Search

Methodology evaluation for decision support systems for health is a time-consuming task. To assess performance of polyp detection methods in colonoscopy videos, clinicians have to deal with the annotation of thousands of images. Current existing tools could be improved in terms of flexibility and ease of use. We introduce GTCreator, a flexible annotation tool for providing image and text annotations to image-based datasets. It keeps the main basic functionalities of other similar tools while extending other capabilities such as allowing multiple annotators to work simultaneously on the same task or enhanced dataset browsing and easy annotation transfer aiming to speed up annotation processes in large datasets. The comparison with other similar tools shows that GTCreator allows to obtain fast and precise annotation of image datasets, being the only one which offers full annotation editing and browsing capabilites. Our proposed annotation tool has been proven to be efficient for large image dataset annotation, as well as showing potential of use in other stages of method evaluation such as experimental setup or results analysis.

In this article, our concern is the early diagnosis of colorectal cancer from a computeraided detection point of view in order to help physicians in their diagnosis during the gold standard examination: optical video colonoscopy. Since many years, some methods and materials have been developed to reduce the polyp miss rate and to improve detection capabilities. Nevertheless, the real challenge lies in the real-time use of these methods. In this context, more precisely, we focus our attention on the hardware implementation of a previous method we recently introduced in the literature for real-time detection of colorectal polyps, lesions that may degenerate into cancer. This implementation is subject to three performance criteria: real-time processing capabilities, detection rate and necessary computational resources. Six different platforms were tested and compared. If we noticed that only workstation computers are able to perform the detection with a good tradeoff between the three aforementioned criteria, possibilities of architecture optimizations are also identified and discussed in order to achieve real-time performance on platforms with low available computational resources like Raspberry Pi for instance. This latter issue is of major importance for possible integration of the detection algorithm inside smallconnected object like videocapsule, a promising alternative to standard colonoscopy.

International audience; Colorectal cancer is the second cause of cancer death inUnited States: precursor lesions (polyps) detection is key for patientsurvival. Though colonoscopy is the gold standard screening tool, somepolyps are still missed. Several computational systems have been pro-posed but none of them are used in the clinical room mainly due to com-putational constraints. Besides, most of them are built over still framedatabases, decreasing their performance on video analysis due to the lackof output stability and not coping with associated variability on imagequality and polyp appearance. We propose a strategy to adapt thesemethods to video analysis by adding a spatio-temporal stability mod-ule and studying a combination of features to capture polyp appearancevariability. We validate our strategy, incorporated on a real-time detec-tion method, on a public video database. Resulting method detects allpolyps under real time constraints, increasing its performance due to ouradaptation strategy.

International audience; Skin cancer is one of the most diagnosed cancer in the world. This cancer is based on two kinds of lesions:carcinomas and melanomas. Both are related to the degeneration of nevi. In this abstract, we aim to propose a tool to help people and physicians to determine automatically if a nevuscould be dangerous or not. To be accessible to the general public, such system must be embeddable at thehighest possible level, like a smartphone application.

International audience; For the last 20 years, wireless videocapsule technology has triggered alot of interest in the gastroenterologist community for the non-invasive early detectionof various gastrointestinal pathologies (ulcers, Chrones disease, polyp detection,etc.). Nevertheless, in most of the European countries videocapsules are notyet considered as a systematic valid alternative to classic endoscopies and colonoscopies.Main reasons are in the existing technological limitations of videocapsulesthat are of two kinds: (i) A limited battery life-time (8 hours usually ensured bythe manufacturer) that does not allow a complete imaging of the gastro intestinaltract, and (ii) the limited performance of the device in terms of detection rate of particularstructures like polyps for instance which degenerations are at the origin ofcolorectal cancer. To overpass these limitations, main idea of our work is to developa generation of smart videocapsules that takes advantage of the constant progress inelectronics and most precisely in embedded signal processing tasks. In this Chapter,we give first a detailed overview of the most recent state-of-the-art related tovideocapsules from the technological perspective in order to clearly positioned ourwork among the existing products and on going projects. In a second time, we proposea synthetic recall of the Cyclope project in the framework of which we arestudying different strategies to improve the performance of current videocapsules inthe particular context of the early diagnosis of colorectal cancer (polyp detection).We then propose a particular focus on the design optimization of the proposed algorithms from an electronic perspective. Most precisely, we give concrete elementsand quantitative estimation (time processing, embedding performance, etc.) to showthat embedding of the signal processing IP inside the videocapsule is feasible consideringthe most recent FPGA-platform performance, and that such an integrationcan bring a positive balance in terms of energy consumption by drastically reducingthe amount of transmitted data.

In this paper a method to perform real-time detection of polyps in videocolonoscopy is introduced. Polyps are at the origins of colorectal cancer which is one of the deadliest diseases in the world. Many methods to improve detection of polyps have been proposed so far. However performance of these methods strongly depends on the available computational resources and, until now, are not able to perform real-time detection during a standard examination. The proposed method, based on active learning, is able to solve these issues. Most precisely, this approach allows us to detect approximately 90% of polyps on a freely available database introduced to the community in 2012, for a F2 score of 65%, and matches real-time constraint by making possible the analysis of a frame in only 0.023s (average value) on a standard computer not necessarily dedicated to that kind of application.

International audience; Colonoscopy is the gold standard for colon cancer screening though still some polyps are missed, thus preventing early disease detection and treatment. Several computational systems have been proposed to assist polyp detection during colonoscopy but so far without consistent evaluation. The lack of publicly available annotated databases has made it difficult to compare methods and to assess if they achieve performance levels acceptable for clinical use. The Automatic Polyp Detection sub-challenge, conducted as part of the Endoscopic Vision Challenge (http://endovis.grand-challenge.org) at the international conference on Medical Image Computing and Computer Assisted Intervention (MICCAI) in 2015, was an effort to address this need. In this paper, we report the results of this comparative evaluation of polyp detection methods, as well as describe additional experiments to further explore differences between methods. We define performance metrics and provide evaluation databases that allow comparison of multiple methodologies. Results show that convolutional neural networks (CNNs) are the state of the art. Nevertheless it is also demonstrated that combining different methodologies can lead to an improved overall performance.