A5042284505- Functional Brain Connectivity Studies
- Neural dynamics and brain function
- EEG and Brain-Computer Interfaces
- Speech and Audio Processing
- Structural Health Monitoring Techniques
- Non-Invasive Vital Sign Monitoring
- Advanced Adaptive Filtering Techniques
- Advanced Neuroimaging Techniques and Applications
- Image and Signal Denoising Methods
- Advanced MRI Techniques and Applications
- Traumatic Brain Injury Research
- Aerodynamics and Acoustics in Jet Flows
- Blind Source Separation Techniques
- Direction-of-Arrival Estimation Techniques
- Context-Aware Activity Recognition Systems
- Healthcare Technology and Patient Monitoring
- Gait Recognition and Analysis
- Indoor and Outdoor Localization Technologies
- Infrastructure Maintenance and Monitoring
- Flow Measurement and Analysis
- Machine Fault Diagnosis Techniques
- Hemodynamic Monitoring and Therapy
- Color Science and Applications
- Automotive and Human Injury Biomechanics
- Agriculture and Farm Safety
Georgia Institute of Technology
2014-2024
Georgia Tech Research Institute
2013-2024
Emory University
2016
Florida State University
2007
University of Oklahoma
2006
This study demonstrates robust human activity recognition from a single triaxial accelerometer via bilateral domain adaptation using semi-supervised deep translation networks. Datasets were obtained previously published studies: University of Michigan (Domain 1) and Georgia Institute Technology 2) where accelerometry was on subjects during defined conditions with the goal recognizing standing rest, walking (level ground), (decline), (incline) without stairs (activity classes). Collected data...
A new full-reference color image quality index based on the universal (UIQI) of Bovik and Wang, (2002) is presented. Because UIQI only takes into consideration distortions luminance component image, a measure chrominance components has been added. The chromatic are represented in an opponent space that better represents perception human eye. cross-correlation between original coded expresses distortion due to compression. Our combines measures through regression analysis using LIVE database...
We employed wearable multimodal sensing (heart rate and triaxial accelerometry) with machine learning to enable early prediction of impending exertional heat stroke (EHS). US Army Rangers Combat Engineers (N = 2,102) were instrumented while participating in rigorous 7-mile 12-mile loaded rucksack timed marches. There three EHS cases, data from 478 analyzed for model building controls. The data-driven approach incorporated estimates physiological strain rate) physical stress (estimated...
The field of brain connectomics develops our understanding the brain's intrinsic organization by characterizing trends in spontaneous activity. Linear correlations blood-oxygen level dependent functional magnetic resonance imaging (BOLD-fMRI) fluctuations are often used as measures connectivity (FC), that is, a quantity describing how similarly two regions behave over time. Given natural spectral scaling BOLD-fMRI signals, it may be useful to represent multiple processes occurring scales....
System identification for structural engineering has received significant attention in the last thirty years. With ever increasing capacity of computing technology, system been applied to important structures such as bridges and aircraft. In case bridges, output can easily be measured by accelerometers. Considerable research on done using output-only models. Of course, it is difficult measure inputs an in-service bridge. this paper, we see how estimated from measurements. We then use...
This work presents a new data-driven method for the identification of functionally connected regions in rat brain, using agglomerative clustering based on discrete wavelet transform (DWT). The proposed approach is evaluated resting state fMRI data and no priori assumptions about distribution signals or anatomical ROIs are made. coefficients DWT used as features algorithm, performance classifier capability to produce clusters that best correlate with known sensorimotor cortex brain. Wavelet...
Recent advances in functional connectivity (FC) analysis of magnetic resonance imaging (fMRI) data facilitate the characterization brain's intrinsic networks (FC-fMRI). Because fMRI signal does not provides a perfect representation neuronal activity, potential for FC-fMRI to identify functionally relevant critically depends upon separating overlapping signals from one another and external noise. As step preconditioning, researchers often band-pass filter range 0.01 Hz 0.1 Hz. However,...
The need for rapid development of tactical system systems solutions military applications requires the use modeling techniques and simulation validation methods to be applied throughout lifecycle system. This combined approach verification is preferred traditional approaches risk mitigation cost effectiveness. paper examines Integrated Blast Effects Sensor Suite developed at Georgia Tech Research Institute its architecture as a complex systems.
