Guaranteeing anatomical shape consistency in cardiac magnetic resonance imaging for left ventricle segmentation is a complex task due to its shape-changing during the cardiac cycle, the low contrast and resolution of images, the size change between apical and basal slices, the similarity with nearby organs, and the presence of cardiomyopathies that can deform the heart. Although producing segmentations very close to the ones produced by experts according to standard evaluation metrics, deep learning networks still often produce anatomically inconsistent segmentations. In this work, we propose a new shape-based loss function that favors shape consistency. The loss function uses shape information extracted from distance maps estimated by the network. We validate our approach with the ACDC and Sunnybrook public datasets by using standard metrics as well as a shape similarity metric. The results indicate that the proposed loss is able to improve shape similarity and demonstrate good generalization ability, while presenting competitive performance in the standard evaluation metrics.

Summary Despite the rapid growth and popularization of virtual reality (VR) applications, which have enabled new concepts for handling and solving existing problems through VR in various domains, practices related to software engineering have not kept up with this growth. Recent studies indicate that one of the topics that is still little explored in this area is software testing, as VR applications can be built for practically any type of purpose, making it difficult to generalize knowledge to be applied. In this paper, we present an approach that combines metamorphic testing, agent‐based testing and machine learning to test VR applications, focusing on finding collision and camera‐related faults. Our approach proposes the use of metamorphic relations to detect faults in collision and camera components in VR applications, as well as the use of intelligent agents for the automatic generation of test data. To evaluate the proposed approach, we conducted an experimental study on four VR applications, and the results showed an of the solution ranging from 93% to 69%, depending on the complexity of the application tested. We also discussed the feasibility of extending the approach to identify other types of faults in VR applications. In conclusion, we discussed important trends and opportunities that can benefit both academics and practitioners.

Breast cancer is responsible for approximately 15% of all cancer-related deaths among women worldwide, and early and accurate diagnosis increases the chances of survival. Over the last decades, several machine learning approaches have been used to improve the diagnosis of this disease, but most of them require a large set of samples for training. Syntactic approaches were barely used in this context, although it can present good results even if the training set has few samples. This article presents a syntactic approach to classify masses as benign or malignant. There were used features extracted from a polygonal representation of masses combined with a stochastic grammar approach to discriminate the masses found in mammograms. The results were compared with other machine learning techniques, and the grammar-based classifiers showed superior performance in the classification task. The best accuracies achieved were from 96% to 100%, indicating that grammatical approaches are robust and able to discriminate the masses even when trained with small samples of images. Syntactic approaches could be more frequently employed in the classification of masses, since they can learn the pattern of benign and malignant masses from a small sample of images achieving similar results when compared to the state of art.

Deep learning networks have obtained promising results for left ventricle segmentation in magnetic resonance images. Training with datasets containing low diversity of examples impacts the generalization ability of these networks, which is a problem in the medical image context. Strategies such as data augmentation and post-processing techniques have been applied to improve learning and generalization, but the influence of pre-processing methods is still underexplored. This work aims to evaluate the impact of pre-processing steps on the generalization of deep learning networks for left ventricle segmentation. Experiments were conducted with public datasets considering different configurations of pre-processing steps. We also compare a novel approach for automatic region of interest (ROI) detection in relation to manual approaches. The results suggest that pre-processing procedures favor the generalization, even when other strategies are not present. New methods to detect the ROI with size invariability are promising to obtain adequate generalization ability in deep learning networks.

The use of Eye Tracking (ET) has been investigated as an auxiliary mechanism to diagnose Autism Spectrum Disorder (ASD). One of the paradigms investigated using ET is Joint Attention (JA), which refers to moments when two individuals are focused on the same object/event so that both are aware that the focus of attention is shared. The computational tools that assist in the diagnosis of ASD have used Image Processing and Machine Learning techniques to process images, videos and ET signals. However, the JA paradigm is still little explored and presents challenges, as it requires analyzing the gaze trajectory and needs innovative approaches. The purpose of this article is to propose a model capable of extracting features from a video used as a stimulus to capture ET signals in order to verify JA and classify individuals as belonging to the ASD or Typical Development (TD) group. The main differential in relation to the approaches in the literature is the definition and implementation of the concept of floating Regions of Interest, which allows monitoring the gaze in relation to an object, considering its semantics, even if the object presents different characteristics throughout the video. A model based on ensembles of Random Forest classifiers was implemented to classify individuals as ASD or TD using the trajectory features extracted from the ET signals. The method reached 0.75 accuracy and 0.82 F1-score, indicating that the proposed approach, based on trajectory and JA, has the potential to be applied to assist in the diagnosis of ASD.

