Predicting Premature Video Skipping and Viewer Interest from EEG Recordings
Ontology highlight
ABSTRACT:
SUBMITTER: Libert A
PROVIDER: S-EPMC7514236 | biostudies-literature | 2019 Oct
REPOSITORIES: biostudies-literature
Similar Datasets
Project description:We performed simultaneous recordings of electroencephalography (EEG) from multiple students in a classroom, and measured the inter-subject correlation (ISC) of activity evoked by a common video stimulus. The neural reliability, as quantified by ISC, has been linked to engagement and attentional modulation in earlier studies that used high-grade equipment in laboratory settings. Here we reproduce many of the results from these studies using portable low-cost equipment, focusing on the robustness of using ISC for subjects experiencing naturalistic stimuli. The present data shows that stimulus-evoked neural responses, known to be modulated by attention, can be tracked for groups of students with synchronized EEG acquisition. This is a step towards real-time inference of engagement in the classroom.
| S-EPMC5339684 | biostudies-literature
Project description:Automated seizure detection in long-term video-EEG recordings is far from being integrated into common clinical practice. Here, we leverage classical and state-of-the-art complexity measures to robustly and automatically detect seizures from scalp recordings. Brain activity is scored through eight features, encompassing traditional time domain and novel measures of recurrence. A binary classification algorithm tailored to treat unbalanced dataset is used to determine whether a time window is ictal or non-ictal from its features. The application of the algorithm on a cohort of ten adult patients with focal refractory epilepsy indicates sensitivity, specificity, and accuracy of 90%, along with a true alarm rate of 95% and less than four false alarms per day. The proposed approach emphasizes ictal patterns against noisy background without the need of data preprocessing. Finally, we benchmark our approach against previous studies on two publicly available datasets, demonstrating the good performance of our algorithm.
| S-EPMC7811137 | biostudies-literature
Project description:Despite considerable effort to remove it, the ballistocardiogram (BCG) remains a major artifact in electroencephalographic data (EEG) acquired inside magnetic resonance imaging (MRI) scanners, particularly in continuous (as opposed to event-related) recordings. In this study, we have developed a new Direct Recording Prior Encoding (DRPE) method to extract and separate the BCG and EEG components from contaminated signals, and have demonstrated its performance by comparing it quantitatively to the popular Optimal Basis Set (OBS) method. Our modified recording configuration allows us to obtain representative bases of the BCG- and EEG-only signals. Further, we have developed an optimization-based reconstruction approach to maximally incorporate prior knowledge of the BCG/EEG subspaces, and of the signal characteristics within them. Both OBS and DRPE methods were tested with experimental data, and compared quantitatively using cross-validation. In the challenging continuous EEG studies, DRPE outperforms the OBS method by nearly sevenfold in separating the continuous BCG and EEG signals.
| S-EPMC4067090 | biostudies-other
Project description:IntroductionCinema is an important part of modern culture, influencing millions of viewers. Research suggested many models for the prediction of film success, one of them being the use of neuroscientific tools. The aim of our study was to find physiological markers of viewer perception and correlate them to short film ratings given by our subjects. Short films are used as a test case for directors and screenwriters and can be created to raise funding for future projects; however, they have not been studied properly with physiological methods.MethodsWe recorded electroencephalography (18 sensors), facial electromyography (corrugator supercilii and zygomaticus major), photoplethysmography, and skin conductance in 21 participants while watching and evaluating 8 short films (4 dramas and 4 comedies). Also, we used machine learning (CatBoost, SVR) to predict the exact rating of each film (from 1 to 10), based on all physiological indicators. In addition, we classified each film as low or high rated by our subjects (with Logistic Regression, KNN, decision tree, CatBoost, and SVC).ResultsThe results showed that ratings did not differ between genres. Corrugator supercilii activity ("frowning" muscle) was larger when watching dramas; whereas zygomaticus major ("smiling" muscle) activity was larger during the watching of comedies. Of all somatic and vegetative markers, only zygomaticus major activity, PNN50, SD1/SD2 (heart rate variability parameters) positively correlated to the film ratings. The EEG engagement indices, beta/(alpha+theta) and beta/alpha correlated positively with the film ratings in the majority of sensors. Arousal (betaF3 + betaF4)/(alphaF3 + alphaF4), and valence (alphaF4/betaF4) - (alphaF3/betaF3) indices also correlated positively to film ratings. When we attempted to predict exact ratings, MAPE was 0.55. As for the binary classification, logistic regression yielded the best values (area under the ROC curve = 0.62) than other methods (0.51-0.60).DiscussionOverall, we revealed EEG and peripheral markers, which reflect viewer ratings and can predict them to a certain extent. In general, high film ratings can reflect a fusion of high arousal and different valence, positive valence being more important. These findings broaden our knowledge about the physiological basis of viewer perception and can be potentially used at the stage of film production.
