding in gastrointestinal tract (GI) is caused by many different diseases and may lead to serious consequence if not being treated correctly. The most effective method to locate bleeding area is using wireless capsule endoscopy (WCE). However, the amount of WCE images of one patient is so large that even a professional physician may take a long time on analyzing them, and human errors may also occur with work time increasing. Thus the computer-aided diagnosis methods become more attractive. In this paper, we propose a novel deep learning based method for WCE images bleeding area segmentation. We design multi-stage architecture and attention blocks to deal with small areas segmentation. The latter stages' input is the combination of feature maps transferred from former stages so that even very deep layers can obtain small areas features, and attention blocks help shallow layers to better extract small areas features thus they can transfer more useful information to latter stages. Extensive experiments are conducted on public available WCE image dataset to show the effect of multi-stage architecture and attention blocks. Compared with other bleeding areas segmentation methods, our approach achieves state-of-the-art performance with 98.5% overall accuracy, 90.1% mean accuracy and 86.3% mean intersection over union (IoU). There is a 10.7% improvement on mean IoU compared with the previous state-of-the-art WCE bleeding segmentation method.

Ultrasound capsule endoscopy (USCE) has attracted increased interest recently. In order to image the walls of the gastrointestinal (GI) tract with USCE efficiently, both high-frequency ultrasound (i.e. microultrasound) imaging and good acoustic coupling are needed. Tissue to be scanned is expected close to the transducer surface, but the near-field tissue echoes may be easily lost in the acoustic ring-down. Here, we present experimental ex vivo and in vivo porcine small bowel imaging results from an USCE prototype. The preliminary results show that the near-field image can be recovered after post-processing. Although some limitations in imaging the GI tract are inherent in the use of focused single element transducers, solving the problem of near-field imaging is relevant to all USCE implementations.

In this paper, an image processing to detect polyps in an intelligent Wireless Capsule Endoscopy (WCE) is presented. This processing will be integrated into the WCE. It is a new screening method to detect colorectal cancer (CRC). A motion estimation algorithm is used to follow a detected polyp and improve the pre-processing of our detection chain. With our methodology, the polyp detection rate is improved by up to 40% from 53% to 93.7%. The improved detection rate was validated with a large database of 20 video-colonoscopies (18,910 images).

Wireless capsule endoscopy (WCE) allows medical doctors to examine the interior of the small intestine with a noninvasive procedure. This methodology is particularly important for Crohn's disease (CD), where an early diagnosis improves treatment outcomes. However, the viewing and evaluation of WCE videos is a time-consuming process for the medical experts. In this work, we present a recurrent attention neural network for the detection in WCE images of CD lesions in the small bowel. Our classifier reaches 90.85% accuracy on our own dataset annotated by experts from the Hospital of Nantes. The model has also been tested on a public endoscopic dataset, the CAD-CAP database used for the GIANA competition, and achieves high performance on detection task with an accuracy of 99.67%. This automatic lesion classifier will greatly reduce the amount of time spent by gastroenterologists in reviewing WCE videos, which will likely foster the development of this technique and speed-up the diagnosis of CD.

Large population of the world suffers from diseases related to gastrointestinal (GI) tract. The most modern technique available to scan the GI tract is capsule endoscopy (CE). It is a non-sedative, non-invasive and patient-friendly alternative to conventional endoscopy. However, CE generates approximately 55000 to 60000 images which make the diagnosis process time consuming and tiresome for physicians. Also the diagnosis varies from expert to expert. Hence a computer-aided diagnosis system is a must. This study, addresses a multi-class medical image analysis problem using image processing and machine learning techniques. It presents a computer aided diagnosis (CAD) system based on conventional machine learning as well as deep learning for automatic abnormality detection in GI tract. The system performs with an accuracy of 95.11% and precision of 93.9%.

