Liliane dos Santos Machado

Using fuzzy hidden Markov models for online training evaluation and classification in virtual reality simulators

The goal of the training evaluation in a simulation is to provide feedback about the user performance in the training environment. Approaches to online or offline evaluation of training in simulators based on virtual reality have been proposed. An online evaluator must have low complexity algorithm to not compromise the performance of the simulator. A new approach to quality of training online evaluation in virtual reality worlds is proposed in this work. This approach uses fuzzy hidden Markov models (FHMM) for modeling and classification of trainee in pre-defined classes of training.The goal of the training evaluation in a simulation is to provide feedback about the user performance in the training environment. Approaches to online or offline evaluation of training in simulators based on virtual reality have been proposed. An online evaluator must have low complexity algorithm to not compromise the performance of the simulator. A new approach to quality of training online evaluation in virtual reality worlds is proposed in this work. This approach uses fuzzy hidden Markov models (FHMM) for modeling and classification of trainee in pre-defined classes of training.

Hidden Markov Models For Learning Evaluation In Virtual-Reality Simulators

Abstract Training on simulator systems based on virtual reality (VR) for learning or learning improvement may be a cost-effective and efficient alternative to traditional training methods. A new approach for online or Web-based evaluation of training performed in simulators based on VR is proposed. This approach uses hidden Markov models for modelling and classification of training in pre-defined classes of training. In this article we show an example of application in a simulator of bone marrow harvest for transplant.

A Virtual Environment for Training and Assessment of Surgical Teams

Collaborative Virtual Environments (CVEs) can improve the way remote users interact with one another while learning and training skills on a given task. One CVEs possibility to the health area is the simulation of medical procedures in which a group of remote users can train and interact simultaneously. Health area has been benefited from the advent of Virtual Reality (VR) especially in education area, where these systems present some advantages over traditional teaching methods such as: cost reduction for training, reducing the use of guinea pigs and anatomical specimens in laboratory practices as well the use of interactive teaching approaches. Another VRs feature is the ability to monitor users actions for assessing training performance. Thus, statistical models are used in order to check whether a group performed the procedure correctly or not. The goal is to allow the formation of teams and the development of individual skills to work together. This work proposes and discusses one CVEs architecture for supporting training and assessment of team skills, during surgery simulation.

Multiple assessment for multiple users in virtual reality training environments

With recent computational advances, several interaction devices can be used by different users who share the same virtual world, allowing the simulation of realistic environments, such as surgical rooms. In order to deal with this feature, assessment systems must be generalized to evaluate, individually, all users of the simulation and to make the aspects of their interactions known. In this paper we propose a new assessment system for training based on virtual reality which can evaluate more than one user at a time. The methodology proposed uses data collected from user interaction and group interactions during training to create user profile and group profile. The main advantages of that approach are: both of reports can be used to increase group performance and the interactions among users, during training, can be monitored to correct and improve group tasks in procedure such as sequential, simultaneous or collaborative tasks.

Avaliação do sistema de referência e contrarreferência do estado da Paraíba segundo os profissionais da Atenção Básica no contexto do 1º ciclo de Avaliação Externa do PMAQ-AB

Este trabalho objetiva caracterizar a organizacao, a articulacao e a assistencia da rede de atencao a saude no estado da Paraiba, com base nos dados do 1o ciclo de Avaliacao Externa do Programa Nacional de Melhoria do Acesso e da Qualidade da Atencao Basica (PMAQ-AB). Realizou-se um estudo do tipo transversal, cujos dados foram analisados descritivamente por meio de valores absolutos e respectivos percentuais. Concluiu-se que existem fragilidades na integracao da rede de atencao a saude na Paraiba em relacao ao ordenamento e definicao do fluxo, em especial na contrarreferencia, fato que pode comprometer a integralidade e o papel da Atencao Basica de coordenar o cuidado e ordenar as redes.

A Peer-to-Peer Multicast Architecture for Supporting Collaborative Virtual Environments (CVEs) in Medicine

Collaborative Virtual Environments (CVEs) can improve the way remote users interact with one another while learning or training skills on a given task. One CVEs application is the possibility of simulating medical procedures in which a group of remote users can train and interact simultaneously. It is important that networking issues and performance evaluation of CVEs allows us to understand how such systems can work in the Internet, as well as the requirements for multisensorial and real-time data. Thus, this paper discloses implementation issues of a peer-to-peer multicast network architecture on the collaborative module of the CyberMed VR framework. The multicast protocol is known to provide better scalability and decrease the use of bandwidth on CVEs, allowing better Quality of Experience (QoE). Finally it presents the results of a performance evaluation experiment.

Gaussian Mixture Models for Supervised Classification of Remote Sensing Multispectral Images

This paper proposes the use of Gaussian Mixture Models as a supervised classifier for remote sensing multispectral images. The main advantage of this approach is provide more adequated adjust to several statistical distributions, including non-symmetrical statistical distributions. We present some results of this method application over a real image of an area of Tapajos River in Brazil and the results are analysed according to a reference image. We perform also a comparison with Maximum Likelihood classifier. The Gaussian Mixture classifier obtained best adjust about image data and best classification performance too.

VR-based simulation for the learning of gynaecological examination

Training is an effective way to acquire learning. Recently, virtual reality systems have been proposed and developed to provide training in medical procedures. This work presents the SITEG, a virtual reality simulator for education in the practice of gynaecological examinations. The SITEG offers a virtual environment where the user can practice the two steps present in a real exam and learn different stages of diseases like Herpes and Human papillomavirus. For that, the system integrates stereoscopic view and a haptic device to allow the identification of malignancies commonly found in this kind of examination. Attached to the simulator there is an assessment tool to classify the user training.

A 3D Intelligent Campus to Support Distance Learning

The Intelligent Campus is an extension of control and monitor systems for interactive models. More than a simple 3D model, the Intelligent Campus is an interactive 3D model of a real campus that allows remote access of learning tools developed and available locally for present learning. It increases the set of learning tools and services provided by the institution and do not require re-development of tools. For access control, the Intelligent Campus implements an account mechanism that requires password for specific services and control access for distance learning classes that can occur in the campus. To manage the accesses were included levels of complexity in the virtual model that control the interactive details of an object presented to the user.

Psychomotor skills assessment in medical training based on virtual reality using a Weighted Possibilistic approach

Virtual reality has been used to provide training systems in several areas, particularly in medicine. In that area, users interactions in a virtual environment are modeled and compared with predefined classes of performance to know how much users are prepared to perform that procedure on human beings. In this paper is proposed a new approach for online Single Users Assessment System (SUAS) using Weighted Possibility and Necessity measures. Those fuzzy measures provide an interval-based estimator to check the compatibility between interactions of an user and previous stored classes of performance. This approach integrates the kernel of a decision support system and it contributes to improve the users assessment taking into account the individual relevance of variables in a procedure. It was verified in performance tests, when this new SUAS approach achieved better results, according to Kappa Coefficient, when compared with other previous approaches.