Errikos M. Ventouras

Sleep spindle detection using artificial neural networks trained with filtered time-domain EEG: A feasibility study

An artificial neural network (ANN) based on the Multi-Layer Perceptron (MLP) architecture is used for detecting sleep spindles in band-pass filtered electroencephalograms (EEG), without feature extraction. Following optimum classification schemes, the sensitivity of the network ranges from 79.2% to 87.5%, while the false positive rate ranges from 3.8% to 15.5%. Furthermore, due to the operation of the ANN on time-domain EEG data, there is agreement with visual assessment concerning temporal resolution. Specifically, the total inter-spindle interval duration and the total duration of spindles are calculated with 99% and 92% accuracy, respectively. Therefore, the present method may be suitable for investigations of the dynamics among successive inter-spindle intervals, which could provide information on the role of spindles in the sleep process, and for studies of pharmacological effects on sleep structure, as revealed by the modification of total spindle duration.

Psychophysiological evidence for altered information processing in delusional misidentification syndromes

Recent research provides evidence that delusional misidentification syndromes (DMS) are associated with cognitive deficits. However, the underlying mechanisms of these deficits are not known. Since the P300 component of event-related potentials (ERPs) is related to fundamental aspects of working memory (WM), the present study is focused on P300 elicited during a WM test in DMS patients, as compared to those of healthy controls. Nine patients with DMS and 11 healthy controls, matched for age, sex and educational level were tested with a computerized version of the digit span test of the Wechsler batteries. Auditory ERPs were measured during the anticipatory period of the test. DMS patients showed significant reductions in P300 amplitude at the right frontal region compared to healthy controls. P300 latency in the central midline brain region was significantly prolonged in the DMS group. Each of these measures classified correctly 90% of the two groups. Moreover, the memory performance of the patient group was significantly lower, relatively to healthy controls. These findings provide evidence supporting the suggestion that DMS is associated with psychophysiological alterations occurring at the right frontal region, which mediates automatic processes, as well as with an irregular allocation of attentional resources, involving the interhemispheric circuitry, possibly due to gray matter degeneration. Finally, present work points to a need for further research investigating the characteristics, causes, course and treatment of severe cognitive deficits associated with DMS.

Psychophysiological differences in schizophrenics with and without delusional misidentification syndromes: a P300 study.

There is a debate on whether delusional misidentification syndromes (DMSs) and schizophrenia are distinct disorders. Information-processing deficits have been found in both. Since the P300 component of event-related potentials (ERPs) reflects attention and working memory (WM) mechanisms, the P300 elicited during a WM test was studied in schizophrenic patients with DMS in comparison to schizophrenic patients without DMS and controls. Nine schizophrenic patients with DMS, 11 without DMS and 11 healthy controls were tested with a computerized version of the digit span test of the Wechsler batteries. Auditory ERPs were measured during the anticipatory period of the test. P300 amplitude in prefrontal areas was found to be significantly reduced in schizophrenics without DMS and markedly less in DMS patients compared to controls. P300 latency in the central midline brain region was significantly prolonged in DMS patients compared to the other groups. Memory performance was significantly reduced in both patient groups as compared to healthy controls. The results may indicate abnormalities in both allocation of attentional resources and automatic orienting in schizophrenic patients with DSM. In contrast, even though schizophrenic patients without DMS exhibit partial similarities with patients suffering from DMS, they show excessive reduction of P300 amplitude located at the left frontal area. Future studies might clarify these issues.

Classification of Error-Related Negativity (ERN) and Positivity (Pe) potentials using kNN and Support Vector Machines

Error processing in subjects performing actions has been associated with the Event-Related Potential (ERP) components called Error-Related Negativity (ERN) and Error Positivity (Pe). In this paper, features based on statistical measures of the sample of averaged ERP recordings are used for classifying correct from incorrect actions. Three feature selection techniques were used and compared. Classification was done by means of a kNN and a Support Vector Machines (SVM) classifier. The use of a leave-one-out approach in the feature selection provided sensitivity and specificity values concurrently higher than or equal to 87.5%, for both classifiers. The classification results were significantly better for the time window that included only the ERN, as compared to time windows including also Pe.

Abnormal P300 in a case of delusional misidentification with coinciding Capgras and Frégoli symptoms.

The Delusional Misidentification Syndrome (DMS) is thought to be related to dissociation between recognition and identification processes. Working memory (WM) is considered responsible for the integration and online manipulation of information, so that it is available for further processing. Since the P300 component of event-related potentials (ERPs) is considered as an index of the on-line updating of WM, the present study is focused on auditory P300 elicited during a WM test in DMS, compared with that in healthy controls. ERPs, elicited during a WM test, in a case suffering from coinciding Capgras and Frégoli symptoms, were recorded. Peak amplitude and latency of the averaged P300 waveforms, as well as memory performance of this case, were compared to the patterns obtained from healthy controls. In relation to normal controls, the patient exhibited significantly attenuated amplitude of P300 at the F4, P3 and Pz abductions. The patient also showed significantly prolonged latencies of P300 at all abductions used. These findings suggest that DMS may be accompanied by WM dysfunction affecting brain regions outside the prefrontal cortex, as well as within, and by diffuse failure to allocate attention resources to a stimulus, as they are reflected by P300 amplitudes and latencies respectively. Additionally, it may be suggested that techniques designed to explore cognitive operations, such as recording of ERPs, and more specifically P300, during WM tasks, could provide further insights into the relationship between neural functioning and the cognitive deficits in DMS.

Effect of frequency deviance direction on performance and mismatch negativity.

