教学课程资源库 更多>>

摘要： Presents the introductory editorial for this issue of the publication.

关键词： Computer science   Design   Biomedical engineering   Multimedia communications  

摘要： Interacting with multimedia content poses a major challenge for people with visual impairments. Making such multimedia content accessible to people with vision impairments has become all the more important because of the ever-increasing prevalence of such content, particularly videos, on the Internet. Recent advances in computer vision offer a promising avenue for tackling challenges in multimedia accessibility. In this thesis, we propose solutions to these accessibility problems by leveraging computer vision techniques. First, we introduce SViM, a saliency-driven video magnifier interface. SViM magnifies salient regions in a video, and offers auto-panning functionality, paving the way towards a better video watching experience for people with low vision. SViM provides three different magnifier interfaces along with accessibility customizations to further enhance the overall user experience. Quantitative and qualitative evaluation of SViM through user studies with low-vision participants demonstrate SViM’s effectiveness compared to state-of-the-art alternatives. Second, we present SaIL, a method to automatically inject ARIA (Accessible Rich Internet Applications) landmarks in a webpage. Although ARIA landmarks have the potential to enhance web accessibility for people with vision impairments, they have remained under-utilized by web developers. SaIL allows screen reader users with visual impairments to rapidly skim and skip to important regions in a webpage. Third, we report our work on assessing the accessibility of videos for people who are blind. Towards this, we collected a dataset comprising subjective evaluations of video accessibility. We extract handcrafted features from both video and audio that show correlation with video accessibility, and utilize these features for predicting user generated video accessibility evaluations. Furthermore, these features play a significant role in diagnosis, namely identifying the causes of video inaccessibility.

关键词： Computer science   Artificial intelligence   Computer engineering   Higher education administration  

摘要： Nowadays, much research is being carried out to find efficient algorithms for optimal automated university course timetable problems (UCTP). UCTP allocates the university's events like lectures, exams to the various resources, including instructors, students, lecture time and classrooms. Class scheduling is one of the biggest challenging problems of educational institutions. In this thesis, the aim is to improve the state-of-art for a class scheduling problem considering some hard and soft constraints. Hard constraints must be satisfied. Soft constraints need not be satisfied, but there is a penalty for each soft constraint violation. We also have a timing penalty for scheduling each class to a specific schedule. The goal is to allocate classes to their schedule so that the total penalty is minimized. The proposed method adopts the meta-heuristic strategy to improve existing solutions. An acceptance criterion is defined on neighbouring solutions with cooling and an energy function to avoid getting stuck at a local optimum. This criterion extends the same in Simulated Annealing (SA) by giving infeasible neighbours a chance to become candidates. We then compared our proposed models for the feasible and infeasible solution on two different datasets based on iteration vs. penalty with the local search algorithm. The results obtained based on our experiment show that the proposed model for a feasible solution outperformed the local search algorithm by about 20% in 20k iterations on average. While the model for the infeasible solution performed about 52% better than the local search algorithm for the 20k iterations on average. However, both of the proposed models take more execution time compared to the local search algorithm.

关键词： Physical anthropology   Ecology   Computer science  

摘要： Biological anthropologists seek to understand the origin and evolution of distinctively human traits, including language, cumulative culture, and intensely cooperative societies. To understand the evolution of human behavior, we need a viable theory of socioecology. Studies of our primate relatives provide opportunities to develop and test socioecological models which can in turn help explain the evolution of human behavior. However, it has become increasingly clear that current socioecological models do not fully explain primate social behavior. Among many important limiting factors are the sheer number of variables to track and the difficulty of operationalizing ecological concepts such that they can be measured feasibly in the field. Microscale (i.e., agent-based) models have great potential for advancing the field because they can handle many different variables and incorporate individual variation, stochasticity, and emergent properties. I reviewed the existing literature on microscale models in behavioral ecology to better understand the state-of-the-art for models relevant to primate socioecology. I found that models are often designed for a single species or single area of research, limiting their application to broader questions. Many models also do not consider important biological constraints such as spatial relationships or rules for birth and death that depend on individual characteristics, nor are they often validated for accuracy. As a contribution towards a more complete understanding of primate socioecology, I developed B3GET, a microscale model that incorporates important biological constraints and can track key socioecological variables in simulated primates. These virtual primates possess decision-making rules encoded in simulated diploid chromosomes, which dictate movement, body growth, inclination to mate, eat, and other behaviors. I developed these rules based on primate socioecological data from the literature and my own field observations. T

关键词： Robotics   Physiology   Biology   Computer science   Disability studies   Electrical engineering   Engineering   Mathematics   Morphology   Surgery  

