A database for publications published by researchers and students at SimulaMet.
Research area
Publication type
- All (686)
- Journal articles (178)
- Books (7)
- Edited books (1)
- Proceedings, refereed (215) Remove Proceedings, refereed <span class="counter">(215)</span> filter
- Book chapters (6)
- Talks, keynote (16)
- PhD theses (1)
- Proceedings, non-refereed (17)
- Posters (10)
- Technical reports (11)
- Manuals (1)
- Talks, invited (155)
- Talks, contributed (13)
- Public outreach (45)
- Miscellaneous (10)
Proceedings, refereed
HTAD: A Home-Tasks Activities Dataset with Wrist-accelerometer and Audio Features
In 27th International Conference on Multimedia Modeling. Springer, 2021.Status: Published
HTAD: A Home-Tasks Activities Dataset with Wrist-accelerometer and Audio Features
| Afilliation | Machine Learning |
| Project(s) | Department of Holistic Systems |
| Publication Type | Proceedings, refereed |
| Year of Publication | 2021 |
| Conference Name | 27th International Conference on Multimedia Modeling |
| Publisher | Springer |
| Keywords | Accelerometer, Activity recognition, Audio, dataset, Sensor fusion |
Reliable Server Pooling Based Workload Offloading with Mobile Edge Computing: A Proof-of-Concept
In Proceedings of the 2nd International Workshop on Recent Advances for Multi-Clouds and Mobile Edge Computing (M2EC) in conjunction with the 35th International Conference on Advanced Information Networking and Applications (AINA). Toronto, Ontario/Canada: Springer, 2021.Status: Published
Reliable Server Pooling Based Workload Offloading with Mobile Edge Computing: A Proof-of-Concept
In recent times, mobile broadband devices have become almost ubiquitous. However, battery-powered devices (such as smartphones), have limitations on energy consumption, computation power and storage space. Cloud computing, and in particular with the upcoming 5G networks, Mobile Edge Computing (MEC) can provide compute and storage services at the vicinity of the user (with a low communication latency). However, the complexity lies in how to simply and efficiently realise MEC services, with the auxiliary public (multi-)cloud resources? In this paper, we propose a proof-of-concept of using Reliable Server Pooling (RSerPool) as a light-weight layer of managing resource pools and handling application sessions with these pools. Our approach is simple, efficient, has low overhead and is available as open-source. Here, we demonstrate the usefulness of our approach by measuring in a test setup, with a 4G testbed connected to MEC and public multi-cloud resources.
| Afilliation | Communication Systems |
| Project(s) | NorNet, The Center for Resilient Networks and Applications, SMIL: SimulaMet Interoperability Lab, 5G-VINNI: 5G Verticals INNovation Infrastructure , Simula Metropolitan Center for Digital Engineering, Simula Metropolitan Center for Digital Engineering, MELODIC: Multi-cloud Execution-ware for Large-scale Optimised Data-Intensive Computing |
| Publication Type | Proceedings, refereed |
| Year of Publication | 2021 |
| Conference Name | Proceedings of the 2nd International Workshop on Recent Advances for Multi-Clouds and Mobile Edge Computing (M2EC) in conjunction with the 35th International Conference on Advanced Information Networking and Applications (AINA) |
| Publisher | Springer |
| Place Published | Toronto, Ontario/Canada |
| Keywords | 5G, Evolved Packet Core (EPC), Mobile Edge Computing (MEC), Multi-Cloud Computing, Network Function Virtualisation (NFV), Reliable Server Pooling (RSerPool) |
Identification of spermatozoa by unsupervised learning from video data
In ESHRE , 2021.Status: Accepted
Identification of spermatozoa by unsupervised learning from video data
Identification of individual sperm is essential to assess a given sperm sample's motility behavior. Existing computer-aided systems need training data based on annotations by professionals, which is resource demanding. On the other hand, data analysed by unsupervised machine learning algorithms can improve supervised algorithms that are more stable for clinical applications. Therefore, unsupervised sperm identification can improve computer-aided sperm analysis systems predicting different aspects of sperm samples. Other possible applications are assessing kinematics and counting of spermatozoa.
