Kamil A. Grajski
Kamil A. Grajski has a total of 23 co-authors. Here are the top co-authors by number of shared publications:
C. Chen
Ford Aerospace
2 shared publicationsC. Kuszmaul
MasPar Computer Corporation
2 shared publicationsG. Chinn
2 shared publicationsRyan Tseng
Qualcomrn Technologies, Incorporated San Diego, California, USA
2 shared publicationsS. Tomboulian
MasPar Computer Corporation
2 shared publicationsBill von Novak
Qualcomrn Technologies, Incorporated San Diego, California, USA
1 shared publication
Latest Publications
- Open Access
Neural voice activity detection with high-gamma ECoG signal correlation structure using a chronically implanted brain-computer interface in an individual with ALS
Chronically implanted brain-computer interfaces (BCI) for speech have demonstrated restoration of speech communication for people with severe motor impairment. Yet maintaining stable long-term performance remains a translational challenge. We investigated whether correlation and partial correlation features of electrocorticographic (ECoG) high-gamma activity (HG-C, HG-PC) could improve robustness compared with high-gamma log-power (HGLP) features for neural voice activity detection (NVAD). Using open-source BCI data from an individual with amyotrophic lateral sclerosis performing a syllable-repetition task, long short-term memory (LSTM) models were trained separately on HGLP, HG-C, and HG-PC features, and evaluated across sessions spanning six months. HG-C and HG-PC achieved comparable or superior NVAD performance to HGLP with smaller long-term averaged loss (HGLP: -17%; HG-C: -9%; HG-PC: -12%) and smaller long-term worst case loss (HGLP: -24%; HG-C: -8%; HG-PC: -16%). Under a simulated local contiguous pattern of neural signal loss, both HG-C and HG-PC outperformed HGLP on averaged long-term loss (HGLP: -22%; HG-C and HG-PC, -10%) and worst-case long-term loss (HGLP: -30%; HG-C: -11%; HG-PC: -13%). The results show that high-gamma correlation-based features captured comparatively more spatially distributed and stable neural speech representations. With further refinement and validation, correlation-based feature representations may contribute to robust longitudinal speech decoding with implanted BCIs.
Open AccessNeural voice activity detection with high-gamma ECoG signal correlation structure using a chronically implanted brain-computer interface in an individual with ALS
Chronically implanted brain-computer interfaces (BCI) for speech have demonstrated restoration of speech communication for people with severe motor impairment. Yet maintaining stable long-term performance remains a translational challenge. We investigated whether correlation and partial correlation features of electrocorticographic (ECoG) high-gamma activity (HG-C, HG-PC) could improve robustness compared with high-gamma log-power (HGLP) features for neural voice activity detection (NVAD). Using open-source BCI data from an individual with amyotrophic lateral sclerosis performing a syllable-repetition task, long short-term memory (LSTM) models were trained separately on HGLP, HG-C, and HG-PC features, and evaluated across sessions spanning six months. HG-C and HG-PC achieved comparable or superior NVAD performance to HGLP with smaller long-term averaged loss (HGLP: -17%; HG-C: -9%; HG-PC: -12%) and smaller long-term worst case loss (HGLP: -24%; HG-C: -8%; HG-PC: -16%). Under a simulated local contiguous pattern of neural signal loss, both HG-C and HG-PC outperformed HGLP on averaged long-term loss (HGLP: -22%; HG-C and HG-PC, -10%) and worst-case long-term loss (HGLP: -30%; HG-C: -11%; HG-PC: -13%). The results show that high-gamma correlation-based features captured comparatively more spatially distributed and stable neural speech representations. With further refinement and validation, correlation-based feature representations may contribute to robust longitudinal speech decoding with implanted BCIs.
