Brain–machine interfaces from motor to mood

Shanechi, Maryam M.

doi:10.1038/s41593-019-0488-y

Cited by 182 publications

(183 citation statements)

References 98 publications

Supporting

Mentioning

175

Contrasting

Order By: Relevance

“…What is more, EEG is a non-stationary signal-its statistics varying over time [1,5,6]. This is especially problematic for online, real-time analysis, where it is inherently models that were trained on past neural data that are used to decode present neural activity [7,8]. Further, for complex machine-learning models, model training time might be lengthy.…”

Section: Introductionmentioning

confidence: 99%

Data augmentation for deep-learning-based electroencephalography

Lashgari

Liang

Maoz

2020

Journal of Neuroscience Methods

253

140

View full text Add to dashboard Cite

Highlights • Data augmentation (DA) is increasingly used with deep learning (DL) on EEG • It enhances decoding accuracy left unexplained by 29% on average on the datasets we review • We analyze which specific DA techniques appear to work best for which EEG tasks • We tested various DA techniques on an open motor-imagery task and compared the accuracy gains to demonstrate the usefulness of DA for DL-based EEG analysis • We propose guidelines for reporting parameters for different DA techniques Abstract-Background Data augmentation (DA) has recently been demonstrated to achieve considerable performance gains for deep learning (DL)-increased accuracy and stability and reduced overfitting. Some electroencephalography (EEG) tasks suffer from low samples-to-features ratio, severely reducing DL effectiveness. DA with DL thus holds transformative promise for EEG processing, possibly like DL revolutionized computer vision, etc.-New method We review trends and approaches to DA for DL in EEG to address: Which DA approaches exist and are common for which EEG tasks? What input features are used? And, what kind of accuracy gain can be expected?-Results DA for DL on EEG begun 5 years ago and is steadily used more. We grouped DA techniques (noise addition, generative adversarial networks, sliding windows, sampling, Fourier transform, recombination of segmentation, and others) and EEG tasks (into seizure detection, sleep stages, motor imagery, mental workload, emotion recognition, motor tasks, and visual tasks). DA efficacy across techniques varied considerably. Noise addition and sliding windows provided the highest accuracy boost; mental workload most benefitted from DA. Sliding window, noise addition, and sampling methods most common for seizure detection, mental workload, and sleep stages, respectively.-Comparing with existing methods Percent of decoding accuracy explained by DA beyond unaugmented accuracy varied between 8% for recombination of segmentation and 36% for noise addition and from 14% for motor imagery to 56% for mental workload-29% on average.-Conclusions DA increasingly used and considerably improved DL decoding accuracy on EEG. Additional publications-if adhering to our reporting guidelines-will facilitate more detailed analysis.

show abstract

Section: Introductionmentioning

confidence: 99%

Data augmentation for deep-learning-based electroencephalography

Lashgari

Liang

Maoz

2020

Journal of Neuroscience Methods

253

140

View full text Add to dashboard Cite

show abstract

“…These results, therefore, could provide an informed basis for the use of intermittent and targeted neuromodulation to aid individuals experiencing severe emotion dysregulation at both extremes of the spectrum. Indeed, this approach, applied to other domains and across cognitive and emotional tasks, could allow us to arrive at a more refined view of how to use neural stimulation to therapeutically alter the circuitry underlying important domains of functioning, such as maladaptive emotional processing and decision making, with implications for a wide array of neuropsychiatric diseases (33,34,51,52).…”

Section: Resultsmentioning

confidence: 99%

“…Second, after establishing a subset of behavioral models which were high-performing, in separate task sessions, we tested whether identified hidden states within these behavioral state space models can be driven by stimulation in specific brain networks. We hypothesized that direct electrical intracranial stimulation in different brain regions would have differential, and causal, effects on these behavioral features as has been hinted at with stimulation in other brain regions in learning and memory (50,51), mood (35,52), and OCD (36,53,54). As a final test, we used the state-space model in closed-loop adaptive stimulation to modulate behavior predictably.…”

Section: Introductionmentioning

confidence: 99%

Bidirectional modulation of human emotional conflict resolution using intracranial stimulation

Yousefi

Ellard

et al. 2019

Preprint

View full text Add to dashboard Cite

The ability to regulate emotions in the service of meeting ongoing goals and task demands is a key aspect of adaptive human behavior in our volatile social world. Consequently, difficulties in processing and responding to emotional stimuli underlie many psychiatric diseases ranging from depression to anxiety, the common thread being effects on behavior. Behavior, which is made up of shifting, difficult to measure hidden states such as attention and emotion reactivity, is a product of integrating external input and latent mental processes. Directly measuring, and differentiating, separable hidden cognitive, emotional, and attentional states contributing to emotion conflict resolution, however, is challenging, particularly when only using task-relevant behavioral measures such as reaction time. State-space representations are a powerful method for investigating hidden states underlying complex systems. Using state-space modeling of behavior, we identified relevant hidden cognitive states and predicted behavior in a standardized emotion regulation task. After identifying and validating models which best fit the behavior and narrowing our focus to one model, we used targeted intracranial stimulation of the emotion regulation-relevant neurocircuitry, including prefrontal structures and the amygdala, to causally modulate separable states. Finally, we focused on this one validated state-space model to perform real-time, bidirectional closed-loop adaptive stimulation in a subset of participants. These approaches enable an improved understanding of how to sample and understand emotional processing in a way which could be leveraged in neuromodulatory therapy for disorders of emotional regulation.

show abstract

“…Based on this approach, various interesting demonstrations have been made 5 , 6 , 10 , 11 , 13 – 15 . However, the design of such systems is still facing many challenges, such as power budget, delay and scalability, especially in order to catch up with the exponentially increasing number of recording sites in state-of-the-art neural probes 12 , 13 , 16 – 18 . Moreover, this conventional approach is fundamentally different from how brain processes information that is in analog and continuous fashion.…”

Section: Introductionmentioning

confidence: 99%

Neural signal analysis with memristor arrays towards high-efficiency brain–machine interfaces

et al. 2020

View full text Add to dashboard Cite

Brain-machine interfaces are promising tools to restore lost motor functions and probe brain functional mechanisms. As the number of recording electrodes has been exponentially rising, the signal processing capability of brain-machine interfaces is falling behind. One of the key bottlenecks is that they adopt conventional von Neumann architecture with digital computation that is fundamentally different from the working principle of human brain. In this work, we present a memristor-based neural signal analysis system, where the bio-plausible characteristics of memristors are utilized to analyze signals in the analog domain with high efficiency. As a proof-of-concept demonstration, memristor arrays are used to implement the filtering and identification of epilepsy-related neural signals, achieving a high accuracy of 93.46%. Remarkably, our memristor-based system shows nearly 400× improvements in the power efficiency compared to state-of-the-art complementary metal-oxide-semiconductor systems. This work demonstrates the feasibility of using memristors for high-performance neural signal analysis in next-generation brain-machine interfaces.

show abstract

Brain–machine interfaces from motor to mood

Cited by 182 publications

References 98 publications

Data augmentation for deep-learning-based electroencephalography

Data augmentation for deep-learning-based electroencephalography

Bidirectional modulation of human emotional conflict resolution using intracranial stimulation

Neural signal analysis with memristor arrays towards high-efficiency brain–machine interfaces

Contact Info

Product

Resources

About