The mismatch negativity (MMN) component of the auditory event-related potential is associated with automatic perceptual inference concerning changes in auditory stimulation. Recent studies have addressed the question whether performance and MMN is affected by the direction of frequency deviance. In the present study, the frequency MMN and performance is investigated during an auditory identification task. Specifically, we examined the effect of positive and negative differences between the present stimulus and the previous response frequencies on performance as well as on the characteristics of stimulus-locked ERPs and brain activation maps. The results show that frequency deviants creating mismatch conditions increase the likelihood of error commission. The decrease in performance achieves statistical significance in the case of positive frequency deviants. In the latter case, ERP amplitude values of the Fz electrode at 164 ms after stimulus onset are statistically larger for mismatch as opposed to no-mismatch condition. This corresponds to significance differences in the activation maps at Brodmann area 11, superior frontal gyrus, and the frontal lobe. The present findings revealed dissociations in behavioral and ERP responses in the processing of positive and negative frequency deviance, lending support to the notion that MMN is more sensitive to increments than to decrements in frequency.

Resting state and task related fMRI in small cell lung cancer patients

Prophylactic cranial irradiation (PCI) is a standard treatment technique for small cell lung cancer patients. However, there is evidence that this technique may contribute to neurocognitive deficits. Therefore the study of anatomical and functional connectivity in patients undergoing PCI as well as their neurocognitive functionality, depending on the type of disease and the phase of treatment and time of clinical examination, is of considerable interest. In this context, we investigate whether there are any differentiations in brain function during resting state and task-related functional magnetic resonance imaging (fMRI) in patients with cancer before PCI compared to healthy subjects. During a finger tapping task, the brain regions that were activated bilaterally for both groups are consistent with previous studies. During rest, the Default Mode Network (DMN) was identified in both groups. The preliminary results presented herein are subject to further investigation with larger patient and control group.

Behavioral and brain pattern differences between acting and observing in an auditory task.

BackgroundRecent research has shown that errors seem to influence the patterns of brain activity. Additionally current notions support the idea that similar brain mechanisms are activated during acting and observing. The aim of the present study was to examine the patterns of brain activity of actors and observers elicited upon receiving feedback information of the actors response.MethodsThe task used in the present research was an auditory identification task that included both acting and observing settings, ensuring concurrent ERP measurements of both participants. The performance of the participants was investigated in conditions of varying complexity. ERP data were analyzed with regards to the conditions of acting and observing in conjunction to correct and erroneous responses.ResultsThe obtained results showed that the complexity induced by cue dissimilarity between trials was a demodulating factor leading to poorer performance. The electrophysiological results suggest that feedback information results in different intensities of the ERP patterns of observers and actors depending on whether the actor had made an error or not. The LORETA source localization method yielded significantly larger electrical activity in the supplementary motor area (Brodmann area 6), the posterior cingulate gyrus (Brodmann area 31/23) and the parietal lobe (Precuneus/Brodmann area 7/5).ConclusionThese findings suggest that feedback information has a different effect on the intensities of the ERP patterns of actors and observers depending on whether the actor committed an error. Certain neural systems, including medial frontal area, posterior cingulate gyrus and precuneus may mediate these modulating effects. Further research is needed to elucidate in more detail the neuroanatomical and neuropsychological substrates of these systems.

Performance evaluation of an Artificial Neural Network automatic spindle detection system

Sleep spindles are transient waveforms found in the electroencephalogram (EEG) of non-rapid eye movement (NREM) sleep. Sleep spindles are used for the classification of sleep stages and have been studied in the context of various psychiatric and neurological disorders, such as Alzheimers disease (AD) and the so-called Mild Cognitive Impairment (MCI), which is considered to be a transitional stage between normal aging and dementia. The visual processing of wholenight sleep EEG recordings is tedious. Therefore, various techniques have been proposed for automatically detecting sleep spindles. In the present work an automatic sleep spindle detection system, that has been previously proposed, using a Multi-Layer Perceptron (MLP) Artificial Neural Network (ANN), is evaluated in detecting spindles of both healthy controls, as well as MCI and AD patients. An investigation is carried also concerning the visual detection process, taking into consideration the feedback information provided by the automatic detection system. Results indicate that the sensitivity of the detector was 81.4%, 62.2%, and 83.3% and the false positive rate was 34%, 11.5%, and 33.3%, for the control, MCI, and AD groups, respectively. The visual detection process had a sensitivity rate ranging from 46.5% to 60% and a false positive rate ranging from 4.8% to 19.2%.

Automatic identification of oculomotor behavior using pattern recognition techniques

In this paper, a methodological scheme for identifying distinct patterns of oculomotor behavior such as saccades, microsaccades, blinks and fixations from time series of eyes angular displacement is presented. The first step of the proposed methodology involves signal detrending for artifacts removal and estimation of eyes angular velocity. Then, feature vectors from fourteen first-order statistical features are formed from each angular displacement and velocity signal using sliding, fixed-length time windows. The obtained feature vectors are used for training and testing three artificial neural network classifiers, connected in cascade. The three classifiers discriminate between blinks and non-blinks, fixations and non-fixations and saccades and microsaccades, respectively. The proposed methodology was tested on a dataset from 1392 subjects, each performing three oculomotor fixation conditions. The average overall accuracy of the three classifiers, with respect to the manual identification of eye movements by experts, was 95.9%. The proposed methodological scheme provided better results than the well-known Velocity Threshold algorithm, which was used for comparison. The findings of the present study indicate that the utilization of pattern recognition techniques in the task of identifying the various eye movements may provide accurate and robust results.