Wireless capsule endoscopy is an advanced application of using ICT in healthcare and medicine. It is a non-invasive method to examine the gastrointestinal (GI) for early diagnosis and treatment. The capsule is equipped with a camera, light source, processing, control unit, and a transmitter to send the captured pictures or videos wirelessly to an external receiver system. The capsule is used to monitor small bowel diseases, esophageal diseases, and colonic diseases. The current data rate of the capsule endoscopes is limited to 1–2 Mbps with 2–3 frames per second and significant compression rates are applied to provide an operation time at least 8 hours inside the body. The data and frame rates might be sufficient for a capsule without locomotion, in which the capsule travels slowly, and the gastric wall movements mainly manage the speed. However, the next generation of the capsules will need remote locomotion and orientation control for fast and precise diagnosis. This application requests high data rate and low image or video compression rates with high frame rates for real-time navigation and control. Such a communication system becomes power hungry and limits the longevity of the device. Design of an ultra-low power wireless transmission scheme can assist the capsule device to save energy to handle the technology requirements. In this paper, we assess the communication systems of the capsule endoscopes and propose a battery-free high data rate backscatter wireless transmission scheme. The proposed system has been tested inside biological phantom in order to measure the path loss over different capsule depths. The average attenuation in a backscatter link inside biological phantom is 4.1 dB/cm in which is a linear function of capsule depth. According to link budget calculations, by using a 20 dBm transmitter at the outside of the body we can receive 8 Mbps data from the capsule at 10 cm depth. Our proposed system demonstrates superior performance compared to state-of-the-art systems in the literature.

In the case of modern diagnosis technology, WCE (Wireless Capsule Endoscopy) is one of the most efficient and effective technologies for the diagnosis of Gastrointestinal tract noninvasively such as bleeding in intestine areas. Since the duration of the WCE video is so long, it contains a large number of images. Consequently, it makes a huge burden for the physicians to recognize affected portion in the real required time and it is the main obstacle for using the WCE in the wider application. For releasing the extra burden of the clinicians, we have proposed a novel automatic computer-aided system for detecting bleeding in the GI tract. In our proposed method, we have applied some special preprocessing for obtaining the required informative portions of blood. A noble Weighted k-Nearest-Neighbor (WKNN) classifier has applied for separating distinct object with some statistical features in HSV color space. We have successfully achieved accuracy 98.8%, sensitivity 99%, specificity 99%, precision 95%, negative predicted value 99% and F1 score 97% after extensive experiment which will outperform some of the existing researches. More importantly, its computational time is very compared to others.

Wireless capsule endoscopy (WCE) becoming a very useful device for detecting diseases in the human digestive tract. Using WCE Physicians diagnose the different abnormalities such as ulcer, bleeding, polyps, chronic diarrhea, small intestinal cancer/tumor and Crohn's disease in the digestive tract in invasive way. Researchers are working for the development and improvement of the performance of WCE using software to detect these diseases at a messive rate of advancement automatically. This paper undergoes an advance methodology for automatically discovering ulcers in the image converted from WCE video footage. The method applied for numerous frames of capsule endoscopic images with logistic regression classifier in HSV color space and found effectiveness for the physicians. We obtained the result of 87.70% accuracy and 94.00% sensitivity which provides the evidence of the accuracy of this automated computer aided system.

In this paper, we propose a scheme to classify different Wireless Capsule Endoscopy (WCE) lesion images for diagnosis. The main contribution is to quantify multi-scale pooled channel-wise information and merge multi-level features together by explicitly modeling interdependencies between all feature maps of different convolution layers. Firstly, feature maps are resized into multi-scale size with bicubic interpolation, and then down-sampling convolution method is adopted to obtain pooled feature maps of the same resolution, and finally one by one convolution kernels are utilized to fuse feature maps after quantization operation based on channel-wise attention mechanism in order to enhance feature extraction of the proposed architecture. Preliminary experimental result shows that our proposed scheme with less model parameters achieves competitive results compared to the state-of-the-art methods in WCE image classification task.

