Abstract: In the domain of medical endoscopy an increasing number of surgeons nowadays store video recordings of their interventions in a huge video archive. Among some other purposes, the videos are used for post-hoc surgical quality assessment, since objective assessment of surgical procedures has been identified as essential component for improvement of surgical quality. Currently, such assessment is performed manually and for selected procedures only, since the amount of data and cumbersome interaction is very time-consuming. In the future, quality assessment should be carried out comprehensively and systematically by means of automated assessment algorithms. In this demo paper, we present a tool that supports human assessors in collecting manual annotations and therefore should help them to deal with the huge amount of visual data more efficiently. These annotations will be analyzed and used as training data in the future.

Abstract: Given a large-scale video archive of surgical interventions and a medical image showing a specific moment of an operation, how to find the most image-related videos efficiently without the utilization of additional semantic characteristics? In this paper, we investigate a novel content-based approach of linking medical images with relevant video segments arising from endoscopic procedures. We propose to approximate the video segments' content-based features by gradient-based signatures and to index these signatures with the Minkowski distance in order to determine the most query-like video segments efficiently. We benchmark our approach on a large endoscopic image and video archive and show that our approach achieves a significant improvement in efficiency in comparison to the state-of-the-art while maintaining high accuracy.

Abstract: Content-based analysis and retrieval of digital images found in scientific articles is often hindered by images consisting of multiple subfigures (compound figures). We address this problem by proposing a method (ComFig) to automatically classify and separate compound figures, which consists of two main steps: (i) a supervised compound figure classifier (ComFig classifier) discriminates between compound and non-compound figures using task-specific image features; and (ii) an image processing algorithm is applied to predicted compound images to perform compound figure separation (ComFig separation). The proposed ComFig classifier is shown to achieve state-of-the-art classification performance on a published dataset. Our ComFig separation algorithm shows superior separation accuracy on two different datasets compared to other known automatic approaches. Finally, we propose a method to evaluate the effectiveness of the ComFig chain combining classifier and separation algorithm, and use it to optimize the misclassification loss of the ComFig classifier for maximal effectiveness in the chain.

Abstract: We propose an image processing algorithm to automatically separate compound figures appearing in scientific articles. We classify compound images into two classes and apply different algorithms for detecting vertical and horizontal separators to each class: the edge-based algorithm aims at detecting visible edges between subfigures, whereas the band-based algorithm tries to detect whitespace separating subfigures (separator bands). The proposed algorithm has been evaluated on two datasets for compound figure separation (CFS) in the biomedical domain and compares well to semi-automatic or more comprehensive state-of-the-art approaches. Additional experiments investigate CFS effectiveness and classification accuracy of various classifier implementations.

Abstract: Fitness- or exergames are ubiquitously available, but often lack the main ingredient of successfully gamified systems: fun. This can be attributed to the typical way of designing such games -- highly focusing on specific physical activities, thus, gamifying fitness. Instead, we propose a novel alternate approach to improve motivation for exergaming, which we call fitnessification: integrating physical exercise into very popular games that have been developed keeping fun in mind and frequently are played for long periods of time -- so-called AAA games. In order to evaluate this concept, we have conducted a comparative study examining voluntary participants' reactions to testing an ergometer controlled casual game as well as a modified AAA game. Results indicate strong tendencies of players preferring the newly introduced AAA approach over the casual fitness game.

Abstract: In this report we describe our approach to the fully automatic Ad-hoc video search task for TRECVID 2016. We describe how we obtain training data from the web, create according CNN models for the provided queries and use them to classify keyframes from a custom sub-shot detection method. The resulting classifications are fed into a Lucene index in order to obtain the shots that match the query. We also discuss our results and point out potentials for further improvements.

Abstract: Our approach to automatically separating compound figures appearing in biomedical articles is split into two image processing algorithms: one is based on detecting separator edges, and the other tries to identify background bands separating subgures. Only one algorithm is applied to a given image, according to the prediction of a binary classifier trained to distinguish graphical illustrations from other images in biomedical articles. Our submission to the ImageCLEF 2015 compound figure separation task achieved an accuracy of 49% on the provided test set of about 3400 compound images. This stays clearly behind the best submission of other participants (85% accuracy), but is by an order of magnitude faster than other approaches reported in the literature.

Abstract: In medical endoscopy more and more surgeons record videos of their interventions in a long-term storage archive for later retrieval. In order to allow content-based search in such endoscopic video archives, the video data needs to be indexed first. However, even the very basic step of content-based indexing, namely content segmentation, is already very challenging due to the special characteristics of such video data. Therefore, we propose to use instrument classification to enable semantic segmentation of laparoscopic videos. In this paper, we evaluate the performance of such an instrument classification approach. Our results show satisfying performance for all instruments used in our evaluation.

Abstract: Medical case retrieval (MCR) is information retrieval in a collection of medical case descriptions, where descriptions of patients' symptoms are used as queries. We apply known text retrieval techniques based on query and document expansion to this problem, and combine them with new algorithms to match queries and documents with Medical Subject Headings (MeSH). We ran comprehensive experiments to evaluate 546 method combinations on the ImageCLEF 2013 MCR dataset. Methods combining MeSH query expansion with pseudo-relevance feedback performed best, delivering retrieval performance comparable to or slightly better than the best MCR run submitted to ImageCLEF 2013.

Abstract: Our approach to the ImageCLEF medical case retrieval task consists of text-only retrieval combined with utilizing the Medical Subject Headings (MeSH) ontology. MeSH terms extracted from the query are used for query expansion or query term weighting. MeSH annotations of documents available from PubMed Central are added to the corpus. Retrieval results improve slightly upon full-text retrieval.

Abstract: In the medical domain it has become common to store recordings of endoscopic surgeries or procedures. The storage of these endoscopic videos provides not only evidence of the work of the surgeons but also facilitates research, the training of new surgeons and supports explanations to the patients. However, an endoscopic video archive, where tens or hundreds of new videos are added each day, needs content-based analysis in order to provide content-based search. A fundamental first step in content analysis is the segmentation of the video. We propose a new method for segmentation of endoscopic videos, based on spatial and temporal differences of motion in these videos. Through an evaluation with 20 videos we show that our approach provides reasonable performance.

Abstract: In recent years, it became common to record video footage of laparoscopic surgeries. This leads to large video archives that are very hard to manage. They often contain a considerable portion of completely irrelevant scenes which waste storage capacity and hamper an efficient retrieval of relevant scenes. In this paper we (1) define three classes of irrelevant segments, (2) propose visual feature extraction methods to obtain irrelevance indicators for each class and (3) present an extensible framework to detect irrelevant segments in laparoscopic videos. The framework includes a training component that learns a prediction model using nonlinear regression with a generalized logistic function and a segment composition algorithm that derives segment boundaries from the fuzzy frame classifications. The experimental results show that our method performs very good both for the classification of individual frames and the detection of segment boundaries in videos and enables considerable storage space savings.

Abstract: Videos of endoscopic procedures typically feature a circular content area in the image center. This area is surrounded by a dark border that carries no relevant information but is subject to noise. Thus, a considerable proportion of the available bitrate has to be wasted to encode the border regions. We propose to superimpose the border regions with a homogenous black mask so that it can be encoded efficiently with skipped macroblocks. To determine the exact position and size of the circular content area we use an efficient circle detection algorithm. Through an evaluation with 138 videos we show that the border overlay can significantly reduce the bitrate without degrading the visual quality of the content area.