Single-Trial Decoding from Local Field Potential Using Bag of Word Representation

Dezfouli, Mohsen Parto; Daliri, Mohammad Reza

doi:10.1007/s10548-019-00726-8

Cited by 6 publications

(3 citation statements)

References 53 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…With the development of brain imaging technologies, such as Electroencephalography (EEG), the local field potential (LFP), Magnetoencephalography (MEG), and functional magnetic resonance imaging (fMRI), neural decoding is becoming a hot topic in brain‐computer interfaces (BCIs) (Ahmadi, Davoudi, Behroozi, & Daliri, 2020; Dezfouli & Daliri, 2020). Visual system is an important part of the human brain for understanding natural world.…”

Section: Introductionmentioning

confidence: 99%

A dual‐channel language decoding from brain activity with progressive transfer training

Huang

Yan

Cheng

et al. 2021

Human Brain Mapping

View full text Add to dashboard Cite

When we view a scene, the visual cortex extracts and processes visual information in the scene through various kinds of neural activities. Previous studies have decoded the neural activity into single/multiple semantic category tags which can caption the scene to some extent. However, these tags are isolated words with no grammatical structure, insufficiently conveying what the scene contains. It is well-known that textual language (sentences/phrases) is superior to single word in disclosing the meaning of images as well as reflecting people's real understanding of the images. Here, based on artificial intelligence technologies, we attempted to build a dual-channel language decoding model (DC-LDM) to decode the neural activities evoked by images into language (phrases or short sentences). The DC-LDM consisted of five modules, namely, Image-Extractor, Image-Encoder, Nerve-Extractor, Nerve-Encoder, and Language-Decoder. In addition, we employed a strategy of progressive transfer to train the DC-LDM for improving the performance of language decoding. The results showed that the texts decoded by DC-LDM could describe natural image stimuli accurately and vividly. We adopted six indexes to quantitatively evaluate the difference between the decoded texts and the annotated texts of corresponding visual images, and found that Word2vec-Cosine similarity (WCS) was the best indicator to reflect the similarity between the decoded and the annotated texts. In addition, among different visual cortices, we found that the text decoded by the higher visual cortex was more consistent with the description of the natural image than the lower one. Our decoding model may provide enlightenment in language-based brain-computer interface explorations.

show abstract

Section: Introductionmentioning

confidence: 99%

A dual‐channel language decoding from brain activity with progressive transfer training

Huang

Yan

Cheng

et al. 2021

Human Brain Mapping

View full text Add to dashboard Cite

show abstract

“…Thus, we can be fairly sure of the localization of a particular signal in the brain, however, the described variability of the responses still requires additional analytic approaches to further relate it to various internal or external influences that might underlie such variability. Although there are approaches developed for trial classification under similar conditions, they usually require training the classifier by presenting a set of stimuli having different parameters while local field potentials are recorded (e.g., Dezfouli and Daliri, 2020 ). However, such approaches are inapplicable to many situations when responses depend on the internally changing parameters, such the changes of the state of vigilance.…”

Section: Discussionmentioning

confidence: 99%

Assessment of a single trial impact on the amplitude of the averaged event related potentials

Fedorov

Levichkina

Limanskaya

et al. 2023

Front. Neural Circuits

View full text Add to dashboard Cite

Widely used in neuroscience the averaging of event related potentials is based on the assumption that small responses to the investigated events are present in every trial but can be hidden under the random noise. This situation often takes place, especially in experiments performed at hierarchically lower levels of sensory systems. However, in the studies of higher order complex neuronal networks evoked responses might appear only under particular conditions and be absent otherwise. We encountered this problem studying a propagation of interoceptive information to the cortical areas in the sleep-wake cycle. Cortical responses to various visceral events were present during some periods of sleep, then disappeared for a while and restored again after a period of absence. Further investigation of the viscero-cortical communication required a method that would allow labeling the trials contributing to the averaged event related responses–“efficient trials,” and separating them from the trials without any response. Here we describe a heuristic approach to solving this problem in the context of viscero-cortical interactions occurring during sleep. However, we think that the proposed technique can be applicable to any situation where neuronal processing of the same events is expected to be variable due to internal or external factors modulating neuronal activity. The method was first implemented as a script for Spike 2 program version 6.16 (CED). However, at present a functionally equivalent version of this algorithm is also available as Matlab code at https://github.com/george-fedorov/erp-correlations.

show abstract

“…Multi-electrode LFP recordings from one or more brain areas can be especially informative. For example, multi-channel LFPs may be used to decode visual stimuli [12,13], auditory stimuli [13,143], speech production [52,112,122], acute pain onset and intensity [165], and even semantic representations [39,97].…”

Section: Single-channel Neural Signalsmentioning

confidence: 99%

Identifying stimulus-driven neural activity patterns in multi-patient intracranial recordings

Manning¹

2022

Preprint

View full text Add to dashboard Cite

Identifying stimulus-driven neural activity patterns is critical for studying the neural basis of cognition. This can be particularly challenging in intracranial datasets, where electrode locations typically vary across patients. This chapter first presents an overview of the major challenges to identifying stimulus-driven neural activity patterns in the general case. Next, we will review several modality-specific considerations and approaches, along with a discussion of several issues that are particular to intracranial recordings. Against this backdrop, we will consider a variety of within-subject and across-subject approaches to identifying and modeling stimulus-driven neural activity patterns in multi-patient intracranial recordings. These approaches include generalized linear models, multivariate pattern analysis, representational similarity analysis, joint stimulus-activity models, hierarchical matrix factorization models, Gaussian process models, geometric alignment models, inter-subject correlations, and inter-subject functional correlations. Examples from the recent literature serve to illustrate the major concepts and provide the conceptual intuitions for each approach.

show abstract

Single-Trial Decoding from Local Field Potential Using Bag of Word Representation

Cited by 6 publications

References 53 publications

A dual‐channel language decoding from brain activity with progressive transfer training

A dual‐channel language decoding from brain activity with progressive transfer training

Assessment of a single trial impact on the amplitude of the averaged event related potentials

Identifying stimulus-driven neural activity patterns in multi-patient intracranial recordings

Contact Info

Product

Resources

About