A5060670102- EEG and Brain-Computer Interfaces
- Neuroscience and Neural Engineering
- Neural dynamics and brain function
- Blind Source Separation Techniques
- Advanced Memory and Neural Computing
- Muscle activation and electromyography studies
- Gaze Tracking and Assistive Technology
- Non-Invasive Vital Sign Monitoring
- Functional Brain Connectivity Studies
- Optical Imaging and Spectroscopy Techniques
- Face Recognition and Perception
- Scientific Computing and Data Management
- ECG Monitoring and Analysis
- Slime Mold and Myxomycetes Research
- Advanced Text Analysis Techniques
- Neurological disorders and treatments
- Neurobiology of Language and Bilingualism
- User Authentication and Security Systems
- Heart Rate Variability and Autonomic Control
- Photoacoustic and Ultrasonic Imaging
- Optical Coherence Tomography Applications
- Philosophy and History of Science
- Action Observation and Synchronization
- Neural Networks and Applications
- CCD and CMOS Imaging Sensors
Johns Hopkins University Applied Physics Laboratory
2021-2024
Johns Hopkins University
2013-2020
Johns Hopkins Medicine
2017-2019
Rensselaer Polytechnic Institute
2011
Brain-computer interfaces (BCIs) that reconstruct and synthesize speech using brain activity recorded with intracranial electrodes may pave the way toward novel communication for people who have lost their ability to speak, or are at high risk of losing this ability, due neurological disorders. Here, we report online synthesis intelligible words a chronically implanted brain-computer interface (BCI) in man impaired articulation ALS, participating clinical trial (ClinicalTrials.gov,...
Abstract Brain‐computer interfaces (BCIs) can be used to control assistive devices by patients with neurological disorders like amyotrophic lateral sclerosis (ALS) that limit speech and movement. For control, it is desirable for BCI systems accurate reliable, preferably minimal setup time. In this study, a participant severe dysarthria due ALS operates computer applications six intuitive commands via chronic electrocorticographic (ECoG) implant over the ventral sensorimotor cortex. Speech...
Current machine learning (ML)-based algorithms for filtering electroencephalography (EEG) time series data face challenges related to cumbersome training times, regularization, and accurate reconstruction. To address these shortcomings, we present an ML filtration algorithm driven by a logistic covariance-targeted adversarial denoising autoencoder (TADA). We hypothesize that the expressivity of targeted, correlation-driven convolutional will enable effective while minimizing compute...
Non-invasive neuroimaging studies have shown that semantic category and attribute information are encoded in neural population activity. Electrocorticography (ECoG) offers several advantages over non-invasive approaches, but the degree to which is ECoG responses not known. We recorded while patients named objects from 12 categories then trained high-dimensional encoding models map attributes spectral-temporal features of task-related responses. Using these models, untrained were decoded with...
Summary Objective This prospective study compared presurgical language localization with visual naming–associated high‐γ modulation ( HGM ) and conventional electrical cortical stimulation ECS in children intracranial electrodes. Methods Patients drug‐resistant epilepsy who were undergoing monitoring included if able to name pictures. Electrocorticography EC oG) signals recorded during picture naming (overt covert) quiet baseline. For each electrode the likelihood of (70–116 Hz) power task...
Recent studies have shown that speech can be reconstructed and synthesized using only brain activity recorded with intracranial electrodes, but until now this has been done retrospective analyses of recordings from able-bodied patients temporarily implanted electrodes for epilepsy surgery. Here, we report online synthesis intelligible words a chronically brain-computer interface (BCI) in clinical trial participant (ClinicalTrials.gov, NCT03567213) dysarthria due to amyotrophic lateral...
Neural keyword spotting could form the basis of a speech brain-computer-interface for menu-navigation if it can be done with low latency and high specificity comparable to "wake-word" functionality modern voice-activated AI assistant technologies. This study investigated neural using motor representations via invasively recorded electrocorticographic signals as proof-of-concept. matched filters were created from monosyllabic consonant-vowel utterances: one utterance, eleven similar...
Any given area in human cortex may receive input from multiple, functionally heterogeneous areas, potentially representing different processing threads. Alpha (8–13 Hz) and beta oscillations (13–20 have been hypothesized by other investigators to gate local cortical processing, but their influence on responses areas is unknown. To study this, we measured the effect of oscillatory power phase elicited single-pulse electrical stimulation (SPES) at distant sites, awake subjects implanted with...
