An affective computing approach to physiological emotion specificity: toward subject-independent and stimulus-independent classification of film-induced emotions

The hypothesis of physiological emotion specificity has been tested using pattern classification analysis (PCA). To address limitations of prior research using PCA, we studied effects of feature selection (sequential forward selection, sequential backward selection), classifier type (linear and quadratic discriminant analysis, neural networks, k-nearest neighbors method), and cross-validation method (subject- and stimulus-(in)dependence). Analyses were run on a data set of 34 participants watching two sets of three 10-min film clips (fearful, sad, neutral) while autonomic, respiratory, and facial muscle activity were assessed. Results demonstrate that the three states can be classified with high accuracy by most classifiers, with the sparsest model having only five features, even for the most difficult task of identifying the emotion of an unknown subject in an unknown situation (77.5%). Implications for choosing PCA parameters are discussed.     

Feature binding and affect: emotional modulation of visuo-motor integration

The primate cortex represents the external world in a distributed fashion, which calls for a mechanism that integrates and binds the features of a perceived or processed event. Animal and patients studies provide evidence that feature binding in the visual cortex is driven by the muscarinic-cholinergic system, whereas visuo-motor integration may be under dopaminergic control. Consistent with this scenario, we present indication that the binding of visual and action features is modulated by emotions through the probable stimulation of the dopaminergic system. Interestingly, the impact of emotions on binding was restricted to tasks in which shape was task-relevant, suggesting that extracting affective information is not automatic but requires attention to shape.     

Early induced beta/gamma activity during illusory contour perception

The temporal binding hypothesis proposes that visual feature binding is achieved by neuronal synchronization. Nevertheless, the existing human neurophysiological evidence for the neuronal synchronization in visual feature binding—the oscillatory induced beta/gamma activity (IB/GA) is under suspicion. The previously observed IB/GA occurs at a later stage (after 200 ms), thus leading to the objection that IB/GA may be related to some later top-down processes rather than the early perceptual processing. However, the present EEG study identified an IB/GA as early as 90 ms after stimulus onset, which was stronger for a Kanizsa-type illusory contour (IC, a classic example of visual feature binding) than for a control stimulus. This finding provides new human evidence for the temporal binding hypothesis that neuronal synchronization occurs at the early stage of visual feature binding.     

A Wavelet Packet-Based Algorithm for the Extraction of Neural Rhythms

Neural rhythms are associated with different brain functions and pathological conditions. These rhythms are often clinically relevant for purposes of diagnosis or treatment, though their complex, time-varying features make them difficult to isolate. The wavelet packet transform has proven itself to be versatile and effective with respect to resolving signal features in both time and frequency. We propose a signal analysis technique, called neural rhythm extraction (NRE) that incorporates wavelet packet analysis along with a threshold-based scheme for separating rhythmic neural features from non-rhythmic ones. We applied NRE to rat in vitro intracellular recordings and human scalp electroencephalogram (EEG) signals, and were able to isolate and classify individual neural rhythms in signals containing large amplitude spikes and other artifacts. NRE is capable of discriminating signal features sharing similar time or frequency localization, as well as extracting low-amplitude, low-power rhythms otherwise masked by spectrally dominant signal components. The algorithm allows for independent retention and reconstruction of rhythmic features, which may serve to enhance other analysis techniques such as independent component analysis (ICA), and aid in application-specific tasks such as detection, classification or tracking.     

Goal-directed learning of features and forward models

The brain is able to perform actions based on an adequate internal representation of the world, where task-irrelevant features are ignored and incomplete sensory data are estimated. Traditionally, it is assumed that such abstract state representations are obtained purely from the statistics of sensory input for example by unsupervised learning methods. However, more recent findings suggest an influence of the dopaminergic system, which can be modeled by a reinforcement learning approach. Standard reinforcement learning algorithms act on a single layer network connecting the state space to the action space. Here, we involve in a feature detection stage and a memory layer, which together, construct the state space for a learning agent. The memory layer consists of the state activation at the previous time step as well as the previously chosen action. We present a temporal difference based learning rule for training the weights from these additional inputs to the state layer. As a result, the performance of the network is maintained both, in the presence of task-irrelevant features, and at randomly occurring time steps during which the input is invisible. Interestingly, a goal-directed forward model emerges from the memory weights, which only covers the state–action pairs that are relevant to the task. The model presents a link between reinforcement learning, feature detection and forward models and may help to explain how reward systems recruit cortical circuits for goal-directed feature detection and prediction.     