Wireless capsule endoscopy (WCE) is a technology used to record colored internal images of the gastrointestinal (GI) tract for the purpose of medical diagnosis. It transmits a large number of frames in a single examination cycle, which makes the process of analyzing and diagnosis of abnormalities extremely challenging and time-consuming. In this paper, we propose a technique to automate the abnormality detection in WCE images following a deep learning approach. The WCE images are split into patches and input to a convolutional neural network (CNN). A trained deep neural network is used to classify patches to be either malign or benign. The patches with abnormalities are marked on the WCE image output. We obtained an area under receiver-operating-characteristic curve (AUROC) value of about 98.65% on a publicly available test data containing nine abnormalities.

Wireless Capsule Endoscopy (WCE) has been widely used for the screening of Gastrointestinal (GI) diseases. However, manually reviewing the huge number of WCE images is time-consuming and error-prone, thus a computer-aided diagnosis (CAD) system is highly demanded in clinical practice. In this paper, we propose a novel Saliency-aware Hybrid Network (SHNet) for automatic GI bleeding detection. Specifically, the SHNet consists of two densely connected convolutional networks (DenseNets) that are respectively considered as global image stream and saliency-aware stream. Moreover, we introduce polar transformation to reduce the noise from the background and highlight the image information. Finally, we employ the ensemble learning strategy to jointly optimize the final diagnosis result. Extensive experiments on the clinical WCE dataset illustrate that our proposed method outperforms the-state-of-the-art algorithms in GI bleeding detection with the F1 score of 0.959.

In this paper, we firstly propose a novel method - Relative Localization, which is a meaningful technique for the localization of wireless capsule endoscopy in human body. The first implementation of relative localization system is presented with a completely new design, the free sensor array. The free sensor array and the established coordinate system follow the patient's belly. The errors of sensor position, sensor orientation and object relative distance increment are only 4.171mm, 2.065° and 1.464mm, respectively. The analysis of the experimental results can be used as guidance for the medical staff.

This paper proposes a miniaturized Ultra-Wide Band (UWB) ingestible capsule antenna used for wireless capsule endoscopy (WCE). The antenna is a compact planar slotted microstrip patch antenna with a modified partial ground plane. Rogers TMM 13i is used as the substrate material of the antenna. The approximate environment of the whole human gastrointestinal (GI) tract with surrounding tissues is designed with CST Microwave Studio and the performance of the antenna is investigated by placing it in the centre of the environment. The size of the proposed antenna is 90 mm3(10 mm × 10 mm × 0.9 mm), resonant frequency is about 4.5 GHz with a −10 dB bandwidth of about 2.9 GHz ranging from 3.1 GHz to 6 GHz, and the radiation pattern is omnidirectional with circular polarization in the biological model. It is observed that the maximum radiation efficiency is −28.39 dB. The SAR value in GI tract is also determined for biocompatibility analysis.

Wireless capsule endoscopy (WCE) is quite an advanced, patient-friendly and novel medical equipment for non-invasive gastrointestinal disease detection, which is able to view the entire gastrointestinal tract without pain. Since there are more than 80,000 WCE images for each examination, it usually takes hours for professional clinicians to diagnose all the video data. Therefore, the automatic computer-aided lesion classification technique is highly perspective needed. In this paper, we propose an effective scheme to classify different lesion images acquired by WCE. Firstly, we obtain feature maps of the same resolution by performing a max pooling operation on different convolutional layers, and then quantify the pooled feature maps by trainable weight parameters, and finally one by one convolution kernels are employed to merge the combined quantized feature maps. We enhance the performance of feature extraction by merging multi-level convolutional features, including both low level and high level features. The preliminary experimental result shows that the classification accuracy rate is up to 95.15%, both running speed and recognition rate are higher than traditional machine learning algorithms.