Computer-aided diagnosis using eye tracking data is classically based on regions of interest in the image. However, in recent years, the modeling of visual attention by saliency maps has shown better results. Wang et al., considering 3-layered saliency model that incorporated pixel-level, object-level, and semantic-level attributes, showed differences in the performance of eye tracking in autism spectrum disorder (ASD) and better characterized these differences by looking at which attributes were used, providing meaningful clinical results about the disorder. Our hypothesis is that the context interpretation would be worse according to the severity of ASD, consequently, the eye tracking data processed based on visual attention model (VAM) could be used to classify patients with ASD according to gravity. In this context, the present work proposes: 1) based on VAM, using Image Processing and Artificial Intelligence to learn a model for each group (severe and non-severe), from eye tracking data, and 2) a supervised classifier that, based on the models learned, performs the severity diagnosis. The classifier using the saliency maps was able to identify and separate the groups with an average accuracy of 88%. The most important features were the presence of face and skin color, in other words, semantic features.

Abstract Approximately 15% of all cancer deaths among women worldwide is due to breast cancer. Mammography is one of the most useful methods for the early detection of this disease. Over the last decade, several papers were published reporting the usage of different computer‐aided diagnosis systems using pattern recognition techniques as a second opinion to obtain a more accurate diagnosis. However, the theory of formal languages has not been explored in this field. In this context, the main contribution of this study is to present the usage of a new syntactic approach that is able to classify breast masses found in mammograms as benign or malignant. The experimental tests were performed using a dataset that contains 111 images from different sources. The grammar‐based classifiers achieved accuracy values ranging from 89% to 100% depending on the features and the model employed. Furthermore, to achieve a feature dimension reduction, a feature selection technique based on the Gini importance of each feature was employed. Additionally, we compared the obtained results with the grammar‐based classifiers to the more traditional classifiers used in this research area, such as artificial neural networks, support vector machines, k‐nearest neighbors, and random forest. The best result achieved by the grammar‐based classifiers was approximately 10% higher, in terms of accuracy, than the best results produced by the traditional classifiers, showing the strength of this grammatical approach.

Abstract As software systems have increased in complexity, manual testing has become harder or even infeasible. In addition, each test phase and application domain may have its idiosyncrasies in relation to testing automation. Techniques and tools to automate test oracles in domains such as graphical user interfaces are available; nevertheless, they are scarce in the virtual reality (VR) realm. We present an approach to automate software testing in VR-based systems with haptic interfaces—interfaces that allow bidirectional communication during human–computer interaction, capturing movements and providing touch feedback. It deals with the complexity and characteristics of haptic interfaces to apply the record and playback technique. Our approach also provides inference rules to identify possible faulty modules of the system under testing. A case study was performed with three systems: a system with primitive virtual objects, a dental anesthesia simulator and a game. Faulty versions of the systems were created by seeding faults manually and by using mutation operators. The results showed that 100% of the manually seeded faults and 93% of mutants were detected. Moreover, the inference rules helped identify the faulty modules of the systems, suggesting that the approach improves the test activity in VR-based systems with haptic interfaces.

Strategies aimed at keeping the user's interest in using computer applications are being studied to provide greater user engagement, and can influence how people interact with computers. One of the approaches that can promote user engagement is Affective Computing (AC), based on the premise of recognizing the user's emotional state and adjusting the computer application to respond to such state in real-time. Although it is a relatively new area, over the past few years many research works have investigated the use of AC in various activities and objectives. To provide an overview on the use of AC in computer applications, this article presents a systematic literature review based on available articles on the main scientific databases of the Computer Science area. The main contribution of this review is the analysis of different types of applications. Based on the 58 articles analyzed, the main emotion recognition techniques and approaches to the adaptation of computer applications, as well as the limitations and challenges to be overcome were compiled. Our conclusions present the limitations and challenges still to be overcome in the area of automatic adaptation of computer applications by means of AC.