| Afilliation | Machine Learning |
| Project(s) | Department of Holistic Systems |
| Publication Type | Proceedings, refereed |
| Year of Publication | 2021 |
| Conference Name | ESHRE |
Generative Adversarial Networks For Creating Realistic Artificial Colon Polyp Images
In DDW, 2021.Status: Accepted
Generative Adversarial Networks For Creating Realistic Artificial Colon Polyp Images
| Afilliation | Machine Learning |
| Project(s) | Department of Holistic Systems |
| Publication Type | Proceedings, refereed |
| Year of Publication | 2021 |
| Conference Name | DDW |
Impact of Image Resolution on Convolutional Neural Networks Performance in Gastrointestinal Endoscopy
In DDW, 2021.Status: Accepted
Impact of Image Resolution on Convolutional Neural Networks Performance in Gastrointestinal Endoscopy
Convolutional neural networks (CNNs) are increasingly used to improve and automate processes in gastroenterology, like the detection of polyps during a colonoscopy. An important input to these methods is images and videos. Up until now, no well-defined, common understanding or standard regarding the resolution of the images and video frames has been defined, and to reduce processing time and resource requirements, images are today almost always down-sampled. However, how such down-sampling and the image resolution influence the performance in context with medical data is unknown. In this work, we investigate how the resolution relates to the performance of convolutional neural networks. This can help set standards for image or video characteristics for future CNN based models in gastrointestinal endoscopy.
| Afilliation | Machine Learning |
| Project(s) | Department of Holistic Systems |
| Publication Type | Proceedings, refereed |
| Year of Publication | 2021 |
| Conference Name | DDW |
DivergentNets: Medical Image Segmentation by Network Ensemble
In EndoCV, 2021.Status: Accepted
DivergentNets: Medical Image Segmentation by Network Ensemble
Detection of colon polyps has become a trending topic in the intersecting fields of machine learning and gastrointestinal endoscopy. The focus has mainly been on per-frame classification. More recently, polyp segmentation has gained attention in the medical community. Segmentation has the advantage of being more accurate than per-frame classification or object detection as it can show the affected area in greater detail. For our contribution to the EndoCV 2021 segmentation challenge, we propose two separate approaches. First, a segmentation model named TriUNet composed of three separate UNet models. Second, we combine TriUNet with an ensemble of well-known segmentation models, namely UNet++, FPN, DeepLabv3, and DeepLabv3+, into a model called DivergentNets to produce more generalizable medical image segmentation masks. In addition, we propose a modified Dice loss that calculates loss only for a single class when performing multi-class segmentation, forcing the model to focus on what is most important. Overall, the proposed methods achieved the best average scores for each respective round in the challenge, with TriUNet being the winning model in Round I and DvergentNets being the winning model in Round II of the segmentation generalization challenge at EndoCV 2021
| Afilliation | Machine Learning |
| Project(s) | Department of Holistic Systems |
| Publication Type | Proceedings, refereed |
| Year of Publication | 2021 |
| Conference Name | EndoCV |
DeepSynthBody: the beginning of the end for data deficiency in medicine
In The International Conference on Applied Artificial Intelligence (ICAPAI), 2021.Status: Accepted
DeepSynthBody: the beginning of the end for data deficiency in medicine
Limited access to medical data is a barrier on developing new and efficient machine learning solutions in medicine such as computer-aided diagnosis, risk assessments, predicting optimal treatments and home-based personal healthcare systems. This paper presents DeepSynthBody: a novel framework that overcomes some of the inherent restrictions and limitations of medical data by using deep generative adversarial networks to produce synthetic data with characteristics similar to the real data, so-called DeepSynth (deep synthetic) data. We show that DeepSynthBody can address two key issues commonly associated with medical data, namely privacy concerns (as a result of data protection rules and regulations) and the high costs of annotations. To demonstrate the full pipeline of applying DeepSynthBody concepts and user-friendly functionalities, we also describe a synthetic medical dataset generated and published using our framework. DeepSynthBody opens a new era of machine learning applications in medicine with a synthetic model of the human body.
