An official website of the United States government
Objectively differentiating patient mental states based on electrical activity, as opposed to overt behavior, is a fundamental neuroscience problem with medical applications, such as identifying patients in locked-in state vs. coma. Electroencephalography (EEG), which detects millisecond-level changes in brain activity across a range of frequencies, allows for assessment of external stimulus processing by the brain in a non-invasive manner. We applied machine learning methods to 26-channel EEG data of 24 fluent Deaf signers watching videos of sign language sentences (comprehension condition), and the same videos reversed in time (non-comprehension condition), to objectively separate vision-based high-level cognition states. While spectrotemporal parameters of the stimuli were identical in comprehension vs. non-comprehension conditions, the neural responses of participants varied based on their ability to linguistically decode visual data. We aimed to determine which subset of parameters (specific scalp regions or frequency ranges) would be necessary and sufficient for high classification accuracy of comprehension state. Optical flow, characterizing distribution of velocities of objects in an image, was calculated for each pixel of stimulus videos using MATLAB Vision toolbox. Coherence between optical flow in the stimulus and EEG neural response (per video, per participant) was then computed using canonical component analysis with NoiseTools toolbox. Peak correlations were extracted for each frequency for each electrode, participant, and video. A set of standard ML algorithms were applied to the entire dataset (26 channels, frequencies from .2 Hz to 12.4 Hz, binned in 1 Hz increments), with consistent out-of-sample 100% accuracy for frequencies in .2-1 Hz range for all regions, and above 80% accuracy for frequencies < 4 Hz. Sparse Optimal Scoring (SOS) was then applied to the EEG data to reduce the dimensionality of the features and improve model interpretability. SOS with elastic-net penalty resulted in out-of-sample classification accuracy of 98.89%. The sparsity pattern in the model indicated that frequencies between 0.2–4 Hz were primarily used in the classification, suggesting that underlying data may be group sparse. Further, SOS with group lasso penalty was applied to regional subsets of electrodes (anterior, posterior, left, right). All trials achieved greater than 97% out-of-sample classification accuracy. The sparsity patterns from the trials using 1 Hz bins over individual regions consistently indicated frequencies between 0.2–1 Hz were primarily used in the classification, with anterior and left regions performing the best with 98.89% and 99.17% classification accuracy, respectively. While the sparsity pattern may not be the unique optimal model for a given trial, the high classification accuracy indicates that these models have accurately identified common neural responses to visual linguistic stimuli. Cortical tracking of spectro-temporal change in the visual signal of sign language appears to rely on lower frequencies proportional to the N400/P600 time-domain evoked response potentials, indicating that visual language comprehension is grounded in predictive processing mechanisms.
more »
« less
Machine Learning Classification of Event-Related Brain Potentials during a Visual Go/NoGo Task
Machine learning (ML) methods are increasingly being applied to analyze biological signals. For example, ML methods have been successfully applied to the human electroencephalogram (EEG) to classify neural signals as pathological or non-pathological and to predict working memory performance in healthy and psychiatric patients. ML approaches can quickly process large volumes of data to reveal patterns that may be missed by humans. This study investigated the accuracy of ML methods at classifying the brain’s electrical activity to cognitive events, i.e., event-related brain potentials (ERPs). ERPs are extracted from the ongoing EEG and represent electrical potentials in response to specific events. ERPs were evoked during a visual Go/NoGo task. The Go/NoGo task requires a button press on Go trials and response withholding on NoGo trials. NoGo trials elicit neural activity associated with inhibitory control processes. We compared the accuracy of six ML algorithms at classifying the ERPs associated with each trial type. The raw electrical signals were fed to all ML algorithms to build predictive models. The same raw data were then truncated in length and fitted to multiple dynamic state space models of order nx using a continuous-time subspace-based system identification algorithm. The 4nx numerator and denominator parameters of the transfer function of the state space model were then used as substitutes for the data. Dimensionality reduction simplifies classification, reduces noise, and may ultimately improve the predictive power of ML models. Our findings revealed that all ML methods correctly classified the electrical signal associated with each trial type with a high degree of accuracy, and accuracy remained high after parameterization was applied. We discuss the models and the usefulness of the parameterization.
