@article{BrandTroyaKrenzeretal.2022, author = {Brand, Markus and Troya, Joel and Krenzer, Adrian and Saßmannshausen, Zita and Zoller, Wolfram G. and Meining, Alexander and Lux, Thomas J. and Hann, Alexander}, title = {Development and evaluation of a deep learning model to improve the usability of polyp detection systems during interventions}, series = {United European Gastroenterology Journal}, volume = {10}, journal = {United European Gastroenterology Journal}, number = {5}, doi = {10.1002/ueg2.12235}, url = {http://nbn-resolving.de/urn:nbn:de:bvb:20-opus-312708}, pages = {477-484}, year = {2022}, abstract = {Background The efficiency of artificial intelligence as computer-aided detection (CADe) systems for colorectal polyps has been demonstrated in several randomized trials. However, CADe systems generate many distracting detections, especially during interventions such as polypectomies. Those distracting CADe detections are often induced by the introduction of snares or biopsy forceps as the systems have not been trained for such situations. In addition, there are a significant number of non-false but not relevant detections, since the polyp has already been previously detected. All these detections have the potential to disturb the examiner's work. Objectives Development and evaluation of a convolutional neuronal network that recognizes instruments in the endoscopic image, suppresses distracting CADe detections, and reliably detects endoscopic interventions. Methods A total of 580 different examination videos from 9 different centers using 4 different processor types were screened for instruments and represented the training dataset (519,856 images in total, 144,217 contained a visible instrument). The test dataset included 10 full-colonoscopy videos that were analyzed for the recognition of visible instruments and detections by a commercially available CADe system (GI Genius, Medtronic). Results The test dataset contained 153,623 images, 8.84\% of those presented visible instruments (12 interventions, 19 instruments used). The convolutional neuronal network reached an overall accuracy in the detection of visible instruments of 98.59\%. Sensitivity and specificity were 98.55\% and 98.92\%, respectively. A mean of 462.8 frames containing distracting CADe detections per colonoscopy were avoided using the convolutional neuronal network. This accounted for 95.6\% of all distracting CADe detections. Conclusions Detection of endoscopic instruments in colonoscopy using artificial intelligence technology is reliable and achieves high sensitivity and specificity. Accordingly, the new convolutional neuronal network could be used to reduce distracting CADe detections during endoscopic procedures. Thus, our study demonstrates the great potential of artificial intelligence technology beyond mucosal assessment.}, language = {en} } @article{VainioValtonenHeberetal.2013, author = {Vainio, Rami and Valtonen, Eino and Heber, Bernd and Malandraki, Olga E. and Papaioannou, Athanasios and Klein, Karl-Ludwig and Afanasiev, Alexander and Agueda, Neus and Aurass, Henry and Battarbee, Markus and Braune, Stephan and Dr{\"o}ge, Wolfgang and Ganse, Urs and Hamadache, Clarisse and Heynderickx, Daniel and Huttunen-Heikinmaa, Kalle and Kiener, J{\"u}rgen and Kilian, Patrick and Kopp, Andreas and Kouloumvakos, Athanasios and Maisala, Sami and Mishev, Alexander and Miteva, Rosita and Nindos, Alexander and Oittinen, Tero and Raukunen, Osku and Riihonen, Esa and Rodriguez-Gasen, Rosa and Saloniemi, Oskari and Sanahuja, Blai and Scherer, Renate and Spanier, Felix and Tatischeff, Vincent and Tziotziou, Kostas and Usoskin, Ilya G. and Vilmer, Nicole}, title = {The first SEPServer event catalogue similar to ~68-MeV solar proton events observed at 1 AU in 1996-2010}, series = {Journal of Space Weather and Space Climate}, volume = {3}, journal = {Journal of Space Weather and Space Climate}, number = {A12}, issn = {2115-7251}, doi = {10.1051/swsc/2013030}, url = {http://nbn-resolving.de/urn:nbn:de:bvb:20-opus-122847}, year = {2013}, abstract = {SEPServer is a three-year collaborative project funded by the seventh framework programme (FP7-SPACE) of the European Union. The objective of the project is to provide access to state-of-the-art observations and analysis tools for the scientific community on solar energetic particle (SEP) events and related electromagnetic (EM) emissions. The project will eventually lead to better understanding of the particle acceleration and transport processes at the Sun and in the inner heliosphere. These