Underwater sonar acoustic returns of objects on the seafloor have many time varying components including specular geometric backscatter, scattering from vibrational dynamics, helical, circumferential, and meridional Rayleigh waves, guided circumferential Lamb multipath wave field reflections both object interest scene. These features return can reveal information about an object's shape, structure, material composition, relationship to background Analysis in time-frequency domain is,...
Abstract Measures of whole-brain activity, from techniques such as functional Magnetic Resonance Imaging, provide a means to observe the brain’s dynamical operations. However, interpretation dynamics has been stymied by inherently high-dimensional structure brain activity. The present research addresses this challenge through series scale transformations in spectral, spatial, and relational domains. Instantaneous multispectral are first developed input data via wavelet filter bank....
This paper presents a data-driven clustering method based on the use of wavelet packet features for study functionally connected regions in brain. In particular, packets are used because their optimal whitening properties 1/f-like processes, association with uniform segmentation frequency axis. Features obtained by transform grouped together using agglomerative standardize Euclidian distance. The results compared known atlas rat brain and from technique standard previously presented same...
Wearable sensor systems represent an increasingly viable approach to short and long term motion classification gait estimation primarily due decreased size, cost broad applicability. Accurate human recognition by means of minimally intrusive low-power has countless applications in healthcare, sports medicine, military operational settings. In this work, locomotion was characterized terms response using for two type activities. A single positioned on the chest used recorded acceleration...
Recent advent of fast imaging techniques for MRI application allow whole brain coverage with sub-second resolution, opening the door new data-driven computational that can harvest information contained in data. This paper examines use wavelet based spectral decomposition and hierarchical clustering resting state functional MRI. Wavelet packets naturally enable short time minimal temporal window lengths across multiple frequency ranges, while is used organizing broadband filtered fMRI data...
Researchers from Georgia Tech, Emory University, and Tech Research Institute previously collaborated on a 2D Heads-Up Display based gamified health assessment focused determining neurologic impairment. As part of continuing gamification the assessment, entire project was translated to 3D Virtual Reality under significant constraints. The system needed test same neural pathways in manner as previous order use data validation for approach. Some aspects user response immersion required changes...
Concussions affect an estimated 1.6 to 3.8 million people in the United States annually. Dizziness, which may manifest as a result of vestibular or oculomotor impairments, is common symptom (>50%) following concussion and associated with increased risk for protracted recovery. PURPOSE:To evaluate novel measure function after sub-acute immersive testing environment. METHODS: 26 participants (age: 17.6 ± 4.8 years) presenting dizziness imbalance (32.9 37.2 days post injury, range: 1-156 days)...
Methods to interpret data obtained from resting state functional magnetic imaging (rs-fMRI) must be developed more thoroughly understand how network structure of the brain supports body and mind. To this end, we examine use agglomerative clustering (AC) as a method for rs-fMRI analysis. AC is driven approach organizing spatially distinct clusters temporally similar activity. Its application produces spatial parcellation areas that share temporal characteristics. The technique scalable,...
In this paper, we present the adaptation to least-square solution for frequency invariant beamforming arbitrary planar arrays. This formulation is easily steerable thanks decoupling of spatial and spectral constraints. achieved with a basis spherical harmonics polynomials, which parametrize array response in term dependent on geometry. Furthermore, parametrization Bessel functions decouples term, becomes independent geometry array, allows steering azimuth elevation independently. The...
This paper presents a modified time-frequency based beamforming technique to use both azimuth and elevation information for signal enhancement of very low Signal-to-Noise Ratio (SNR) multichannel auditory MAV (Micro Aerial Vehicle) recordings. The method identities acoustic sources embedded in strong harmonic rotor noise by exploiting the sparsity source content. Spatially informed Wiener filters are applied individual recording channels different array steering directions, maximum kurtosis...
We present an application for the acoustic detection and tracking of Unmanned Aerial Vehicles (UAVs) using a spherical harmonic multiple signal classification technique implemented commercially available array. Preliminary results obtained during field testing demonstrate ability array system to simultaneously accurately track up two UAVs in flight, provide insight on detectable target range, robustness against wind environmental noise. Additionally, effectiveness this is investigated...