A5002487909- Video Analysis and Summarization
- Colorectal Cancer Screening and Detection
- Peer-to-Peer Network Technologies
- Distributed systems and fault tolerance
- Advanced Data Storage Technologies
- Mobile Agent-Based Network Management
- Sports Performance and Training
- Cloud Computing and Resource Management
- Image Retrieval and Classification Techniques
- Radiomics and Machine Learning in Medical Imaging
- AI in cancer detection
- Distributed and Parallel Computing Systems
- Gastrointestinal Bleeding Diagnosis and Treatment
- Advanced Image and Video Retrieval Techniques
- Human Pose and Action Recognition
- Sports injuries and prevention
- Security and Verification in Computing
- Sports Analytics and Performance
- Anomaly Detection Techniques and Applications
- Cloud Data Security Solutions
- Advanced Vision and Imaging
- Caching and Content Delivery
- Advanced Neural Network Applications
- Music and Audio Processing
- Multimedia Communication and Technology
UiT The Arctic University of Norway
2016-2025
Centre for Arctic Gas Hydrate, Environment and Climate
2014-2024
University of Bergen
2020
Høyskolen Kristiania
2020
University of California, Irvine
2020
Simula Metropolitan Center for Digital Engineering
2020
University of Oslo
2018
Aalborg University
2011
Cornell University
1994-2010
Accurate computer-aided polyp detection and segmentation during colonoscopy examinations can help endoscopists resect abnormal tissue thereby decrease chances of polyps growing into cancer. Towards developing a fully automated model for pixel-wise segmentation, we propose ResUNet++, which is an improved ResUNet architecture colonoscopic image segmentation. Our experimental evaluations show that the suggested produces good results on publicly available datasets. Furthermore, ResUNet++...
Semantic image segmentation is the process of labeling each pixel an with its corresponding class. An encoder-decoder based approach, like U-Net and variants, a popular strategy for solving medical tasks. To improve performance on various tasks, we propose novel architecture called DoubleU-Net, which combination two architectures stacked top other. The first uses pre-trained VGG-19 as encoder, has already learned features from ImageNet can be transferred to another task easily. capture more...
Automatic detection of diseases by use computers is an important, but still unexplored field research. Such innovations may improve medical practice and refine health care systems all over the world. However, datasets containing images are hardly available, making reproducibility comparison approaches almost impossible. In this paper, we present KVASIR, a dataset from inside gastrointestinal (GI) tract. The collection classified into three important anatomical landmarks clinically...
Abstract Artificial intelligence is currently a hot topic in medicine. However, medical data often sparse and hard to obtain due legal restrictions lack of personnel for the cumbersome tedious process manually label training data. These constraints make it difficult develop systems automatic analysis, like detecting disease or other lesions. In this respect, article presents HyperKvasir , largest image video dataset gastrointestinal tract available today. The collected during real gastro-...
Computer-aided detection, localisation, and segmentation methods can help improve colonoscopy procedures. Even though many have been built to tackle automatic detection of polyps, benchmarking state-of-the-art still remains an open problem. This is due the increasing number researched computer vision that be applied polyp datasets. Benchmarking novel provide a direction development automated tasks. Furthermore, it ensures produced results in community are reproducible fair comparison...
Abstract Artificial intelligence (AI) is predicted to have profound effects on the future of video capsule endoscopy (VCE) technology. The potential lies in improving anomaly detection while reducing manual labour. Existing work demonstrates promising benefits AI-based computer-assisted diagnosis systems for VCE. They also show great improvements achieve even better results. Also, medical data often sparse and unavailable research community, qualified personnel rarely time tedious labelling...
The increase of available large clinical and experimental datasets has contributed to a substantial amount important contributions in the area biomedical image analysis. Image segmentation, which is crucial for any quantitative analysis, especially attracted attention. Recent hardware advancement led success deep learning approaches. However, although models are being trained on datasets, existing methods do not use information from different epochs effectively. In this work, we leverage...
The TACOMA project is concerned with implementing operating system support for agents, processes that migrate through a network. Two prototypes have been completed; this paper outlines our experiences in building and using them. A mechanism exchanging electronic cash was explored, as well agent-based schemes scheduling fault-tolerance.
