00 Informatik, Wissen, Systeme
Filtern
Dokumenttyp
- Wissenschaftlicher Artikel (Fachzeitschriften) (18) (entfernen)
Sprache
- Englisch (18)
Volltext vorhanden
- ja (18)
Gehört zur Bibliographie
- nein (18)
Schlagworte
- Maschinelles Lernen (5)
- machine learning (4)
- Convolutional Neural Network (3)
- Künstliche Intelligenz (3)
- Alkoholmissbrauch (2)
- Anomalie <Medizin> (2)
- Anomalieerkennung (2)
- App <Programm> (2)
- Intervention <Medizin> (2)
- Jugend (2)
This paper describes the project “Visual Knowledge Communication”, a joint project that started recently. The partners are psychologists and computer scientists from four universities of the German state Rhineland-Palatinate. The starting point for the project was the fact that visualizations have attracted considerable interest in psychology as well as computer science within the last years. However, psychologists and computer scientists pursued their investigations independently from each other in the past. This project has as its main goal the support and fostering of cooperation between psychologists and computer scientists in several visualization research projects.
The paper sketches the overall project. It then discusses in more detail the authors' subproject which deals with a peer review process for animations developed by students. The basic ideas, the main goals, and the project plan are described.
This paper is a work-in-progress report. Therefore, it does not contain any results.
Background: Problem drinking, particularly risky single-occasion drinking is widespread among adolescents and young adults in most Western countries. Mobile phone text messaging allows a proactive and cost-effective delivery of short messages at any time and place and allows the delivery of individualised information at times when young people typically drink alcohol. The main objective of the planned study is to test the efficacy of a combined web- and text messaging-based intervention to reduce problem drinking in young people with heterogeneous educational level.
Methods/Design: A two-arm cluster-randomised controlled trial with one follow-up assessment after 6 months will be conducted to test the efficacy of the intervention in comparison to assessment only. The fully-automated intervention program will provide an online feedback based on the social norms approach as well as individually tailored mobile phone text messages to stimulate (1) positive outcome expectations to drink within low-risk limits, (2) self-efficacy to resist alcohol and (3) planning processes to translate intentions to resist alcohol into action. Program participants will receive up to two weekly text messages over a time period of 3 months. Study participants will be 934 students from approximately 93 upper secondary and vocational schools in Switzerland. Main outcome criterion will be risky single-occasion drinking in the past 30 days preceding the follow-up assessment.
Discussion: This is the first study testing the efficacy of a combined web- and text messaging-based intervention to reduce problem drinking in young people. Given that this intervention approach proves to be effective, it could be easily implemented in various settings, and it could reach large numbers of young people in a cost-effective way.
Background: Tobacco smoking prevalence continues to be high, particularly among adolescents and young adults with lower educational levels, and is therefore a serious public health problem. Tobacco smoking and problem drinking often co-occur and relapses after successful smoking cessation are often associated with alcohol use. This study aims at testing the efficacy of an integrated smoking cessation and alcohol intervention by comparing it to a smoking cessation only intervention for young people, delivered via the Internet and mobile phone.
Methods/Design: A two-arm cluster-randomised controlled trial with one follow-up assessment after 6 months will be conducted. Participants in the integrated intervention group will: (1) receive individually tailored web-based feedback on their drinking behaviour based on age and gender norms, (2) receive individually tailored mobile phone text messages to promote drinking within low-risk limits over a 3-month period, (3) receive individually tailored mobile phone text messages to support smoking cessation for 3 months, and (4) be offered the option of registering for a more intensive program that provides strategies for smoking cessation centred around a self-defined quit date. Participants in the smoking cessation only intervention group will only receive components (3) and (4). Study participants will be 1350 students who smoke tobacco daily/occasionally, from vocational schools in Switzerland. Main outcome criteria are 7-day point prevalence smoking abstinence and cigarette consumption assessed at the 6-month follow up.
Discussion: This is the first study testing a fully automated intervention for smoking cessation that simultaneously addresses alcohol use and interrelations between tobacco and alcohol use. The integrated intervention can be easily implemented in various settings and could be used with large groups of young people in a cost-effective way.
Ahmad et al. in their paper for the first time proposed to apply sharp function for classification of images. In continuation of their work, in this paper we investigate the use of sharp function as an edge detector through well known diffusion models. Further, we discuss the formulation of weak solution of nonlinear diffusion equation and prove uniqueness of weak solution of nonlinear problem. The anisotropic generalization of sharp operator based diffusion has also been implemented and tested on various types of images.
One key for successful and fluent human-robot-collaboration in disassembly processes is equipping the robot system with higher autonomy and intelligence. In this paper, we present an informed software agent that controls the robot behavior to form an intelligent robot assistant for disassembly purposes. While the disassembly process first depends on the product structure, we inform the agent using a generic approach through product models. The product model is then transformed to a directed graph and used to build, share and define a coarse disassembly plan. To refine the workflow, we formulate "the problem of loosening a connection and the distribution of the work" as a search problem. The created detailed plan consists of a sequence of actions that are used to call, parametrize and execute robot programs for the fulfillment of the assistance. The aim of this research is to equip robot systems with knowledge and skills to allow them to be autonomous in the performance of their assistance to finally improve the ergonomics of disassembly workstations.