The cerebral hemispheres differ in their capacity for content-to-level binding but not for identification: Evidence from conjunction errors obtained with bilateral hierarchical stimuli

It is widely assumed that the right and left cerebral hemispheres are specialised for processing the global and local information of hierarchical stimuli, respectively. This idea has further been specified in the content-to-level binding theory (Hübner & Volberg, 2005) by stating that the hemispheres differ in their efficiency for binding the contents of a stimulus to their respective level. In contrast, it is assumed that the hemispheres do not differ in their capacity for the identification of the information at the two levels. This latter hypothesis was tested in the present experiment by presenting a hierarchical letter to each visual field. As expected, there were visual field effects only for errors involving the erroneous binding between a letter and a level. For errors that result from the mislocalisation of a letter, there were no visual field effects. Together, the data support the hypothesis that the hemispheres do not differ in their identification capacity.     

A Scientific Software Product Line for the Bioinformatics domain

ContextMost specialized users (scientists) that use bioinformatics applications do not have suitable training on software development. Software Product Line (SPL) employs the concept of reuse considering that it is defined as a set of systems that are developed from a common set of base artifacts. In some contexts, such as in bioinformatics applications, it is advantageous to develop a collection of related software products, using SPL approach. If software products are similar enough, there is the possibility of predicting their commonalities, differences and then reuse these common features to support the development of new applications in the bioinformatics area.ObjectivesThis paper presents the PL-Science approach which considers the context of SPL and ontology in order to assist scientists to define a scientific experiment, and to specify a workflow that encompasses bioinformatics applications of a given experiment. This paper also focuses on the use of ontologies to enable the use of Software Product Line in biological domains.MethodIn the context of this paper, Scientific Software Product Line (SSPL) differs from the Software Product Line due to the fact that SSPL uses an abstract scientific workflow model. This workflow is defined according to a scientific domain and using this abstract workflow model the products (scientific applications/algorithms) are instantiated.ResultsThrough the use of ontology as a knowledge representation model, we can provide domain restrictions as well as add semantic aspects in order to facilitate the selection and organization of bioinformatics workflows in a Scientific Software Product Line. The use of ontologies enables not only the expression of formal restrictions but also the inferences on these restrictions, considering that a scientific domain needs a formal specification.ConclusionsThis paper presents the development of the PL-Science approach, encompassing a methodology and an infrastructure, and also presents an approach evaluation. This evaluation presents case studies in bioinformatics, which were conducted in two renowned research institutions in Brazil.     

Iris recognition based on key image feature extraction

In iris recognition, feature extraction can be influenced by factors such as illumination and contrast, and thus the features extracted may be unreliable, which can cause a high rate of false results in iris pattern recognition. In order to obtain stable features, an algorithm was proposed in this paper to extract key features of a pattern from multiple images. The proposed algorithm built an iris feature template by extracting key features and performed iris identity enrolment. Simulation results showed that the selected key features have high recognition accuracy on the CASIA Iris Set, where both contrast and illumination variance exist.     

Neuro-fuzzy feature evaluation with theoretical analysis

The article provides a fuzzy set theoretic feature evaluation index and a connectionist model for its evaluation along with their theoretical analysis. A concept of weighted membership function is introduced which makes the modeling of the class structures more appropriate. A neuro-fuzzy algorithm is developed for determining the optimum weighting coefficients representing the feature importance. It is shown theoretically that the evaluation index has a fixed upper bound and a varying lower bound, and it monotonically increases with the lower bound. A relation between the evaluation index, interclass distance and weighting coefficients is established. Effectiveness of the algorithms for evaluating features both individually and in a group (considering their independence and dependency) is demonstrated along with comparisons on speech, Iris, medical and mango-leaf data. The results are also validated using scatter diagram and k-NN classifier.     

Exudate-based diabetic macular edema detection in fundus images using publicly available datasets

Diabetic macular edema (DME) is a common vision threatening complication of diabetic retinopathy. In a large scale screening environment DME can be assessed by detecting exudates (a type of bright lesions) in fundus images. In this work, we introduce a new methodology for diagnosis of DME using a novel set of features based on colour, wavelet decomposition and automatic lesion segmentation. These features are employed to train a classifier able to automatically diagnose DME through the presence of exudation. We present a new publicly available dataset with ground-truth data containing 169 patients from various ethnic groups and levels of DME. This and other two publicly available datasets are employed to evaluate our algorithm. We are able to achieve diagnosis performance comparable to retina experts on the MESSIDOR (an independently labelled dataset with 1200 images) with cross-dataset testing (e.g., the classifier was trained on an independent dataset and tested on MESSIDOR). Our algorithm obtained an AUC between 0.88 and 0.94 depending on the dataset/features used. Additionally, it does not need ground truth at lesion level to reject false positives and is computationally efficient, as it generates a diagnosis on an average of 4.4 s (9.3 s, considering the optic nerve localisation) per image on an 2.6 GHz platform with an unoptimised Matlab implementation.