- 7 December 2020
- Published by Wiley in Alzheimer's & Dementia
- Vol. 16 (S4)
Medial temporal lobe (MTL) ‐ default‐mode network (DMN) functional connectivity disruption in the progression of Alzheimer’s disease
Background: A multi‐modal analysis of resting‐state functional connectivity disruption (fMRI) and morphometric changes characteristic of neurodegeneration (T1 structural MRI) in Alzheimer's disease (AD) was performed to develop biomarkers for the transition from cognitively normal (CN) to significant memory concern (SMC) to early and late mild cognitive impairment (EMCI, LMCI).Methods: False Detection Rate (FDR) corrected (q=0.005) parallel group‐level one‐way four‐factor (CN, SMC, EMCI, LMCI) ANOVA and post hoc pairwise t tests of ROI pairs [1]; same‐session functional connectivity (AFNI) and morphometry (FreeSurfer); ADNI‐3 sub‐population with baseline diagnosis CN (N=33), SMC (N=35), EMCI (N=13), and LMCI (N=14) matched for age (p=0.33), education (p=0.63), and gender (p=0.127).Results: A 3‐mm grid of ROI seeds (r=6mm) in left MTL yielded 435 ROI “anchors”. 3D (Fisher Z‐transformed) correlation maps were computed for each subject for each anchor ROI and input to a group‐level one‐way four‐factor 3D ANOVA. Cluster analysis (N vox 20; p=0.02; P=3.45) of the 3D F Statistic maps yielded 4,185 target ROIs. An r=6mm ROI was placed at the center of mass of each target cluster. ROI anchor‐target pair correlations were input to a group‐level one‐way four‐factor ANOVA and FDR corrected (q=0.005; p*=0.003) (estimated by permutation, N=1000) which identified 2,420 significant tests (ROI pairs). Post hoc pairwise t tests only on the significant ANOVA tests were FDR‐corrected (q=0.005) and yielded 31 significant SMC‐CN tests (p*=6.3x10‐5); 54 EMCI‐CN tests (p*=1.1x10‐4); and 209 LMCI‐CN tests (p*=4.2x10‐4). The spatial distribution of target ROIs included (bilaterally) MTL and DMN‐associated regions. Of the significant SMC‐CN tests, 39% were positive (increased connectivity). In the significant EMCI‐CN tests, 100% were positive. However, in the significant LMCI‐CN comparisons, the pattern reversed and only 5% were positive. An independent analysis of right hemisphere yielded similar results.Conclusions: An hybrid a priori data‐driven approach identified a population of MTL‐anchored ROI pairs whose spatial distribution of target ROIs included regions associated with DMN and MTL, and whose functional connectivity revealed a striking pattern of change with clinical progression when compared to CN: SMC, a mixture of hyper‐ and hypo‐connectivity; EMCI, widespread hyper‐connectivity; and LMCI, widespread hypo‐connectivity. Reference: (1) Grajski, K. A., Bressler, S., and ADNI. (2019). Neuroimage Clin. 2019;23:101860.
- 7 December 2020
- Published by Wiley in Alzheimer's & Dementia
- Vol. 16 (S4)
Medial temporal lobe (MTL) ‐ default‐mode network (DMN) functional connectivity disruption in the progression of Alzheimer’s disease
Background: A multi‐modal analysis of resting‐state functional connectivity disruption (fMRI) and morphometric changes characteristic of neurodegeneration (T1 structural MRI) in Alzheimer's disease (AD) was performed to develop biomarkers for the transition from cognitively normal (CN) to significant memory concern (SMC) to early and late mild cognitive impairment (EMCI, LMCI).Methods: False Detection Rate (FDR) corrected (q=0.005) parallel group‐level one‐way four‐factor (CN, SMC, EMCI, LMCI) ANOVA and post hoc pairwise t tests of ROI pairs [1]; same‐session functional connectivity (AFNI) and morphometry (FreeSurfer); ADNI‐3 sub‐population with baseline diagnosis CN (N=33), SMC (N=35), EMCI (N=13), and LMCI (N=14) matched for age (p=0.33), education (p=0.63), and gender (p=0.127).Results: A 3‐mm grid of ROI seeds (r=6mm) in left MTL yielded 435 ROI “anchors”. 3D (Fisher Z‐transformed) correlation maps were computed for each subject for each anchor ROI and input to a group‐level one‐way four‐factor 3D ANOVA. Cluster analysis (N vox 20; p=0.02; P=3.45) of the 3D F Statistic maps yielded 4,185 target ROIs. An r=6mm ROI was placed at the center of mass of each target cluster. ROI anchor‐target pair correlations were input to a group‐level one‐way four‐factor ANOVA and FDR corrected (q=0.005; p*=0.003) (estimated by permutation, N=1000) which identified 2,420 significant tests (ROI pairs). Post hoc pairwise t tests only on the significant ANOVA tests were FDR‐corrected (q=0.005) and yielded 31 significant SMC‐CN tests (p*=6.3x10‐5); 54 EMCI‐CN tests (p*=1.1x10‐4); and 209 LMCI‐CN tests (p*=4.2x10‐4). The spatial distribution of target ROIs included (bilaterally) MTL and DMN‐associated regions. Of the significant SMC‐CN tests, 39% were positive (increased connectivity). In the significant EMCI‐CN tests, 100% were positive. However, in the significant LMCI‐CN comparisons, the pattern reversed and only 5% were positive. An independent analysis of right hemisphere yielded similar results.Conclusions: An hybrid a priori data‐driven approach identified a population of MTL‐anchored ROI pairs whose spatial distribution of target ROIs included regions associated with DMN and MTL, and whose functional connectivity revealed a striking pattern of change with clinical progression when compared to CN: SMC, a mixture of hyper‐ and hypo‐connectivity; EMCI, widespread hyper‐connectivity; and LMCI, widespread hypo‐connectivity. Reference: (1) Grajski, K. A., Bressler, S., and ADNI. (2019). Neuroimage Clin. 2019;23:101860.
