Faculty Bibliography

Localizing neuronal patterns that generate pathological brain signals may assist with tissue resection and intervention strategies in patients with neurological diseases. Precise localization requires high spatiotemporal recording from populations of neurons while minimizing invasiveness and adverse events. We describe a large-scale, high-density, organic material-based, conformable neural interface device ("NeuroGrid") capable of simultaneously recording local field potentials (LFPs) and action potentials from the cortical surface. We demonstrate the feasibility and safety of intraoperative recording with NeuroGrids in anesthetized and awake subjects. Highly localized and propagating physiological and pathological LFP patterns were recorded, and correlated neural firing provided evidence about their local generation. Application of NeuroGrids to brain disorders, such as epilepsy, may improve diagnostic precision and therapeutic outcomes while reducing complications associated with invasive electrodes conventionally used to acquire high-resolution and spiking data.

Glioblastoma (GBM) is a deadly primary brain malignancy with extensive intratumoral hypoxia. Hypoxic regions of GBM contain stem-like cells and are associated with tumor growth and angiogenesis. The molecular mechanisms that regulate tumor growth in hypoxic conditions are incompletely understood. Here, we use primary human tumor biospecimens and cultures to identify GPR133 (ADGRD1), an orphan member of the adhesion family of G-protein-coupled receptors, as a critical regulator of the response to hypoxia and tumor growth in GBM. GPR133 is selectively expressed in CD133+ GBM stem cells (GSCs) and within the hypoxic areas of PPN in human biospecimens. GPR133 mRNA is transcriptionally upregulated by hypoxia in hypoxia-inducible factor 1alpha (Hif1alpha)-dependent manner. Genetic inhibition of GPR133 with short hairpin RNA reduces the prevalence of CD133+ GSCs, tumor cell proliferation and tumorsphere formation in vitro. Forskolin rescues the GPR133 knockdown phenotype, suggesting that GPR133 signaling is mediated by cAMP. Implantation of GBM cells with short hairpin RNA-mediated knockdown of GPR133 in the mouse brain markedly reduces tumor xenograft formation and increases host survival. Analysis of the TCGA data shows that GPR133 expression levels are inversely correlated with patient survival. These findings indicate that GPR133 is an important mediator of the hypoxic response in GBM and has significant protumorigenic functions. We propose that GPR133 represents a novel molecular target in GBM and possibly other malignancies where hypoxia is fundamental to pathogenesis.

The neural mechanisms that support working memory (WM) depend on persistent neural activity. Within topographically organized maps of space in dorsal parietal cortex, spatially selective neural activity persists during WM for location. However, to date the necessity of these topographic subregions of human parietal cortex for WM remain unknown. To test the causal relationship of these areas to WM, we compared the performance of patients with lesions to topographically organized parietal cortex to controls on a memory-guided saccade (MGS) task as well as a visually-guided saccade (VGS) task. The MGS task allowed us to measure WM precision continuously with great sensitivity, while the VGS task allowed us to control for any deficits in general spatial or visuomotor processing. Compared to controls, patients generated memory-guided saccades that were significantly slower and less accurate, while visually-guided saccades were unaffected. These results provide key missing evidence for the causal role of topographic areas in human parietal cortex for WM, as well as the neural mechanisms supporting WM.

Surface-based cortical thickness (CT) analyses are increasingly being used to investigate variations in brain morphology across the spectrum of brain health, from neurotypical to neuropathological. An outstanding question is whether individual differences in cortical morphology, such as regionally increased or decreased CT, are associated with domain-specific performance deficits in healthy adults. Since CT studies are correlational, they cannot establish causality between brain morphology and cognitive performance. A direct comparison with classic lesion methods is needed to determine whether the regional specificity of CT-cognition correlations is similar to that observed in patients with brain lesions. We address this question by comparing the neuroanatomical overlap of effects when 1) whole brain vertex-wise CT is tested as a correlate of performance variability on a commonly used neuropsychological test of executive function, Trailmaking Test Part B (TMT-B), in healthy adults and 2) voxel-based lesion-symptom mapping (VBLSM) is used to map lesion location to performance decrements on the same task in patients with frontal lobe lesions. We found that reduced performance on the TMT-B was associated with increased CT in bilateral prefrontal regions in healthy adults and that results spatially overlapped in the left dorsomedial prefrontal cortex with findings from the VBLSM analysis in patients with frontal brain lesions. Findings indicate that variations in the structural integrity of the left dorsomedial prefrontal lobe, ranging from individual CT differences in healthy adults to structural lesions in patients with neurological disorders, are associated with poor performance on the TMT-B. These converging results suggest that the left dorsomedial prefrontal region houses a critical region for the complex processing demands of TMT-B, which include visuomotor tracking, sequencing, and cognitive flexibility.