| S-EPMC10291053 | biostudies-literature
Project description:Wireless electroencephalography (EEG) of small animal subjects typically utilizes miniaturized EEG devices which require a robust recording and electrode assembly that remains in place while also being well-tolerated by the animal so as not to impair the ability of the animal to perform normal living activities or experimental tasks. We developed simple and fast electrode assembly and method of electrode implantation using electrode wires and wire-wrap technology that provides both higher survival and success rates in obtaining recordings from the electrodes than methods using screws as electrodes. The new wire method results in a 51% improvement in the number of electrodes that successfully record EEG signal. Also, the electrode assembly remains affixed and provides EEG signal for at least a month after implantation. Screws often serve as recording electrodes, which require either drilling holes into the skull to insert screws or affixing screws to the surface of the skull with adhesive. Drilling holes large enough to insert screws can be invasive and damaging to brain tissue, using adhesives may interfere with conductance and result in a poor signal, and soldering screws to wire leads results in fragile connections. The methods presented in this article provide a robust implant that is minimally invasive and has a significantly higher success rate of electrode implantation. In addition, the implant remains affixed and produces good recordings for over a month, while using economical, easily obtained materials and skills readily available in most animal research laboratories.
| S-EPMC5694541 | biostudies-literature
Project description:Neonatal seizures are a common emergency in the neonatal intensive care unit (NICU). There are many questions yet to be answered regarding the temporal/spatial characteristics of seizures from different pathologies, response to medication, effects on neurodevelopment and optimal detection. The dataset presented in this descriptor contains EEG recordings from human neonates, the visual interpretation of the EEG by the human experts, supporting clinical data and codes to assist access. Multi-channel EEG was recorded from 79 term neonates admitted to the NICU at the Helsinki University Hospital. The median recording duration was 74 min (IQR: 64 to 96 min). The presence of seizures in the EEGs was annotated independently by three experts. An average of 460 seizures were annotated per expert in the dataset; 39 neonates had seizures and 22 were seizure free, by consensus. The dataset can be used as a reference set of neonatal seizures, in studies of inter-observer agreement and for the development of automated methods of seizure detection and other EEG analyses.
| S-EPMC6400100 | biostudies-literature
Project description:BackgroundInternational consensus on best practices for calculating and reporting vestibular function is lacking. Quantitative vestibulo-ocular reflex (VOR) gain using a video head impulse test (HIT) device can be calculated by various methods.ObjectiveTo compare different gain calculation methods and to analyze interactions between artifacts and calculation methods.MethodsWe analyzed 1300 horizontal HIT traces from 26 patients with acute vestibular syndrome and calculated the ratio between eye and head velocity at specific time points (40 ms, 60 ms) after HIT onset ('velocity gain'), ratio of velocity slopes ('regression gain'), and ratio of area under the curves after de-saccading ('position gain').ResultsThere was no mean difference between gain at 60 ms and position gain, both showing a significant correlation (r2 = 0.77, p < 0.001) for artifact-free recordings. All artifacts reduced high, normal-range gains modestly (range -0.06 to -0.11). The impact on abnormal, low gains was variable (depending on the artifact type) compared to artifact-free recordings.ConclusionsThere is no clear superiority of a single gain calculation method for video HIT testing. Artifacts cause small but significant reductions of measured VOR gains in HITs with higher, normal-range gains, regardless of calculation method. Artifacts in abnormal HITs with low gain increased measurement noise. A larger number of HITs should be performed to confirm abnormal results, regardless of calculation method.