- Open Access
- 1 January 2019
- Published by Elsevier in NeuroImage: Clinical
- Vol. 23, 101860
Differential medial temporal lobe and default-mode network functional connectivity and morphometric changes in Alzheimer's disease
- 1 January 2019
- Published by Elsevier in NeuroImage: Clinical
- Vol. 23, 101860
Open AccessDifferential medial temporal lobe and default-mode network functional connectivity and morphometric changes in Alzheimer's disease
- Open Access
Differential medial temporal lobe and default-mode network functional connectivity and morphometric changes in Alzheimer’s Disease
We report group level differential detection of medial temporal lobe resting-state functional connectivity disruption and morphometric changes in the transition from cognitively normal to early mild cognitive impairment in an age-, education- and gender-matched 105 subjects Alzheimer’s Disease Neuroimaging Initiative dataset. In mild Alzheimer’s Disease, but not early mild cognitive impairment, characteristic brain atrophy was detected in FreeSurfer estimates of cortical thickness and subcortical and hippocampal subfield volumes. By contrast, functional connectivity analysis detected earlier significant changes. In early mild cognitive impairment these changes involved medial temporal lobe regions of transentorhinal, perirhinal and entorhinal cortices (associated with the earliest stages of neurofibrillary changes in Alzheimer’s Disease), hippocampus, parahippocampal gyrus and temporal pole, and cortical regions comprising or co-activated with the default-mode network, including rostral and medial prefrontal cortex, anterior cingulate cortex, precuneus and inferior temporal cortex. Key findings include: a) focal and bilaterally symmetric spatial organization of affected medial temporal lobe regions; b) mutual hyperconnectivity bilaterally involving ventral medial temporal lobe structures (temporal pole, uncus); and c) dorsal medial temporal lobe hypoconnectivity with anterior and posterior midline default-mode network nodes. These findings position medial temporal lobe resting state functional connectivity as a candidate biomarker of an Alzheimer’s Disease pathophysiological cascade, potentially in advance of clinical biomarkers, and coincident with biomarkers of the earliest stages of Alzheimer’s neuropathology. Our results indicate that medial temporal lobe resting-state functional connectivity should be further investigated as a potential biomarker in the diagnosis of Alzheimer’s Disease. Highlights: Functional connectivity change seen before structural change in Alzheimer’s Disease Medial temporal lobes mutually hyper-connect in mild cognitive impairment Medial temporal lobe and default mode network decouple in mild cognitive impairment Loci of functional change in hippocampi are focal with bilaterally symmetric features Nonmonotonic functional connectivity changes in Alzheimer’s Disease progression
Open AccessDifferential medial temporal lobe and default-mode network functional connectivity and morphometric changes in Alzheimer’s Disease
We report group level differential detection of medial temporal lobe resting-state functional connectivity disruption and morphometric changes in the transition from cognitively normal to early mild cognitive impairment in an age-, education- and gender-matched 105 subjects Alzheimer’s Disease Neuroimaging Initiative dataset. In mild Alzheimer’s Disease, but not early mild cognitive impairment, characteristic brain atrophy was detected in FreeSurfer estimates of cortical thickness and subcortical and hippocampal subfield volumes. By contrast, functional connectivity analysis detected earlier significant changes. In early mild cognitive impairment these changes involved medial temporal lobe regions of transentorhinal, perirhinal and entorhinal cortices (associated with the earliest stages of neurofibrillary changes in Alzheimer’s Disease), hippocampus, parahippocampal gyrus and temporal pole, and cortical regions comprising or co-activated with the default-mode network, including rostral and medial prefrontal cortex, anterior cingulate cortex, precuneus and inferior temporal cortex. Key findings include: a) focal and bilaterally symmetric spatial organization of affected medial temporal lobe regions; b) mutual hyperconnectivity bilaterally involving ventral medial temporal lobe structures (temporal pole, uncus); and c) dorsal medial temporal lobe hypoconnectivity with anterior and posterior midline default-mode network nodes. These findings position medial temporal lobe resting state functional connectivity as a candidate biomarker of an Alzheimer’s Disease pathophysiological cascade, potentially in advance of clinical biomarkers, and coincident with biomarkers of the earliest stages of Alzheimer’s neuropathology. Our results indicate that medial temporal lobe resting-state functional connectivity should be further investigated as a potential biomarker in the diagnosis of Alzheimer’s Disease. Highlights: Functional connectivity change seen before structural change in Alzheimer’s Disease Medial temporal lobes mutually hyper-connect in mild cognitive impairment Medial temporal lobe and default mode network decouple in mild cognitive impairment Loci of functional change in hippocampi are focal with bilaterally symmetric features Nonmonotonic functional connectivity changes in Alzheimer’s Disease progression