We present a new approach to extracting low-dimensional neural trajectories that summarize the electrocorticographic (ECoG) signals recorded with high-channel-count electrode arrays implanted subdurally. In our approach, Hidden-Markov Factor Analysis (HMFA), a finite set of factor analyzers are used to model the relationship between the high-dimensional ECoG neural space and a low-dimensional latent neural space; the factor analyzers at different time points are in turn linked together with a hidden Markov model. The recorded ECoG signals were band-pass filtered such that our analysis was focused on a sub-band (76-100Hz) of high gamma. HMFA affords the quantization of the ECoG neural space and dimensionality reduction in a common probabilistic space. We applied this method to the ECoG recordings of 2 subjects who responded with button presses to audiovisual stimuli in an experimental task. Using a goodness-of-fit metric that measures how well the ECoG activity of each electrode can be predicted by all the other electrodes, we found that HMFA performed best when compared with Gaussian-Process Factor Analysis (GPFA) and other related spatiotemporal modeling techniques. In contradistinction to HMFA, GPFA and the other techniques integrate temporal smoothing with dimensionality reduction. We believe that this method will provide a powerful tool for relating high-channel-count ECoG signals to the perception and behavior of subjects.

To augment neural monitoring, a minimally intrusive multi-modal capture system was designed and implemented in the epilepsy clinic. This system provides RGB-D audio-video synchronized with patient electrocorticography (ECoG), which records neural activity across cortex. We propose an automated approach to studying the human brain in a naturalistic setting. We demonstrate coarse functional mapping of ECoG electrodes correlated to contralateral arm movements. Motor electrode mapping was generated by analyzing continuous movement data recorded over several hours from epilepsy patients in hospital rooms. From these recordings we estimate the kinematics of patient hand movement behaviors using computer vision algorithms. We compare movement behaviors to neural data collected from ECoG, specifically high-γ (70-110 Hz) spectral features. We present a functional map of electrode responses to natural arm movements, generated using a statistical test. We demonstrate that our approach has the potential to aid in the development of automated functional brain mapping using continuous video and neural recordings of patients in clinical settings.

A dominant theory, based on electrophysiological and lesion evidence from nonhuman primate studies, posits that the dorsolateral prefrontal cortex (dlPFC) stores and maintains working memory (WM) representations. Yet, neuroimaging studies have consistently failed to translate these results to humans; these studies normally find that neural activity persists in the human precentral sulcus (PCS) during WM delays. Here, we attempt to resolve this discrepancy. To test the degree to which dlPFC is necessary for WM, we compared the performance of patients with dlPFC lesions and neurologically healthy controls on a memory-guided saccade task that was used in the monkey studies to measure spatial WM. We found that dlPFC damage only impairs the accuracy of memory-guided saccades if the damage impacts the PCS; lesions to dorsolateral dlPFC that spare the PCS have no effect on WM. These results identify the necessary subregion of the frontal cortex for WM and specify how this influential animal model of human cognition must be revised. SIGNIFICANCE STATEMENT: High-level cognition depends on working memory (WM) as a critical building block, and many symptoms of psychiatric disorders may be the direct result of impaired WM. Canonical theory posits a critical role for the dorsolateral prefrontal cortex (dlPFC) in WM based on studies of nonhuman primates. However, we find that spatial WM in humans is intact after dlPFC damage unless it impacts the more caudal PCS. Therefore, the human dlPFC is not necessary for spatial WM and highlights the need for careful translation of animal models of human cognition.

BACKGROUND: Vagus nerve stimulation (VNS) is an established surgical treatment for medically intractable epilepsy with more than 75 000 devices implanted worldwide. While there are many reports documenting efficacy, complications, and clinical use, there are very few reports concerning VNS battery replacement and revision surgeries. OBJECTIVE: To review our experience with VNS battery replacement and revision surgery. METHODS: We retrospectively reviewed 1144 consecutive VNS procedures performed by a single surgeon between 1998 and 2012. Six hundred forty-four of those procedures were the initial placement of the VNS device. These patients were then followed to determine when a battery change occurred and what type of revision or removal was necessary. RESULTS: In the study, 46% of patients required at least 1 or more type of battery replacement or revision surgery. The most common types of surgery were for generator battery depletion (27%), poor efficacy (9%), and lead malfunction (8%). Only 2% of patients were noted to have an infection. CONCLUSION: VNS battery replacement, revisions, and removals account for almost one-half of all VNS procedures. Our findings suggest important long-term expectations for VNS including expected complications, battery life, and other surgical issues. Review of the literature suggests that this is the first large review of VNS revisions by a single center. Our findings are important to better characterize long-term surgical expectations of VNS therapy. A significant portion of patients undergoing VNS therapy will eventually require revision. ABBREVIATION: VNS, vagus nerve stimulation.