processes lead to SEP events that form one of the key elements of space weather. In this paper we present the first results from the systematic analysis work performed on the following datasets: SOHO/ERNE, SOHO/EPHIN, ACE/EPAM, Wind/WAVES and GOES X-rays. A catalogue of SEP events at 1 AU, with complete coverage over solar cycle 23, based on high-energy (similar to 68-MeV) protons from SOHO/ERNE and electron recordings of the events by SOHO/EPHIN and ACE/EPAM are presented. A total of 115 energetic particle events have been identified and analysed using velocity dispersion analysis (VDA) for protons and time-shifting analysis (TSA) for electrons and protons in order to infer the SEP release times at the Sun. EM observations during the times of the SEP event onset have been gathered and compared to the release time estimates of particles. Data from those events that occurred during the European day-time, i.e., those that also have observations from ground-based observatories included in SEPServer, are listed and a preliminary analysis of their associations is presented. We find that VDA results for protons can be a useful tool for the analysis of proton release times, but if the derived proton path length is out of a range of 1 AU < s less than or similar to 3 AU, the result of the analysis may be compromised, as indicated by the anti-correlation of the derived path length and release time delay from the associated X-ray flare. The average path length derived from VDA is about 1.9 times the nominal length of the spiral magnetic field line. This implies that the path length of first-arriving MeV to deka-MeV protons is affected by interplanetary scattering. TSA of near-relativistic electrons results in a release time that shows significant scatter with respect to the EM emissions but with a trend of being delayed more with increasing distance between the flare and the nominal footpoint of the Earth-connected field line.}, language = {en} } @article{MehnertKochSchulzetal.2012, author = {Mehnert, Anja and Koch, Uwe and Schulz, Holger and Wegscheider, Karl and Weis, Joachim and Faller, Hermann and Keller, Monika and Br{\"a}hler, Elmar and H{\"a}rter, Martin}, title = {Prevalence of mental disorders, psychosocial distress and need for psychosocial support in cancer patients - study protocol of an epidemiological multi-center study}, volume = {12}, number = {70}, doi = {10.1186/1471-244X-12-70}, url = {http://nbn-resolving.de/urn:nbn:de:bvb:20-opus-153296}, year = {2012}, abstract = {Background Empirical studies investigating the prevalence of mental disorders and psychological distress in cancer patients have gained increasing importance during recent years, particularly with the objective to develop and implement psychosocial interventions within the cancer care system. Primary purpose of this epidemiological cross-sectional multi-center study is to detect the 4-week-, 12-month-, and lifetime prevalence rates of comorbid mental disorders and to further assess psychological distress and psychosocial support needs in cancer patients across all major tumor entities within the in- and outpatient oncological health care and rehabilitation settings in Germany. Methods/Design In this multicenter, epidemiological cross-sectional study, cancer patients across all major tumor entities will be enrolled from acute care hospitals, outpatient cancer care facilities, and rehabilitation centers in five major study centers in Germany: Freiburg, Hamburg, Heidelberg, Leipzig and W{\"u}rzburg. A proportional stratified random sample based on the nationwide incidence of all cancer diagnoses in Germany is used. Patients are consecutively recruited in all centers. On the basis of a depression screener (PHQ-9) 50\% of the participants that score below the cutoff point of 9 and all patients scoring above are assessed using the Composite International Diagnostic Interview for Oncology (CIDI-O). In addition, all patients complete validated questionnaires measuring emotional distress, information and psychosocial support needs as well as quality of life. Discussion Epidemiological data on the prevalence of mental disorders and distress provide detailed and valid information for the estimation of the demands for the type and extent of psychosocial support interventions. The data will provide information about specific demographic, functional, cancer- and treatment-related risk factors for mental comorbidity and psychosocial distress, specific supportive care needs and use of psychosocial support offers.}, language = {en} }