more »
« less
- Award ID(s):
- 1632377
- PAR ID:
- 10507215
- Publisher / Repository:
- MDPI
- Date Published:
- Journal Name:
- Entropy
- Volume:
- 26
- Issue:
- 3
- ISSN:
- 1099-4300
- Page Range / eLocation ID:
- 220
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Abstract Prior research has supported some aspects of a theorized prejudice self-regulation model. We provide the first test of the full model-based process of bias regulation as it unfolds in real time. Event-related potentials (ERPs) were recorded from White undergraduates at two large American universities (N = 130; 40% female) during a racial stereotype priming task. Attention to Black male face primes, indexed by the P2 ERP, increased following self-regulation failures. In turn, within-person, trial-to-trial variability in attention to Black male faces predicted variability in bias expression. The latter effect was moderated by individual differences in internal motivation to respond without prejudice (IMS). Specifically, among lower-IMS individuals, trials in which Black faces elicited relatively larger P2 amplitudes (relative to an individual’s own average P2 amplitude) were associated with increased behavioral race bias. In contrast, and consistent with theory, among higher-IMS individuals trials in which Black faces elicited larger relative P2 amplitudes were associated with decreased bias. Findings provide direct evidence supporting the temporal sequencing of race-bias regulation and identify within-person variability in attention to race as a potential mechanism for determining when and in whom bias will be regulated.more » « less
-
The prospect of electroencephalography (EEG) and functional near-infrared spectroscopy (fNIRS) in the presence of topological information of participants is often left unexplored in most of the brain-computer interface (BCI) systems. Additionally, the usage of these modalities together in the field of multimodality analysis to support multiple brain signals toward improving BCI performance is not fully examined. This study first presents a multimodal data fusion framework to exploit and decode the complementary synergistic properties in multimodal neural signals. Moreover, the relations among different subjects and their observations also play critical roles in classifying unknown subjects. We developed a context-aware graph neural network (GNN) model utilizing the pairwise relationship among participants to investigate the performance on an auditory task classification. We explored standard and deviant auditory EEG and fNIRS data where each subject was asked to perform an auditory oddball task and has multiple trials regarded as context-aware nodes in our graph construction. In experiments, our multimodal data fusion strategy showed an improvement up to 8.40% via SVM and 2.02% via GNN, compared to the single-modal EEG or fNIRS. In addition, our context-aware GNN achieved 5.3%, 4.07% and 4.53% higher accuracy for EEG, fNIRS and multimodal data based experiments, compared to the baseline models.more » « less
-
Abstract Electroencephalogram (EEG) alpha power (8–13 Hz) is a characteristic of various creative task conditions and is involved in creative ideation. Alpha power varies as a function of creativity-related task demands. This study investigated the event-related potentials (ERPs), alpha power activation, and potential machine learning (ML) to classify the neural responses of engineering students involved with creativity task. All participants performed a modified alternate uses task (AUT), in which participants categorized functions (or uses) for everyday objects as either creative, nonsense, or common. At first, this study investigated the fundamental ERPs over central and parietooccipital temporal areas. The bio-responses to understand creativity in engineering students demonstrates that nonsensical and creative stimuli elicit larger N400 amplitudes (−1.107 mV and −0.755 mV, respectively) than common uses (0.0859 mV) on the 300–500 ms window. N400 effect was observed on 300–500 ms window from the grand average waveforms of each electrode of interest. ANOVA analysis identified a significant main effect: decreased alpha power during creative ideation, especially over (O1/2, P7/8) parietooccipital temporal area. Machine learning is used to classify the specific temporal area data’s neural responses (creative, nonsense, and common). A k-nearest neighbors (kNN) classifier was used, and results were evaluated in terms of accuracy, precision, recall, and F1- score using the collected datasets from the participants. With an overall 99.92% accuracy and area under the curve at 0.9995, the kNN classifier successfully classified the participants’ neural responses. These results have great potential for broader adaptation of machine learning techniques in creativity research.more » « less
-
Gadekallu, Thippa Reddy (Ed.)As of March 30 2021, over 5,193 COVID-19 clinical trials have been registered through Clinicaltrial.gov. Among them, 191 trials were terminated, suspended, or withdrawn (indicating the cessation of the study). On the other hand, 909 trials have been completed (indicating the completion of the study). In this study, we propose to study underlying factors of COVID-19 trial completion vs . cessation, and design predictive models to accurately predict whether a COVID-19 trial may complete or cease in the future. We collect 4,441 COVID-19 trials from ClinicalTrial.gov to build a testbed, and design four types of features to characterize clinical trial administration, eligibility, study information, criteria, drug types, study keywords, as well as embedding features commonly used in the state-of-the-art machine learning. Our study shows that drug features and study keywords are most informative features, but all four types of features are essential for accurate trial prediction. By using predictive models, our approach achieves more than 0.87 AUC (Area Under the Curve) score and 0.81 balanced accuracy to correctly predict COVID-19 clinical trial completion vs . cessation. Our research shows that computational methods can deliver effective features to understand difference between completed vs . ceased COVID-19 trials. In addition, such models can also predict COVID-19 trial status with satisfactory accuracy, and help stakeholders better plan trials and minimize costs.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Journal Article:
- https://doi.org/10.3390/e26030220