Technology development enables progress in numerous areas and one of the relatively recent examples is wireless capsule endoscopy. It is used for detailed examination of a digestive track. Capsule size camera is swallowed by a patient and it takes thousands of images during the travel through digestive tract. The obtained images are used to detect different anomalies such as bleedings. In this paper we propose a method for automatic bleeding detection in capsule endoscopy images based on color and texture features. The proposed method is region based and it uses HSI and CIE Lab color spaces along with uniform local binary pattern for describing each region. Based on these features, regions are classified by support vector machine into three groups: background, bleeding or non-bleeding region. The proposed method was tested on benchmark dataset and the results were compared with other state-of-the-art method. Our proposed method shows competitive results based on the Dice similarity coefficient and misclassification error.

Video capsule endoscopy (VCE) was originally developed to address the limitation of conventional endoscopy in accessing the small bowel as a remote part of the gastrointestinal tract. To further enhance the diagnostic ability of VCE, microultrasound capsule endoscopy is under development for identification of disease at an earlier stage and visualisation of subsurface tissue features. This paper presents an evaluation of two approaches to improve signal to noise ratio (SNR) in rapid prototyped capsule endoscopes. First, noise reduction techniques are applied to the integrated front-end circuits in the prototype capsules. Secondly, multiple types of coded excitation transmission are tested and benchmarked with respect to non-coded transmission. Results are presented for both bench top phantom imaging and in vivo translational trial imaging.

Wireless Capsule Endoscopy (WCE) has become increasingly popular in clinical gastrointestinal (GI) disease diagnosis, benefiting from its painless and noninvasive examination. However, reviewing a large number of images is time-consuming for doctors, thus a computer-aided diagnosis (CAD) system is in high demand. In this paper, we present an automatic bleeding detection algorithm that consists of three stages. The first stage is the preprocessing, including key frame extraction and edge removal. In the second stage, we discriminate the bleeding frames using a novel superpixelcolor histogram (SPCH) feature based on the principle color spectrum, and then the decision is made by a subspace KNN classifier. Thirdly, we further segment the bleeding regions by extracting a 9-D color feature vector from the multiple color spaces at the superpixel level. Experimental results with an accuracy of 0.9922 illustrate that our proposed method outperforms the state-of-the-art methods in GI bleeding detection with low computational costs.

Wireless capsule endoscopy is widely used as an efficient way to receive images of gastrointestinal tract for medical diagnostics. Due to its energy compaction property for correlated image pixels, DCT) is highly preferred in image compression for this real time application. To accommodate the tight power budget and small size condition in the wireless capsule endoscopy application, an energy efficient DCT based on CMOS implementation is proposed in this paper. Three approximating methods are combined and applied in a multiplier-less DCT architecture. The simulation results indicate that the proposed DCT reduces 53.85% energy and 25.32% area with a PSNR penalty of 1.9dB when compared with the integer DCT. Compared with CORDIC-Loeffler DCT, it achieves 19.38% reduction on energy dissipation and 14.4dB more PSNR at the cost of only 1.39% area overhead.

Recently, external electromagnetic actuation(EMA) system was introduced to control the locomotion of the capsule endoscopy(CE) using magnetic force. EMA system provides the manual user interface to control the system, but inspectors suffered from fatigues due to the long examination time. We proposed an autonomous driving algorithm for the capsule endoscope. The algorithm searched the target point based on the image processing where the capsule should orient, and the steering was automatically manipulated until the capsule oriented the target point. Then, the propulsion was made until the capsule deviated from the target point. In order to verify the feasibility of the algorithm, simulated endoscopic images were acquired from the commercially available endoscopic capsule by using intestine phantom and a linear and rotation motion stage. The driving simulator was tested on the arc-shaped paths having the various curvatures under the various propulsion forces. In the most conditions, the proposed algorithm succeeded in driving autonomously in the given paths. In some conditions, having a large curvature and a large propulsion, the target point was missed, but scanning algorithm for the missed target point may overcome this problem. In conclusion, the proposed algorithm could be utilized in the active capsule endoscope system and provide the autonomous driving mode in the capsule endoscopy without additional sensors or devices.