Deep learning in gastrointestinal endoscopy can assist to improve clinical performance and be helpful assess lesions more accurately. To this extent, semantic segmentation methods that perform automated real-time delineation of a region-of-interest, e.g., boundary identification cancer or pre-cancerous lesions, benefit both diagnosis interventions. However, accurate endoscopic images is extremely challenging due its high operator dependence high-definition image quality. utilize settings, it...
This paper presents a dataset of body-sensor traces and corresponding videos from several professional soccer games captured in late 2013 at the Alfheim Stadium Tromsø, Norway. Player data, including field position, heading, speed are sampled 20Hz using highly accurate ZXY Sport Tracking system. Additional per-player statistics, like total distance covered different classes, also included with 1Hz sampling rate. The provided high-definition two stationary camera arrays positioned an elevated...
Bowel preparation (cleansing) is considered to be a key precondition for successful colonoscopy (endoscopic examination of the bowel). The degree bowel cleansing directly affects possibility detect diseases and may influence decisions on screening follow-up intervals. An accurate assessment quality therefore important. Despite use reliable validated scales, grading vary from one doctor another. objective automated would contribute reduce such inequalities optimize medical resources. This...
There is a rapid growing body of knowledge regarding physical aspects football match due to studies using computer-assisted motion analysis. The present study used time-motion analysis and triaxial-accelerometers obtain new insights about differences in profiles elite players across playing-positions. Player performance data 23 official home matches from professional club, during two seasons were collected for Eighteen five different playing positions (central backs: n = 3; full-backs: 5;...
Methods based on convolutional neural networks have improved the performance of biomedical image segmentation. However, most these methods cannot efficiently segment objects variable sizes and train small biased datasets, which are common for use cases. While exist that incorporate multi-scale fusion approaches to address challenges arising with sizes, they usually complex models more suitable general semantic segmentation problems. In this paper, we propose a novel architecture called...
Analysis of medical videos from the human gastrointestinal (GI) tract for detection and localization abnormalities like lesions diseases requires both high precision recall. Additionally, it is important to support efficient, real-time processing live feedback during (i) standard colonoscopies (ii) scalability massive population-based screening, which we conjecture can be done using a wireless video capsule endoscope (camera-pill). Existing related work in this field does neither provide...
Sports analytics is a growing area of interest, both from computer system view to manage the technical challenges and sport performance aid development athletes. In this paper, we present Bagadus, prototype sports application using soccer as case study. Bagadus integrates sensor system, annotations video processing camera array. A currently installed at Alfheim Stadium in Norway, describe how can follow zoom on particular player(s). Next, will playout events games stitched panorama or...
Video analysis including classification, segmentation or tagging is one of the most challenging but also interesting topics multimedia research currently try to tackle. This often related videos from surveillance cameras social media. In last years, medical institutions produce more and video image content. Some areas analysis, like radiology brain scans, are well covered, there a much broader potential content analysis. For example, in colonoscopy, 20% polyps missed incompletely removed on...
Quantification of training and match load is an important method to personalize the stimulus' prescription players according their demands. The present study used time-motion analysis triaxial-accelerometer quantify compare: a) most demanding passages play in sessions matches (5-min peaks); b) accumulated typical microcycles official matches, by playing position. Players performance data 15 home 11 in-season were collected for analysis. divided into four different positions: Centre-backs,...
Health care has a long history of adopting technology to save lives and improve the quality living. Visual information is frequently applied for disease detection assessment, established fields computer vision medical imaging provide essential tools. It is, however, misconception that assessment are provided exclusively by these they solution all challenges. Integration analysis data from several sources, real-time processing, usefulness end-users core competences multimedia community...
Introduction The COVID-19 outbreak with partial lockdown has inevitably led to an alteration in training routines for football players worldwide. Thus, coaches had face the novel challenge of minimizing potential decline fitness during this period disruption. Methods In observational pre- posttest study involving Norwegian female (18.8 ± 1.9 years, height 1.68 0.4 m, mass 61.3 3.7 kg), we investigated effects a prescribed home-based and group-based intervention, implemented lockdown, on...
European Parliament, 'At a glance The CJEU judgement in the Schrems II case' (2020) <https://www.
Efficient matching of incoming events to persistent queries is fundamental event pattern matching, complex processing, and publish/subscribe systems. Recent processing engines based on non-deterministic finite automata (NFAs) have demonstrated scalability in the number that can be efficiently executed a single machine. However, existing NFA systems are limited Consequently, their capacity cannot increased by adding more machines.In this paper, we present an experimental evaluation different...