Sustainable software products - Towards assessment criteria for resource and energy efficiency
(2018)
Many authors have proposed criteria to assess the “environmental friendliness” or “sustainability” of software products. However, a causal model that links observable properties of a software product to conditions of it being green or (more general) sustainable is still missing. Such a causal model is necessary because software products are intangible goods and, as such, only have indirect effects on the physical world. In particular, software products are not subject to any wear and tear, they can be copied without great effort, and generate no waste or emissions when being disposed of. Viewed in isolation, software seems to be a perfectly sustainable type of product. In real life, however, software products with the same or similar functionality can differ substantially in the burden they place on natural resources, especially if the sequence of released versions and resulting hardware obsolescence is taken into account. In this article, we present a model describing the causal chains from software products to their impacts on natural resources, including energy sources, from a life-cycle perspective. We focus on (i) the demands of software for hardware capacities (local, remote, and in the connecting network) and the resulting hardware energy demand, (ii) the expectations of users regarding such demands and how these affect hardware operating life, and (iii) the autonomy of users in managing their software use with regard to resource efficiency. We propose a hierarchical set of criteria and indicators to assess these impacts. We demonstrate the application of this set of criteria, including the definition of standard usage scenarios for chosen categories of software products. We further discuss the practicability of this type of assessment, its acceptability for several stakeholders and potential consequences for the eco-labeling of software products and sustainable software design.
Companies have made considerable progress in assessing the sustainability of their processes and products, including the information and communication technology (ICT) sector. However, it is surprising that little attention has been given to the sustainability performance of software products. For this article, we chose a case study approach to explore the extent, to which software manufacturers have considered sustainability criteria for their products. We selected a manufacturer of sustainability management software on the assumption that they would be more likely to integrate elements of sustainability performance in their products. In the case study, we applied a previously developed set of criteria for sustainable software (SCSS) using a questionnaire and experiments, to assess a web-based sustainability management software product regarding its sustainability performance. The assessment finds that despite a sustainability conscious manufacturer, a systematic assessment of sustainability regarding software products is missing in the case study. This implies that sustainability assessment for software products is still novel, corresponding knowledge is missing and suitable tools are not yet being widely applied in the industry. The SCSS presents a suitable approach to close this gap, but it does require further refinement, for example regarding its applicability to web-based software on external servers.
Containerization is one of the most important topics for modern data centers and web developers. Since the number of containers on one- and multi-node systems is growing, knowledge about the energy consumption behavior of single web-service containers is essential in order to save energy and, of course, money. In this article, we are going to show how the energy consumption behavior of single containerized web services/web apps changes while creating replicas of the service in order to scale and balance the web service.
Background: High numbers of consumable medical materials (eg, sterile needles and swabs) are used during the daily routine of intensive care units (ICUs) worldwide. Although medical consumables largely contribute to total ICU hospital expenditure, many hospitals do not track the individual use of materials. Current tracking solutions meeting the specific requirements of the medical environment, like barcodes or radio frequency identification, require specialized material preparation and high infrastructure investment. This impedes the accurate prediction of consumption, leads to high storage maintenance costs caused by large inventories, and hinders scientific work due to inaccurate documentation. Thus, new cost-effective and contactless methods for object detection are urgently needed.
Objective: The goal of this work was to develop and evaluate a contactless visual recognition system for tracking medical consumable materials in ICUs using a deep learning approach on a distributed client-server architecture.
Methods: We developed Consumabot, a novel client-server optical recognition system for medical consumables, based on the convolutional neural network model MobileNet implemented in Tensorflow. The software was designed to run on single-board computer platforms as a detection unit. The system was trained to recognize 20 different materials in the ICU, while 100 sample images of each consumable material were provided. We assessed the top-1 recognition rates in the context of different real-world ICU settings: materials presented to the system without visual obstruction, 50% covered materials, and scenarios of multiple items. We further performed an analysis of variance with repeated measures to quantify the effect of adverse real-world circumstances.
Results: Consumabot reached a >99% reliability of recognition after about 60 steps of training and 150 steps of validation. A desirable low cross entropy of <0.03 was reached for the training set after about 100 iteration steps and after 170 steps for the validation set. The system showed a high top-1 mean recognition accuracy in a real-world scenario of 0.85 (SD 0.11) for objects presented to the system without visual obstruction. Recognition accuracy was lower, but still acceptable, in scenarios where the objects were 50% covered (P<.001; mean recognition accuracy 0.71; SD 0.13) or multiple objects of the target group were present (P=.01; mean recognition accuracy 0.78; SD 0.11), compared to a nonobstructed view. The approach met the criteria of absence of explicit labeling (eg, barcodes, radio frequency labeling) while maintaining a high standard for quality and hygiene with minimal consumption of resources (eg, cost, time, training, and computational power).
Conclusions: Using a convolutional neural network architecture, Consumabot consistently achieved good results in the classification of consumables and thus is a feasible way to recognize and register medical consumables directly to a hospital’s electronic health record. The system shows limitations when the materials are partially covered, therefore identifying characteristics of the consumables are not presented to the system. Further development of the assessment in different medical circumstances is needed.
Fuzzy system based on two-step cascade genetic optimization strategy for tobacco tar prediction
(2019)
There are many challenges in accurately measuring cigarette tar constituents. These include the need for standardized smoke generation methods related to unstable mixtures. In this research were developed algorithms using fusion of artificial intelligence methods to predict tar concentration. Outputs of development are three fuzzy structures optimized with genetic algorithms resulting in genetic algorithm (GA)-FUZZY, GA-adaptive neuro fuzzy inference system (ANFIS), GA-GA-FUZZY algorithms. Proposed algorithms are used for the tar prediction in the cigarette production process. The results of prediction are compared with gas chromatograph (high-performance liquid chromatography (HPLC)) readings.