Temporal data visualization supports planning and decision-making processes as it helps understanding patterns and relationships among time-based data. In the Healthcare area, the anamnesis procedure offers to physicians a large volume of valuable information, which is usually analyzed considering temporal aspects. Contributing to overcome the limited use of three-dimensional (3D) space, in this article we present a VR approach named 3D Block ARL to support interactive visualization of medical temporal data where the interface design is based on VA concepts. Additionally, we use a rule-based learning method to associate users' preferences to graphical elements aiming to personalize the proposed 3D visualization interface. Our results indicate that VA can be a valuable resource to improve the design of Information Visualization interface tools in the context of temporal medical data as well as to personalize the visualizations according to the preferences of users.

Breast cancer is one of the most common cancers that affect women worldwide being responsible for about 15% of all deaths related to cancer in the world. Mammography is one of the main techniques to help early detection of breast cancer. Although there are some characteristics that should be considered to discriminate benign and malignant masses, only about 15 to 30% of the cases sent to biopsies are malignant. To aid in the diagnosis of this disease, several CAD systems were proposed and developed to make a second opinion to the physicians, but the theory of formal languages is underexplored in this field. This paper presents a new syntactic approach to discriminate benign and malignant masses in digital mammography. Preliminary results showed that this approach is very promising, since our classifier achieved accuracies from 80% to 100% depending on the model and features used, applied on two different databases.

Breast cancer is one of the most common and deadliest cancers that affect mainly women worldwide, and mammography examination is one of the main tools to help early detection. Several papers have been published in the last decades reporting on techniques to automatically recognize breast cancer by analyzing mammograms. These techniques were used to create computer systems to help physicians and radiologists obtain a more precise diagnosis. The objective of this paper is to present an overview regarding the use of machine learning and pattern recognition techniques to discriminate masses in digitized mammograms. The main differences we found in the literature between the present paper and the other reviews are: 1) we used a systematic review method to create this survey; 2) we focused on mass classification problems; 3) the broad scope and spectrum used to investigate this theme, as 129 papers were analyzed to find out whether mass classification in mammograms is a problem solved. In order to achieve this objective, we performed a systematic review process to analyze papers found in the most important digital libraries in the area. We noticed that the three most common techniques used to classify mammographic masses are artificial neural network, support vector machine and k-nearest neighbors. Furthermore, we noticed that mass shape and texture are the most used features in classification, although some papers presented the usage of features provided by specialists, such as BI-RADS descriptors. Moreover, several feature selection techniques were used to reduce the complexity of the classifiers or to increase their accuracies. Additionally, the survey conducted points out some still unexplored research opportunities in this area, for example, we identified that some techniques such as random forest and logistic regression are little explored, while others, such as grammars or syntactic approaches, are not being used to perform this task.

Summary Software testing is regarded as an important method for fault revealing. Despite this advantage, it has been poorly used within the scope of virtual reality (VR) applications because they are highly complex and have peculiar features. Most testing performed of this VR applications are usability, which is conducted manually and only at final of the development process. Although some works try to propose criteria for this domain, there are no approaches that automatize the generation of test data from requirements specification in the VR domain. This paper proposes an approach called virtual reality—requirements specification and testing (VR‐ReST) to assist the requirements specification through a semiformal language and uses structural test criteria to generate test requirements and test data automatically for VR applications using scene graph concepts. The paper also examines the empirical results concerning the cost‐effectiveness of the approach for three different VR applications through two experiments. Mutation testing was used to evaluate effectiveness. We found that the approach achieved a high mutation score outperforming random testing, by 20%, on average. Our results also demonstrate that the approach is promising since it assists in writing and validating the requirements, as well as in reducing the risks of requirement specification by adopting a semiformal language.

Although many studies investigate the automatic adaptation in serious games with the goal to improve the users motivation, the majority of Affective Computing approaches requires a high development cost and usually does not consider the intervention of health professionals in adapting the game. This paper describes an approach to enable affective adaptation in serious games for motor rehabilitation with physiotherapists aid. Our approach consists of the definition and implementation of a framework. Its architecture reduces the development cost of a game with affective adaptation whilist enabling physiotherapists to configure adaptations in it according to patients profile. The results of an experiment with physiotherapists show that the system presents a high level of acceptance.