BCI2000 has been a popular platform for development of real-time brain computer interfaces (BCIs). Since BCI2000's initial release, web browsers have evolved considerably, enabling rapid internet-enabled applications and interactive visualizations. Linking the amplifier abstraction signal processing native to with host technologies ease afforded by modern could enable new generation browser-based BCIs We developed server filter module called BCI2000Web providing an HTTP connection capable...
Abstract Objective . Brain–computer interfaces (BCIs) have the potential to preserve or restore speech in patients with neurological disorders that weaken muscles involved production. However, successful training of low-latency synthesis and recognition models requires alignment neural activity intended phonetic acoustic output high temporal precision. This is particularly challenging who cannot produce audible speech, as ground truth which pinpoint synchronized not available. Approach In...
Dimensionality is a serious challenge in human neuroimaging. Across imaging modalities, large pools of potential neural features (e.g. responses from particular voxels, electrodes, and temporal windows) have to be related typically limited sets stimuli samples. To deal with stimulus sets, zero-shot encoding decoding models been introduced classify classes outside the training set. However, these found particularly susceptible curse dimensionality risk over-fitting. While no systematic study...
Abstract Objective . Brain-Computer Interfaces (BCIs) hold significant promise for restoring communication in individuals with partial or complete loss of the ability to speak due paralysis from amyotrophic lateral sclerosis (ALS), brainstem stroke, and other neurological disorders. Many approaches speech decoding reported BCI literature have required time-aligned target representations allow successful training – a major challenge when translating such people who already lost their voice....
Abstract Background Brain-computer interfaces (BCIs) can restore communication for movement- and/or speech-impaired individuals by enabling neural control of computer typing applications. Single command click detectors provide a basic yet highly functional capability. Methods We sought to test the performance and long-term stability decoding using chronically implanted high density electrocorticographic (ECoG) BCI with coverage sensorimotor cortex in human clinical trial participant...
Cortical networks for speech production are believed to be widely distributed and highly organized over temporal, parietal, frontal lobes areas in the human brain cortex. Effective connectivity demonstrates an inherent element of directional information propagation, is therefore dense measure relevant activity different cortical regions. Connectivity analysis electrocorticographic (ECoG) recordings has been studied its excellent signal-to-noise ratio as well high temporal spatial...
Brain-computer interfaces (BCIs) can restore communication in movement- and/or speech-impaired individuals by enabling neural control of computer typing applications. Single command "click" decoders provide a basic yet highly functional capability.
Frequency-domain (FD) fNIRS is attractive for non-invasive brain imaging because phase-sensitive detection leads to increased resolution and may exhibit improved robustness motion artifacts. We present an FD-fNIRS system with silicon photomultiplier (SiPM) receivers, where the sensitivity dynamic range approach those of a first-class continuous-wave (CW-) system. This represents significant step toward fully exploiting phase degree freedom provided by FD-fNIRS. The transmitter subsystem...
After providing some context via (i) earlier work on literary creativity carried out by Bringsjord et al., and (ii) an account of espoused Cope, which stands in rather direct opposition to Bringsjord's account, we summarize our nascent attempt engineer artificial conductor: Handle. Handle is a microcosmic version part larger, much more ambitious system: CAIRA. Both are under development courtesy three-year CreativeIT grant from the National Science Foundation (PI Braasch, Co-PIs Oliveros &...
This paper presents the design and implementation of a signal simulator that emulates event-related human electrocorticographic (ECoG) signals. real-time renders representative model ECoG encompassing prominent physiological modulation in time domain (e.g., potentials, or ERPs) frequency alpha/mu, beta, high gamma band). The simulated signals were generated MATLAB SIMULINK framework output through National Instruments PCI card for recording by standard research-grade amplifier system....
A process is presented for analyzing electrocorticographic (ECoG) recordings and prototyping brain computer interfaces in which complex signal processing chains are able to be rapidly developed iterated digital audio workstation (DAW) software. DAW software includes many built-in "drag drop" blocks that perform common, low-level algorithms such as filtering envelope extraction. In addition being optimized real-time performance, also produces output, allowing listening raw processed signals....
We present a 32-transmitter, 32-receiver dual-wavelength frequency-domain (FD) fNIRS system comprised of commercially available avalanche photodiodes, laser drivers and mounts. The custom frequency domain is used to interrogate cerebral tissue with optodes positioned at the posterior occipital region head. Data are collected from human subjects watching movie scenes no sound. applied cross-validated PCA identify number dimensions retained in neural signal recorded using FD-fNIRS for...