| S-EPMC9037838 | biostudies-literature
Project description:ObjectiveWe sought to test the performance of three strategies for binary classification (logistic regression, support vector machines, and deep learning) for the problem of predicting successful episodic memory encoding using direct brain recordings obtained from human stereo EEG subjects. We also sought to test the impact of applying t-distributed stochastic neighbor embedding (tSNE) for unsupervised dimensionality reduction, as well as testing the effect of reducing input features to a core set of memory relevant brain areas. This work builds upon published efforts to develop a closed-loop stimulation device to improve memory performance.ApproachWe used a unique data set consisting of 30 stereo EEG patients with electrodes implanted into a core set of five common brain regions (along with other areas) who performed the free recall episodic memory task as brain activity was recorded. Using three different machine learning strategies, we trained classifiers to predict successful versus unsuccessful memory encoding and compared the difference in classifier performance (as measured by the AUC) at the subject level and in aggregate across modalities. We report the impact of feature reduction on the classifiers, including reducing the number of input brain regions, frequency bands, and the impact of tSNE.ResultsDeep learning classifiers outperformed both support vector machines (SVM) and logistic regression (LR). A priori selection of core brain regions also improved classifier performance for LR and SVM models, especially when combined with tSNE.SignificanceWe report for the first time a direct comparison among traditional and deep learning methods of binary classification to the problem of predicting successful memory encoding using human brain electrophysiological data. Our findings will inform the design of brain machine interface devices to affect memory processing.
| S-EPMC9020643 | biostudies-literature
Project description:We describe driver behaviour and brain dynamics acquired from a 90-minute sustained-attention task in an immersive driving simulator. The data included 62 sessions of 32-channel electroencephalography (EEG) data for 27 subjects driving on a four-lane highway who were instructed to keep the car cruising in the centre of the lane. Lane-departure events were randomly induced to cause the car to drift from the original cruising lane towards the left or right lane. A complete trial included events with deviation onset, response onset, and response offset. The next trial, in which the subject was instructed to drive back to the original cruising lane, began 5-10?seconds after finishing the previous trial. We believe that this dataset will lead to the development of novel neural processing methodology that can be used to index brain cortical dynamics and detect driving fatigue and drowsiness. This publicly available dataset will be beneficial to the neuroscience and brain-computer interface communities.
| S-EPMC6472414 | biostudies-literature
Project description:BackgroundTo probe the functional role of brain oscillations, transcranial alternating current stimulation (tACS) has proven to be a useful neuroscientific tool. Because of the excessive tACS-caused artifact at the stimulation frequency in electroencephalography (EEG) signals, tACS + EEG studies have been mostly limited to compare brain activity between recordings before and after concurrent tACS. Critically, attempts to suppress the artifact in the data cannot assure that the entire artifact is removed while brain activity is preserved. The current study aims to evaluate the feasibility of specific artifact correction techniques to clean tACS-contaminated EEG data.New methodIn the first experiment, we used a phantom head to have full control over the signal to be analyzed. Driving pre-recorded human brain-oscillation signals through a dipolar current source within the phantom, we simultaneously applied tACS and compared the performance of different artifact-correction techniques: sine subtraction, template subtraction, and signal-space projection (SSP). In the second experiment, we combined tACS and EEG on one human subject to demonstrate the best-performing data-correction approach in a proof of principle.ResultsThe tACS artifact was highly attenuated by SSP in the phantom and the human EEG; thus, we were able to recover the amplitude and phase of the oscillatory activity. In the human experiment, event-related desynchronization could be restored after correcting the artifact.Comparison with existing methodsThe best results were achieved with SSP, which outperformed sine subtraction and template subtraction.ConclusionOur results demonstrate the feasibility of SSP by applying it to a phantom measurement with pre-recorded signal and one human tACS + EEG dataset. For a full validation of SSP, more data are needed.
| S-EPMC7775555 | biostudies-literature