A clinical breast examination (CBE) is a thorough examination of the breast and the underarm area by trained healthcare professional to check for abnormalities. A CBE may be done if a woman finds a lump or change in her breasts, or as part of a woman's regular physical examination. A periodic screening with CBE and mammography results in reduced breast cancer mortality. A simulator for virtual training can help in the acquisition of skills to perform this procedure, but it has to simulate breast structures in a realistic way. Although some simulators are cited in the literature, they do not show the necessary attention towards the definition of the correct force feedback parameters for palpation simulation. The objective of this paper is to present the definition of the parameters for realistic representation of nodules in a virtual reality simulator for CBE. Based on nodule characteristics related to normal and abnormal cases, we defined equations aiming at calculating the appropriate force feedback to represent different cases configured by an instructor. Results of an assessment conducted by a physician indicated that the parameters set for each of the characteristics were considered realistic, with limitations in the representation of spiculated nodules.

Three-dimensional models have been used to aid medical diagnoses, using images generated by modalities like Magnetic Resonance Imaging. They can provide a more complete vision of objects since their depth is taken into account. Content-based Image Retrieval (CBIR) has also been used to aid the diagnosis. One important step in Three-dimensional CBIR (Model Retrieva) systems is the comparison between two models by using a set of features extracted and stored in a database. In this paper we present a novel method to compare two models, using the Bipartite graphs technique, with the aim to improve the retrieval precision. This technique retrieves 3D medical models of the left ventricle in order to aid the diagnosis of Congestive Heart Failure. Results showed that the novel method improved the precision by 10% when compared to the Similarity Function of Euclidean and Manhattan distance. These results confirmed that bipartite graph techniques can be used to improve the accuracy of Model Retrieval systems.

This paper describes a novel application of Content-Based Image Retrieval (CBIR) to search a medical database consisting of 3D models for diagnosis purposes. The 3D models, which are generated using Magnetic Resonance Imaging and include depth information, are used to search for similarity a database of 3D annotated medical cases using their pairwise feature similarity. The 3D models consist of both local and global feature descriptors that consider the surface of the 3D model and the overall geometry of the medical artifact. The models are then matched using spectral clustering that embeds the Euclidean distance for affinity and partitions the models into two groups, Congestive Heart Failure (CHF) and non-CHF. This suffices to demarcate using pairwise similarity the existence of CHF for the left ventricle. Experimental results using thirty 3D models show the utility of the new 3D method compared to existing methods. In particular, the novel method yields 83% overall accuracy.

The ability to process and identify facial emotions is an essential factor for an individuals social interaction. There are certain psychiatric disorders that can limit an individuals ability to recognize emotions in facial expressions. This problem could be confronted by making use of computational techniques in order to develop learning environments for the diagnosis, evaluation and training in identifying facial emotions. This paper presents an approach that uses image processing techniques, formal languages, anthropometry and Facial Action Coding System (FACS) to generate caricatures that represent facial movements related to neutral, satisfaction, sadness, anger, disgust, fear and surprise emotions. The rules that define the emotions were determined using an AND-OR graph to enable generating these images in a flexible manner. An evaluation conducted with healthy volunteers showed that some emotions are more easily recognized, while for other emotions the caricatures need to be further improved. This is a promising approach, since the parameters used provide flexibility to define the emotional intensity that must be represented.

Content-Based Image Retrieval (CBIR) systems constitute an innovative approach to store, to compare and to query images in a database. Visual aspects such as color, texture or shape are used to perform such operations. Recently, CBIR concepts were applied to build testing oracles for image processing programs, where test verdicts (approval/disapproval) are based on similarity measures between images produced by the program and reference images. However, the results of a CBIR system may vary depending on the components employed in the system (feature extractors and similarity functions), and few studies assessing this influence have been found in the literature. Our aim is to present an empirical analysis of ten similarity functions in CBIR systems within the context of software testing with graphic outputs. A case study with images obtained from a computer-aided diagnosis system in mammography indicated some variability among image test verdicts (approval/disapproval) according to the similarity function choice. The case study also indicates the existence of some clusters of similarity functions with high correlation coefficients.

