Michael Häfner

Successful antibiotic treatment of Helicobacter pylori negative gastric mucosa associated lymphoid tissue lymphomas

Background and aims: The role of antibiotic treatment in early stage gastric mucosa associated lymphoid tissue (MALT) lymphoma not associated with Helicobacter pylori infection has not been investigated. Patients and methods: Six patients with localised gastric MALT lymphoma underwent antibiotic treatment with clarithromycin, metronidazole, and pantoprazole. Staging, including endosonography plus gastroscopy, computed tomography of the thorax and abdomen, colonoscopy, magnetic resonance imaging of the salivary glands, and bone marrow biopsy were performed to rule out distant spread of the disease. In addition, MALT specific genetic changes, including reverse transcriptase-polymerase chain reaction for t(11;18)(q21;q21), were tested in all patients. H pylori infection was ruled out by histology, urease breath test, serology, and stool antigen testing. Results: All six patients had MALT lymphoma restricted to the stomach, and no evidence of infection with H pylori was found. Only one patient tested positive for t(11;18)(q21;q21) while the remaining five displayed no genetic aberrations. Following antibiotic treatment, endoscopic controls were performed every three months. Five patients responded with lymphoma regression between three and nine months following antibiotic treatment (one partial remission and four complete responses). One patient had stable disease for 12 months and was then referred for chemotherapy. Conclusions: Patients with early stage gastric MALT lymphoma negative for H pylori might still benefit from antibiotic treatment as the sole treatment modality.

Color treatment in endoscopic image classification using multi-scale local color vector patterns

Graphical abstract In this work we propose a novel multi-scale operator which is based on the full color information within an image. In order to evaluate the method, we extract features from endoscopic images using this operator and classify the images according to the respective class of polyps. Highlights ► Compared to other LBP-based operators LCVP uses all color information available, yet yielding a more compact descriptor for an image. ► LCVP is up to 7.5 times faster compared to other LBP-based methods evaluated. ► In terms of a classification of polyps the accuracy of LCVP differs insignificantly only from previously developed methods.

Austrian consensus on the definition and treatment of portal hypertension and its complications (Billroth II)

SummaryIn November 2004, the Austrian Society of Gastroenterology and Hepatology (ÖGGH) held for the first time a consensus meeting on the definitions and treatment of portal hypertension and its complications in the Billroth-Haus in Vienna, Austria (Billroth I-Meeting). This meeting was preceded by a meeting of international experts on portal hypertension with some of the proponents of the Baveno consensus conferences (http://www.oeggh.at/videos.asp). The consensus itself is based on the Baveno III consensus with regard to portal hypertensive bleeding and the suggestions of the International Ascites Club regarding the treatment of ascites. Those statements were modified by new knowledge derived from the recent literature and also by the current practice of medicine as agreed upon by the participants of the consensus meeting. In October 2011, the ÖGGH organized the second consensus meeting on portal hypertension and its complications in Vienna (Billroth II-Meeting). The Billroth II-Guidelines on the definitions and treatment of portal hypertension and its complications take into account the developments of the last 7 years, including the Baveno-V update and several key publications.ZusammenfassungIm November 2004 hielt die Österreichische Gesellschaft für Gastroenterologie und Hepatologie (ÖGGH) den ersten Konsensus über die Definitionen und die Therapie der Portalen Hypertension und ihrer Komplikationen im Billroth-Haus in Wien, Österreich ab (Billroth I Meeting). Diesem Treffen ging ein internationales Expertenmeeting über die Portale Hypertension mit einigen wichtigen Proponenten der Baveno Konsensus-Konferenzen vorraus (http://www.oeggh.at/videos.asp). Der Konsensus selber basiert auf dem Baveno III Konsensus im Hinblick auf die portal-hypertensive Blutung und den Vorschlägen des International Ascites Club in Hinblick auf die Therapie des Aszites. Deren Aussagen wurden mit neuen Erkenntnissen aus der rezenten Literatur und auch entsprechend der praktischen Erfahrung der Teilnehmer des Konsensus-Treffens modifiziert. Im Oktober 2001 organisierte die ÖGGH das zweite Konsensus Treffen über die portale Hypertension und ihrer Komplikationen (Billroth II Meeting). Die Billroth II Leitlinien über die Definitionen und die Therapie der Portalen Hypertension und ihrer Komplikationen lassen die Entwicklungen der letzten 7 Jahre inklusive des Baveno V Updates und etlicher Schlüsselpublikationen mit einfließen und stellen den neuen Standard im Management der Portalen Hypertension in Österreich dar.