- Open Access
- 25 July 2016
- Published by Frontiers Media SA in Frontiers in Computational Neuroscience
- Vol. 10, 72
Emergent Spatial Patterns of Excitatory and Inhibitory Synaptic Strengths Drive Somatotopic Representational Discontinuities and their Plasticity in a Computational Model of Primary Sensory Cortical Area 3b
Mechanisms underlying the emergence and plasticity of representational discontinuities in the mammalian primary somatosensory cortical representation of the hand are investigated in a computational model. The model consists of an input lattice organized as a three-digit hand forward-connected to a lattice of cortical columns each of which contains a paired excitatory and inhibitory cell. Excitatory and inhibitory synaptic plasticity of feedforward and lateral connection weights is implemented as a simple covariance rule and competitive normalization. Receptive field properties are computed independently for excitatory and inhibitory cells and compared within and across columns. Within digit representational zones intracolumnar excitatory and inhibitory receptive field extents are concentric, single-digit, small, and unimodal. Exclusively in representational boundary-adjacent zones, intracolumnar excitatory and inhibitory receptive field properties diverge: excitatory cell receptive fields are single-digit, small, and unimodal; and the paired inhibitory cell receptive fields are bimodal, double-digit, and large. In simulated syndactyly (webbed fingers), boundary-adjacent intracolumnar receptive field properties reorganize to within-representation type; divergent properties are reacquired following syndactyly release. This study generates testable hypotheses for assessment of cortical laminar-dependent receptive field properties and plasticity within and between cortical representational zones. For computational studies, present results suggest that concurrent excitatory and inhibitory plasticity may underlie novel emergent properties.
- 25 July 2016
- Published by Frontiers Media SA in Frontiers in Computational Neuroscience
- Vol. 10, 72
Open AccessEmergent Spatial Patterns of Excitatory and Inhibitory Synaptic Strengths Drive Somatotopic Representational Discontinuities and their Plasticity in a Computational Model of Primary Sensory Cortical Area 3b
Mechanisms underlying the emergence and plasticity of representational discontinuities in the mammalian primary somatosensory cortical representation of the hand are investigated in a computational model. The model consists of an input lattice organized as a three-digit hand forward-connected to a lattice of cortical columns each of which contains a paired excitatory and inhibitory cell. Excitatory and inhibitory synaptic plasticity of feedforward and lateral connection weights is implemented as a simple covariance rule and competitive normalization. Receptive field properties are computed independently for excitatory and inhibitory cells and compared within and across columns. Within digit representational zones intracolumnar excitatory and inhibitory receptive field extents are concentric, single-digit, small, and unimodal. Exclusively in representational boundary-adjacent zones, intracolumnar excitatory and inhibitory receptive field properties diverge: excitatory cell receptive fields are single-digit, small, and unimodal; and the paired inhibitory cell receptive fields are bimodal, double-digit, and large. In simulated syndactyly (webbed fingers), boundary-adjacent intracolumnar receptive field properties reorganize to within-representation type; divergent properties are reacquired following syndactyly release. This study generates testable hypotheses for assessment of cortical laminar-dependent receptive field properties and plasticity within and between cortical representational zones. For computational studies, present results suggest that concurrent excitatory and inhibitory plasticity may underlie novel emergent properties.
- 1 May 2013
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 79-83
Introduction to the alliance for wireless power loosely-coupled wireless power transfer system specification version 1.0
The Alliance for Wireless Power (A4WP) Version 1.0 Baseline System Specification (BSS) is an interoperability specification for loosely-coupled (LC) wireless power transfer (WPT) systems that meet next-generation user experience and industrial design requirements as part of the vision of Ubiquitous Power for portable, hand-held, consumer electronic devices. This survey of the A4WP Version 1.0 BSS reviews an LC WPT reference model, power transmitting and power receiving unit device classification, power system interfaces and key parameters, the power control specification, the signaling specification, and the framework for certification and acceptance testing based on reference power transmitting and power receiving resonators. The selection of ISM band (6.78MHz) as the WPT operating frequency and ISM band (2.4GHz) for WPT management protocol via BluetoothLE is discussed.