In order to achieve the realism required in simulating the physical behavior of tissues during deformation, the methods which allow the use of physical parameters are the most widely employed. This paper presents a method of simulating the deformation of three-dimensional objects that represent human organs through multiple layers of tissue, which utilizes the following physical parameters: modulus of elasticity, thickness and density. The number of layers and the allocation of physical parameters can be set according to the different tissues that make up the human body and the behavior to be simulated. Thus, it is possible to achieve visual realism, realistic haptics and real-time interaction, with acceptable computational cost. For the simulation of medical procedures, both visual realism and haptic technology are required in order to provide the user with sensations similar to those found in real procedures. The organ chosen for the experiments was the breast, and the results obtained in relation to the physical behavior of deformation were acceptable in relation to visual realism and haptics.

Virtual Reality (VR) systems are a trend in the educational field. We can observe in academic literature that VR applications are widely adopted as Three-Dimensional Virtual Learning Environments (3D VLEs) in different fields of knowledge. However, one of the discussions highlighted in this context refers to the contribution that these environments really offer for student knowledge acquisition. In this scenario, this paper presents the results obtained from an experimental study conducted with high school students interested in learning Plane and Spatial Geometry. The volunteers explored the 3D VLE applied in the experiment, in which it is possible to create and visualize spatial figures from selected plane figures. The interactions in the virtual environment were registered for statistic analysis of the interaction level of participants. To assess the level of knowledge acquisition, a Theoretical Model for Assessment of Knowledge Acquisition was applied The results of the experiment conducted in this research indicated that it is possible to evaluate learning in 3D VLE by applying the Model abovementioned and showed that participants gained knowledge about the object of study using the 3D VLE learning method. Thus, it was possible to verify if the 3D VLEs are really contributing to the learner's knowledge acquisition.

Time and resource constraints should be taken into account in software testing activities, and thus optimizing the test suite is fundamental in the development process. In this context, the test case selection aims to eliminate redundant or unnecessary test data, which is crucial for the definition of test strategies. This paper presents a systematic review on the test case selection conducted through a selection of 449 articles published in leading journals and conferences in Computer Science. We addressed the state-of-art by collecting and comparing existing evidence on the methods used in the different software domains and the methods used to evaluate the test case selection. Our study identified 32 papers that met the research objectives, which featured 18 different selection methods and were evaluated through 71 case studies. The most commonly reported methods are adaptive random testing, genetic algorithms and greedy algorithm. Most approaches rely on heuristics, such as diversity of test cases and code or model coverage. This paper also discusses the key concepts and approaches, areas of application and evaluation metrics inherent to the methods of test case selection available in the literature.

An approach is presented in this paper, combining methods and models that are efficient enough to simulate elastic deformation, obtaining equilibrium between visual and haptic realism. Many medical training computational applications manipulate 3D objects that represent organs and human tissues. These representations, in function of the training requirements for which they are meant, may include parameter such as shape, topology, color, volume texture and, in certain cases, physical properties such as elasticity and stiffness. Based on this model, visual and/or hap-tic outputs are generated for users, and need to be realistic. In other words, they need to provide the learner with sensations close enough to those they would have if the training were provided with real life objects. However, its computational cost is too high to simultaneously provide visual and haptic realism in real-time. The results from deformation response time are compatible with those required for haptic interaction and the visual results from using meshes composed of a large number of polygons.

SUMMARY Automation of testing is an essential requirement to render it viable for software development. Although there are several testing techniques and criteria in many different domains, developing methods to test programs with complex outputs remains an unsolved challenge. This setting includes programs with graphical output, which produce images or interface windows. One possible approach towards automating the testing activity is the use of automatic oracles in which a reference image, taken as correct, can be used to establish a correctness measure in the tested program execution. A method that uses concepts of content‐based image retrieval to facilitate oracle automation in the domain of programs with graphics output is presented. Two case studies, one using a computer‐aided diagnostic system and one using a Web application, are presented, including some reflections and discussions that demonstrate the feasibility of the proposed approach. Copyright © 2011 John Wiley & Sons, Ltd.

Computer-aided diagnosis (CAD) schemes have been developed in many research centers to help the early detection of breast cancer. However, dense breast images are a challenge to CAD schemes due to the low contrast between structures of interest (such as microcalcifications-small size structures-which usually are associated to several breast tumors) and the background. This work describes a method to eliminate the background of a digitized mammogram image as well as two specific techniques to enhance the contrast in dense breast digital images as part of a CAD scheme under development in our group. The results indicate that these techniques can improve the performance of the scheme, and, thus, it can help in the early detection of breast cancer.