Computer-assisted pit-pattern classification in different wavelet domains for supporting dignity assessment of colonic polyps

In this paper, we show that zoom-endoscopy images can be well classified according to the pit-pattern classification scheme by using texture-analysis methods in different wavelet domains. We base our approach on three different variants of the wavelet transform and propose that the color channels of the RGB and LAB color model are an important source for computing image features with high discriminative power. Color-channel information is incorporated by either using simple feature vector concatenation and cross-cooccurrence matrices in the wavelet domain. Our experimental results based on k-nearest neighbor classification and forward feature selection exemplify the advantages of the different wavelet transforms and show that color-image analysis is superior to grayscale-image analysis regarding our medical image classification problem.

Feature extraction from multi-directional multi-resolution image transformations for the classification of zoom-endoscopy images

In this article, we discuss the discriminative power of a set of image features, extracted from detail subbands of the Gabor wavelet transform and the dual-tree complex wavelet transform for the purpose of computer-assisted zoom-endoscopy image classification. We incorporate color channel information into the classification process and show that this leads to superior classification results, compared to luminance-channel-only-based image analysis.

Exploring Deep Learning and Transfer Learning for Colonic Polyp Classification

Recently, Deep Learning, especially through Convolutional Neural Networks (CNNs) has been widely used to enable the extraction of highly representative features. This is done among the network layers by filtering, selecting, and using these features in the last fully connected layers for pattern classification. However, CNN training for automated endoscopic image classification still provides a challenge due to the lack of large and publicly available annotated databases. In this work we explore Deep Learning for the automated classification of colonic polyps using different configurations for training CNNs from scratch (or full training) and distinct architectures of pretrained CNNs tested on 8-HD-endoscopic image databases acquired using different modalities. We compare our results with some commonly used features for colonic polyp classification and the good results suggest that features learned by CNNs trained from scratch and the “off-the-shelf” CNNs features can be highly relevant for automated classification of colonic polyps. Moreover, we also show that the combination of classical features and “off-the-shelf” CNNs features can be a good approach to further improve the results.

Combining Gaussian Markov random fields with the discrete-wavelet transform for endoscopic image classification

In this work we present a method for automated classification of endoscopic images according to the pit pattern classification scheme. Images taken during colonoscopy are transformed to the wavelet domain using the pyramidal discrete wavelet transform. Then, Gaussian Markov random fields are used to extract features from the resulting wavelet coefficients. Finally, these features are used for a classification using the k-NN classifier and the Bayes classifier.

Pit pattern classification using extended Local Binary Patterns

In this work we present a method for automated classification of endoscopic images according to the pit pattern classification scheme. Images taken during colonoscopy are transformed using a modified version of the local binary patterns operator (LBP). Then, two-dimensional histograms based on the LBP data from different color channels are created. Finally, the classification is carried out by employing the nearest-neighbors (1-NN) classifier in conjunction with the Bhattacharyya distance metric. The experimental results show that the extended LBP operator delivers superior results and an automated classification of endoscopic images based on the pit pattern classification scheme is feasible.

Delaunay triangulation-based pit density estimation for the classification of polyps in high-magnification chromo-colonoscopy

Highlights ► Exploiting the visual nature of pit patterns on the colonic mucosa. ► Roughly four times faster compared to a previously developed approach. ► Significantly higher classification rates compared to our previous work. ► More robust against overfitting when compared to other methods.

Endoscopic Image Classification Using Edge-Based Features

We present a system for an automated colon cancer detection based on the pit pattern classification. In contrast to previous work we exploit the visual nature of the underlying classification scheme by extracting features based on detected edges. To focus on the most discriminative subset of features we use a greedy forward feature subset selection. The classification is then carried out using the k-nearest neighbors (k-NN) classifier. The results obtained are very promising and show that an automated classification of the given imagery is feasible by using the proposed method.