- 1 May 2013
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 79-83
Introduction to the alliance for wireless power loosely-coupled wireless power transfer system specification version 1.0
The Alliance for Wireless Power (A4WP) Version 1.0 Baseline System Specification (BSS) is an interoperability specification for loosely-coupled (LC) wireless power transfer (WPT) systems that meet next-generation user experience and industrial design requirements as part of the vision of Ubiquitous Power for portable, hand-held, consumer electronic devices. This survey of the A4WP Version 1.0 BSS reviews an LC WPT reference model, power transmitting and power receiving unit device classification, power system interfaces and key parameters, the power control specification, the signaling specification, and the framework for certification and acceptance testing based on reference power transmitting and power receiving resonators. The selection of ISM band (6.78MHz) as the WPT operating frequency and ISM band (2.4GHz) for WPT management protocol via BluetoothLE is discussed.
- 1 May 2012
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 9-14
Loosely-coupled wireless power transfer: Physics, circuits, standards
Loosely-coupled (LC) wireless power transfer (WPT) offers unique, next-generation improvements in user experience, and product design and innovation towards the vision of ubiquitous power, and infinite stand-by time for a wide range of consumer electronic devices. We outline the vision as it relates to today's mobile, battery-powered, hand-held, consumer electronics devices, and consider the unique capabilities of loosely-coupled (LC) WPT. We then describe the physics of the magnetically coupled resonant coil subsystem, which lies at the heart of LC WPT, discuss approaches and challenges in implementing WPT electronic circuits, and review industry progress towards timely, high-quality WPT technical standards.
- 1 May 2012
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 9-14
Loosely-coupled wireless power transfer: Physics, circuits, standards
Loosely-coupled (LC) wireless power transfer (WPT) offers unique, next-generation improvements in user experience, and product design and innovation towards the vision of ubiquitous power, and infinite stand-by time for a wide range of consumer electronic devices. We outline the vision as it relates to today's mobile, battery-powered, hand-held, consumer electronics devices, and consider the unique capabilities of loosely-coupled (LC) WPT. We then describe the physics of the magnetically coupled resonant coil subsystem, which lies at the heart of LC WPT, discuss approaches and challenges in implementing WPT electronic circuits, and review industry progress towards timely, high-quality WPT technical standards.
- 1 December 1993
- Published by Acoustical Society of America (ASA) in The Journal of the Acoustical Society of America
- Vol. 94 (6) , 3534
Apparatus and method for detecting vibration patterns
- 1 December 1993
- Published by Acoustical Society of America (ASA) in The Journal of the Acoustical Society of America
- Vol. 94 (6) , 3534
Apparatus and method for detecting vibration patterns
- 1 March 1990
- Published by MIT Press in Neural Computation
- Vol. 2 (1) , 71-84
Hebb-Type Dynamics is Sufficient to Account for the Inverse Magnification Rule in Cortical Somatotopy
The inverse magnification rule in cortical somatotopy is the experimentally derived inverse relationship between cortical magnification (area of somatotopic map representing a unit area of skin surface) and receptive field size (area of restricted skin surface driving a cortical neuron). We show by computer simulation of a simple, multilayer model that Hebb-type synaptic modification subject to competitive constraints is sufficient to account for the inverse magnification rule.
- 1 March 1990
- Published by MIT Press in Neural Computation
- Vol. 2 (1) , 71-84
Hebb-Type Dynamics is Sufficient to Account for the Inverse Magnification Rule in Cortical Somatotopy
The inverse magnification rule in cortical somatotopy is the experimentally derived inverse relationship between cortical magnification (area of somatotopic map representing a unit area of skin surface) and receptive field size (area of restricted skin surface driving a cortical neuron). We show by computer simulation of a simple, multilayer model that Hebb-type synaptic modification subject to competitive constraints is sufficient to account for the inverse magnification rule.
- 1 January 1990
- Published by Springer Nature
- p. 673
Neural Network Simulation on the MasPar MP-1 Massively Parallel Processor
- 1 January 1990
- Published by Springer Nature
- p. 673
Neural Network Simulation on the MasPar MP-1 Massively Parallel Processor