A5075663234- Image Retrieval and Classification Techniques
- Semantic Web and Ontologies
- Neural Networks and Applications
- Fuzzy Logic and Control Systems
- Advanced Image and Video Retrieval Techniques
- Video Analysis and Summarization
- Information Retrieval and Search Behavior
- Natural Language Processing Techniques
- Music and Audio Processing
- Topic Modeling
- Recommender Systems and Techniques
- Advanced MRI Techniques and Applications
- Data Management and Algorithms
- Advanced Text Analysis Techniques
- Web Data Mining and Analysis
- Medical Imaging Techniques and Applications
- Medical Image Segmentation Techniques
- Data Visualization and Analytics
- Advanced X-ray and CT Imaging
- Data Mining Algorithms and Applications
- Speech and dialogue systems
- Advanced Neural Network Applications
- Music Technology and Sound Studies
- Advanced Clustering Algorithms Research
- Brain Tumor Detection and Classification
Otto-von-Guericke University Magdeburg
2016-2025
Center for Behavioral Brain Sciences
2021-2024
University Hospital Magdeburg
2004-2022
University of California, Berkeley
2002-2014
National Institute of Informatics
2014
ATA Engineering (United States)
2011
BT Research
2004
The enormous amount of information stored in unstructured texts cannot simply be used for further processing by computers, which typically handle text as simple sequences character strings.Therefore, specific (pre-)processing methods and algorithms are required order to extract useful patterns.Text mining refers generally the process extracting interesting knowledge from text.In this article, we discuss a young interdisciplinary field intersection related areas retrieval, machine learning,...
Deep learning methods have become the de-facto standard for challenging image processing tasks such as classification. One major hurdle of deep approaches is that large sets labeled data are necessary, which can be prohibitively costly to obtain, particularly in medical diagnosis applications. Active techniques alleviate this labeling effort. In paper we investigate some recently proposed active with high-dimensional and convolutional neural network classifiers. We compare ensemble-based...
A brain tumour is a mass or cluster of abnormal cells in the brain, which has possibility becoming life-threatening because its ability to invade neighbouring tissues and also form metastases. An accurate diagnosis essential for successful treatment planning magnetic resonance imaging principal modality diagnostic tumours their extent. Deep Learning methods computer vision applications have shown significant improvement recent years, most can be credited fact that sizeable amount data...
The outbreak of COVID-19 has shocked the entire world with its fairly rapid spread, and challenged different sectors. One most effective ways to limit spread is early accurate diagnosing infected patients. Medical imaging, such as X-ray computed tomography (CT), combined potential artificial intelligence (AI), plays an essential role in supporting medical personnel diagnosis process. Thus, this article, five deep learning models (ResNet18, ResNet34, InceptionV3, InceptionResNetV2,...
Offline evaluations are the most common evaluation method for research paper recommender systems. However, no thorough discussion on appropriateness of offline has taken place, despite some voiced criticism. We conducted a study in which we evaluated various recommendation approaches with both and online evaluations. found that results often contradict each other. discuss this finding detail conclude may be inappropriate evaluating systems, many settings.
Over 80 approaches for academic literature recommendation exist today. The were introduced and evaluated in more than 170 research articles, as well patents, presentations blogs. We reviewed these found most evaluations to contain major shortcomings. Of the proposed, 21% not evaluated. Among approaches, 19% against a baseline. user studies performed, 60% had 15 or fewer participants did report on number of participants. Information runtime coverage was rarely provided. Due several other...
In this demo paper we present Docear's research recommender system. Docear is an academic literature suite to search, organize, and create articles. The users' data (papers, references, annotations, etc.) managed in mind maps these are utilized for the recommendations. Using content-based filtering methods, achieves click-through rates around 6%, some scenarios even over 10%.
While eye tracking is becoming more and relevant as a promising input channel, diverse applications using gaze control in natural way are still rather limited. Though several researchers have indicated the particular high potential of gaze-based interaction for pointing tasks, often gaze-only approaches investigated. However, time-consuming dwell-time activations limit this potential. To overcome this, we present gaze-supported fisheye lens combination with (1) keyboard (2) tilt-sensitive...
Blood vessels of the brain provide human with required nutrients and oxygen. As a vulnerable part cerebral blood supply, pathology small can cause serious problems such as Cerebral Small Vessel Diseases (CSVD). It has also been shown that CSVD is related to neurodegeneration, Alzheimer's disease. With advancement 7 Tesla MRI systems, higher spatial image resolution be achieved, enabling depiction very in brain. Non-Deep Learning-based approaches for vessel segmentation, e.g., Frangi's...
Abstract Objectives The shape is commonly used to describe the objects. State-of-the-art algorithms in medical imaging are predominantly diverging from computer vision, where voxel grids, meshes, point clouds, and implicit surface models used. This seen growing popularity of ShapeNet (51,300 models) Princeton ModelNet (127,915 models). However, a large collection anatomical shapes (e.g., bones, organs, vessels) 3D surgical instruments missing. Methods We present MedShapeNet translate...
We present a language-independent spell-checker that is based on an enhancement of the n-gram model. The spell checker proposing correction suggestions by selecting most promising candidates from ranked list derived statistics and lexical resources. Besides motivating describing developed techniques, we briefly discuss use proposed approach in application for keyword- semantic-based search support. In addition, tool was compared with state-of-the-art spelling approaches. evaluation showed it...
In the past few years, we have developed a research paper recommender system for our reference management software Docear.In this paper, introduce architecture of and four datasets.The comprises multiple components, e.g. crawling PDFs, generating user models, calculating content-based recommendations.It supports researchers developers in building their own systems, is, to best knowledge, most comprehensive that has been released field.The datasets contain metadata 9.4 million academic...
Previous chapter Next Full AccessProceedings Proceedings of the 2008 SIAM International Conference on Data Mining (SDM)Creating a Cluster Hierarchy under Constraints Partially Known HierarchyKorinna Bade and Andreas NürnbergerKorinna Nürnbergerpp.13 - 24Chapter DOI:https://doi.org/10.1137/1.9781611972788.2PDFBibTexSections ToolsAdd to favoritesExport CitationTrack CitationsEmail SectionsAboutAbstract Although clustering constraints is current research topic, hierarchical setting, in which...
We present first results of a logfile analysis on web search engines for children. The aim this research is to analyse fundamental facts about how children's behaviour differs from that adults. show differences previous results, which are often based small lab experiments. Our large-scale suggests children queries more information-oriented and shorter average. Children indeed make lot spelling errors repeat searches revisit pages.