| Afilliation | Machine Learning |
| Project(s) | Department of Holistic Systems |
| Publication Type | Proceedings, refereed |
| Year of Publication | 2021 |
| Conference Name | The International Conference on Applied Artificial Intelligence (ICAPAI) |
Preliminary Evaluation of a Survey Checklist in the Context of Evidence-based Software Engineering Education
In 16th International Conference on Evaluation of Novel Approaches to Software EngineeringProceedings of the 16th International Conference on Evaluation of Novel Approaches to Software Engineering. Online Streaming, --- Select a Country ---: SCITEPRESS - Science and Technology Publications, 2021.Status: Published
Preliminary Evaluation of a Survey Checklist in the Context of Evidence-based Software Engineering Education
Background: In order to judge evidence it is important to be able to assess study quality. Checklists are means to objectify the assessment. In an earlier study we proposed and evaluated a checklist for surveys, which was assessed by experts. Objective: (1) To assess whether the use of the checklist enables students with limited experience in research to consistently and accurately assess the quality of a research paper. (2) To elicit qualitative feedback to identify improvements to the checklist. Method: The students reviewed a survey in a one-group posttest-only quasi-experiment using the checklist. In total 13 students participated in the context of the course Evidence-based software engineering as part of the study program Information Systems at Flensburg University of Applied Sciences. Results: In total the students achieved 74% percent of agreement among each other. However, the Kappa values indicated mostly a poor level of agreement considering the classification by Fleiss. In addition, the students were quite inaccurate assessing the questions. Though, they performed well on questions for research objectives and the identification of population. Conclusion: Findings indicate that students do not assess reliably. However, further investigations are needed to substantiate the findings.
| Afilliation | Software Engineering |
| Project(s) | Department of IT Management |
| Publication Type | Proceedings, refereed |
| Year of Publication | 2021 |
| Conference Name | 16th International Conference on Evaluation of Novel Approaches to Software EngineeringProceedings of the 16th International Conference on Evaluation of Novel Approaches to Software Engineering |
| Publisher | SCITEPRESS - Science and Technology Publications |
| Place Published | Online Streaming, --- Select a Country --- |
| ISBN Number | 978-989-758-508-1 |
| Keywords | Checklist, One-group Quasi-experiment, Students, survey |
| URL | https://www.scitepress.org/DigitalLibrary/Link.aspx?doi=10.5220/00104962... |
| DOI | 10.5220/0010496204370444 |
Progressively Normalized Self-Attention Network for Video Polyp Segmentation
In MICCAI 2021. Springer, 2021.Status: Accepted
Progressively Normalized Self-Attention Network for Video Polyp Segmentation
Existing video polyp segmentation (VPS) models typically employ convolutional neural networks (CNNs) to extract features. However, due to their limited receptive fields, CNNs can not fully exploit the global temporal and spatial information in successive video frames, resulting in false positive segmentation results. In this paper, we propose the novel PNS-Net(Progressively Normalized Self-attention Net-work), which can efficiently learn representations from polyp videos with real-time speed (∼140fps) on a single RTX 2080 GPU and no post-processing. OurPNS-Netis based solely on a basic normalized self-attention block, equipping with recurrence and CNNs entirely. Experiments on challenging VPS datasets demonstrate that the proposed PMS-Netachieves state-of-the-art performance. We also conduct extensive experiments to study the effectiveness of the channel split, soft-attention, and progressive learning strategy. We find that our PNS-Networks well under different settings, making it a promising solution to the VPS task.
| Afilliation | Machine Learning |
| Project(s) | Department of Holistic Systems |
| Publication Type | Proceedings, refereed |
| Year of Publication | 2021 |
| Conference Name | MICCAI 2021 |
| Publisher | Springer |
| Keywords | colonoscopy, Normalized self-attention, Polyp segmentation |
Automatic Polyp Segmentation using U-Net-ResNet50
In Medico MediaEval 2020. MediaEval 2020, 2021.Status: Published
Automatic Polyp Segmentation using U-Net-ResNet50
Polyps are the predecessors to colorectal cancer which is considered as one of the leading causes of cancer-related deaths worldwide. Colonoscopy is the standard procedure for the identification, localization, and removal of colorectal polyps. Due to variability in shape, size, and surrounding tissue similarity, colorectal polyps are often missed by the clinicians during colonoscopy. With the use of an automatic, accurate, and fast polyp segmentation method during the colonoscopy, many colorectal polyps can be easily detected and removed. The ``Medico automatic polyp segmentation challenge'' provides an opportunity to study polyp segmentation and build an efficient and accurate segmentation algorithm. We use the U-Net with pre-trained ResNet50 as the encoder for the polyp segmentation. The model is trained on Kvasir-SEG dataset provided for the challenge and tested on the organizer's dataset and achieves a dice coefficient of 0.8154, Jaccard of 0.7396, recall of 0.8533, precision of 0.8532, accuracy of 0.9506, and F2 score of 0.8272, demonstrating the generalization ability of our model.
| Afilliation | Machine Learning |
| Project(s) | Department of Holistic Systems |
| Publication Type | Proceedings, refereed |
| Year of Publication | 2021 |
| Conference Name | Medico MediaEval 2020 |
| Publisher | MediaEval 2020 |