摘要： Lower-limb amputations severely affects the ability of individuals to perform functional tasks, which affects their quality of life. Robotic lower limb prostheses have been developed that are capable of generating the same torque and power requirements as those of able-bodied joints. These existing robotic lower-limb prostheses use autonomous control to address cyclic, locomotive tasks, but they are inadequate to operate the prosthesis for daily activities that are non-cyclic and unpredictable. To address this challenge, this study aims to design a novel electromyography (EMG)-driven musculoskeletal model for volitional control of a robotic ankle-foot prosthesis. This controller places the user in continuous control of the device, allowing them to freely manipulate the prosthesis behavior at will. The Hill-type muscle model was used to model a dorsiflexor and a plantarflexor, which functioned around a virtual ankle joint. The model parameters were determined by fitting the model prediction to the experimental data collected from an able-bodied subject. EMG signals recorded from ankle agonist and antagonist muscle pair were used to activate the virtual muscle models. This model was validated via offline simulations and real-time prosthesis control. Additionally, the feasibility of the proposed prosthesis control on assisting the user’s functional tasks was demonstrated. The present control may further improve the function of robotic prosthesis for supporting versatile activities in individuals with lower-limb amputations.

关键词： Plasma physics   Computer science   Load   Propagation   Simulation   Numerical analysis   Plasma   Energy   Electrons   Electric fields  

关键词： 计算机科学   技术   应用现状   发展趋势  

摘要： 本文主要针对计算机科学与技术的应用现状与发展趋势展开研究,提出了计算机科学与技术的应用特点,对计算机科学与技术的应用现状进行阐述,主要包括办公自动化、物流自动化、智能交通、人工智能技术,总结了几点计算机科学与技术的发展趋势,主要为生物计算机、量子计算机、光子计算机、纳米计算机,充分彰显出计算机科学与技术的应用价值,使其应用范围更为广阔化,并其应用潜能发挥到极致,其现实意义显著.

关键词： Medical imaging   Computer science   Mechanics   Medicine   Physics   Polymer chemistry   Surgery  

关键词： Artificial intelligence   Bioinformatics   Computer science  

摘要： Proteins are fundamental building blocks of cellular function. They systematically interact with other proteins to make life happen. Understanding these protein-protein interactions is important for obtaining a detailed understanding of protein function and to enable the process of drug and vaccine design. Experimental methods for studying protein interfaces including X-ray crystallography, NMR, and Cryo-electron microscopy, are expensive, time consuming, and sometimes unsuccessful due to the unstable nature of many protein-protein interactions. Computational docking experiments are a cheap and fast alternative. Docking algorithms produce a large number of potential solutions that are then ranked by quality. However, current scoring methods are not good enough for finding a docking solution that is close to the native structure. That has led to the development of machine learning methods for this task. These methods typically involve extensive engineering of features to describe the protein complex, and are not very successful at identifying good quality solutions among the top ranks. In this thesis, we propose a scoring technique that uses graph neural networks that function at the atomic level to learn the interfaces of docked proteins without the need for feature engineering. We evaluate our model and show that it performs better than commonly used docking methods and deep learning methods that use 3D CNNs.

关键词： Computer science   Public health  

摘要： Rapidly developing sequencing methods and analytical techniques is enabling a deeper level of understanding of the relationship between the microbiome and human health. Numerous studies have demonstrated differences in the composition of host-associated microbial communities between healthy and disease states. An important emerging challenge in the microbiome analysis is the description of microbial community dynamics that occur during progressive disease. In order to gain further insight into the role of pathogens and develop better tools with clinical utility for disease diagnosis and prognosis, much more precise understanding is needed of the variations in microbial community structures during disease progression. Due to the requirements of sample size and extended study period, it is difficult to model microbial dynamics in chronic illness through longitudinal studies. Thus, there is an increasing need for novel approaches delineating the microbial dynamics in disease progression that is economically and logistically feasible. To mitigate the aforementioned longstanding issues that have challenged the research community, we presented a novel bioinformatics pipeline designed to delineate microbial community dynamics over the course of disease development using static sample data. Based on high-dimensional 16S rRNA sequence data, disease-stage-related operational taxonomic units (OTUs) are used to learn a principal-tree that depicts the microbial community progression structure embedded in the data. Reverse mapping of clinical variables to the microbial progression model can reveal information previously unavailable using conventional association analyses. We demonstrated the utility of the proposed pipeline on a previously published data set of Crohn's disease presenting alternate disease progression trajectories. By integrating static sample data, the proposed pipeline can facilitate the construction of a microbiome progression model and provide new insights int