Clinical score predicting a successful endoscopic approach of esophageal anastomotic leaks: external validation.

BACKGROUND: Recently, a clinical prediction rule has been proposed to predict the chance of successful endoscopic stenting in benign esophageal anastomotic leakage, perforation and fistula. We aimed to validate this score in a cohort of patients with anastomotic leaks managed with self-expanding metal esophageal stents, by assessing technical and clinical success rates and comparing the agreement between the predicted and the actual clinical success. METHODS: A multicenter retrospective cohort study including patients submitted to endoscopic stenting due to anastomotic leak was conducted. Variables of the score (leak size, location and C-reactive protein) were collected and the chance of success (≤50, 50-70 and ≥70%) and its accuracy was assessed. RESULTS: Fifty-three patients, submitted to esophageal stenting after cancer (n = 47) and bariatric surgery were included. Clinical success was achieved in 62% of patients. The area under the ROC curve to differentiate between successful and failed therapies showed a good discriminative power of the score (AUC 0.705; P < 0.01). For a predicted chance of success >50%, the positive predictive value was 72.5%; for a chance of success ≤50%, the negative predictive value was 69.2%. CONCLUSIONS: The application of this predictive model in patients with anastomotic leaks proved to be valid in a different cohort from that in which it was derived. Its usefulness in clinical practice may be anticipated, favoring stenting in patients with a chance of success >50%. However, we must be cautious in patients with a lower probability of success and a case-by-case decision should be made.

Integral scale histogram local binary patterns for classification of narrow-band gastroenterology images.

The introduction of various novel imaging technologies such as narrow-band imaging have posed novel image processing challenges to the design of computer assisted decision systems. In this paper, we propose an image descriptor referred to as integrated scale histogram local binary patterns. We propagate an aggregated histogram of local binary patterns of an image at various resolutions. This results in low dimensional feature vectors for the images while incorporating their multiresolution analysis. The descriptor was used to classify gastroenterology images into four distinct groups. Results produced by the proposed descriptor exhibit around 92% accuracy for classification of gastroenteroloy images outperforming other state-of-the-art methods, endorsing the effectiveness of the proposed descriptor.

Impact of visual features on the segmentation of gastroenterology images using normalized cuts.

Gastroenterology imaging is an essential tool to detect gastrointestinal cancer in patients. Computer-assisted diagnosis is desirable to help us improve the reliability of this detection. However, traditional computer vision methodologies, mainly segmentation, do not translate well to the specific visual characteristics of a gastroenterology imaging scenario. In this paper, we propose a novel method for the segmentation of gastroenterology images from two distinct imaging modalities and organs: chromoendoscopy (CH) and narrow-band imaging (NBI) from stomach and esophagus, respectively. We have used various visual features individually and their combinations (edgemaps, creaseness, and color) in normalized cuts image segmentation framework to segment ground truth datasets of 142 CH and 224 NBI images. Experiments show that an integration of edgemaps and creaseness in normalized cuts gives the best segmentation performance resulting in high-quality segmentations of the gastroenterology images.

Invariant Gabor texture descriptors for classification of gastroenterology images.

Automatic classification of lesions for gastroenterology imaging scenarios poses novel challenges to computer-assisted decision systems, which are mostly attributed to the dynamics of the image acquisition conditions. Such challenges demand that automatic systems are able to give robust characterizations of tissues irrespective of camera rotation, zoom, and illumination gradients when viewing the inner surface of the gastrointestinal tract. In this paper, we study the invariance properties of Gabor filters and propose a novel descriptor, the autocorrelation Gabor features (AGF). We show that our proposed AGF is invariant to scale, rotation, and illumination changes in the images. We integrate these new features in a texton framework (Texton-AGF) to classify images from two complementary gastroenterology imaging scenarios (chromoendoscopy and narrow-band imaging) broadly into three different groups: normal, precancerous, and cancerous. Results show that they compare favorably to using state-of-the-art texture descriptors for both imaging modalities.

Quantitative comparison of segmentation methods for in-body images.

In this paper, we present a numerical comparison of how well segmentation algorithms approximate the manual segmentation of gastroenterologists for a set of endoscopic images. Different areas in these images demand different levels of analysis by a clinician and some provide critical information about the patient. Our objective is thus to segment endoscopic images so that the results mimic as closely as possible the areas that were considered relevant by doctors. We focus on a detailed quantitative comparison of two popular segmentation algorithms, mean shift and normalized cuts, when applied to in-body images, most specifically for vital-stained magnification endoscopy. Segmentation results are compared with the manual annotations of the same images performed by two specialist clinicians. Results show that if we simply consider the most relevant segmented patch, normalized cuts performs better. However, if we allow the annotated area to be represented by multiple patches, mean shift is clearly a better choice, although automatic ways to determine its kernel's bandwidth are highly desirable.