Search

Parkinsons disease (PD) diagnosis is challenging due to subtle early clinical signs. F-DOPA PET is commonly used for early PD diagnosis. We explore the potential of machine-learning (ML) based EEG features extracted from single-channel EEG during auditory cognitive assessment as a noninvasive, low-cost support for PD diagnosis. The study included data collected from 32 participants who underwent an F-DOPA PET scan as part of their standard treatment and 20 cognitively healthy controls. Participants performed an auditory cognitive assessment recorded with Neurosteer EEG device. Data processing involved wavelet-packet decomposition and ML. First, a prediction model was developed to predict 1/3 of the undisclosed F-DOPA results. Then, generalized linear mixed models were calculated to distinguish between PD and non-PD subjects on the frequency bands and ML-based EEG features (A0 and L1) previously associated with cognitive functions. The prediction model accurately labeled patients with unrevealed scores as positive F-DOPA. Novel EEG feature A0 and the Delta band showed significant separation between study groups, with healthy controls exhibiting higher activity than PD patients. EEG feature L1 activity was significantly lower in resting state compared to high-cognitive load. This effect was absent in the PD group, suggesting that lower activity in resting state is lacking in PD patients. This study successfully demonstrated the ability to separate patients with positive vs. negative F-DOPA PET results with an easy-to-use single-channel EEG during an auditory cognitive assessment. Future longitudinal studies should further explore the potential utility of this tool for early PD diagnosis and as a potential biomarker in PD.

Psychophysical findings accumulated over the past several decades indicate that perceptual tasks such as similarity judgment tend to be performed on a low-dimensional representation of the sensory data. Low dimensionality is especially important for learning, as the number of examples required for attaining a given level of performance grows exponentially with the dimensionality of the underlying representation space. In this chapter, we argue that, whereas many perceptual problems are tractable precisely because their intrinsic dimensionality is low, the raw dimensionality of the sensory data is normally high, and must be reduced by a nontrivial computational process, which, in itself, may involve learning. Following a survey of computational techniques for dimensionality reduction, we show that it is possible to learn a low-dimensional representation that captures the intrinsic low-dimensional nature of certain classes of visual objects, thereby facilitating further learning of tasks involving those objects.

Nearest-neighbor correlation-based similarity computation in the space of outputs of complex-type receptive fields can support robust recognition of 3D objects. Our experiments with four collections of objects resulted in mean recognition rates between 84% (for subordinate-level discrimination among 15 quadruped animal shapes) and 94% (for basic-level recognition of 20 everyday objects), over a 40deg X 40deg range of viewpoints, centered on a stored canonical view and related to it by rotations in depth (comparable figures were obtained for image-plane translations). This result has interesting implications for the design of a front end to an artificial object recognition system, and for the understanding of the faculty of object recognition in primate vision.

We consider training classifiers for multiple tasks as a method for improving generalization and obtaining a better low-dimensional representation. To that end, we introduce a hybrid training methodology for MLP networks; the utility of the hidden-unit representation is assessed by embedding it into a 2D space using multidimensional scaling. The proposed methodology is tested on a highly nonlinear image classification task.

Simulation-based training is an effective tool for acquiring practical skills, specifically to train new surgeons in a controlled and hazard-free environment, it is however important to measure participants cognitive load to decide whether they are ready to go into a real surgery. In the present study we measured performance on a surgery simulator of medical students and interns, while their brain activity was monitored by a mobile EEG device. 38 medical studentswere underwent 3 experiments undergoing a task with Simbionix simulator, while their brain activity was measured using a single-channel EEG device (Aurora by Neurosteer). On each experiment, participants performed 3 repeats of a simulator task using laparoscopic hands. The retention between tasks was different on each experiment, to examine changes in performance and cognitive load biomarkers that occur during the task or as a results of night sleep consolidation. The participants behavioral performance improved with trial repetition in all 3 experiments. In Exps. 1 & 2, the theta band activity significantly decreased with better individual performance, as exhibited by some of the behavioral measurements of the simulator. The novel VC9 biomarker (previously shown to correlate with cognitive load), exhibited a significant decrease with better individual performance shown by all behavioral measurements. In correspondence with previous research, theta decreased with lower cognitive load and higher performance and the novel biomarker, VC9, showed higher sensitivity to load changes. Together, these measurements might be for neuroimaging assessment of cognitive load while performing simulator laparoscopic tasks. This could potentially be expanded to evaluate efficacy of different medical simulations to provide more efficient training to medical staff and to measure cognitive and mental load in real laparoscopic surgeries.

Post-operative cognitive decline is a well-known phenomenon and of crucial importance especially in the elderly. General anesthesia can be accomplished by inhalation-based (volatile) or total intravenous anesthesia (TIVA). While their effects on post-operative symptoms have been investigated, little is known about their influence on brain functionalities during the surgery itself. To assess differences 17 patients were divided to receive either volatile anesthesia (n=9), or TIVA (n=8). The level of anesthesia was kept to be equal in both groups. A single bipolar EEG electrode (Neurosteer system) was placed on the participants foreheads. It presented real-time activity and collected their data during the surgery. The dependent variables included frequency bands (delta, theta, alpha, and beta), and three features (VC9, ST4, and A0) previously extracted with the device and provided by Neurosteer. All surgeries were uneventful, and all patients showed bispectral index (BIS) score less than 60. Feature activity under volatile anesthesia (in comparison to TIVA) was significantly lower for the delta, theta and alpha frequency bands and for the three features. Further analysis showed that the largest difference between anesthesia types was for feature A0. The EEG frequency bands and novel brain activity features provide evidence that volatile anesthesia further reduces components of brain activity in comparison to TIVA anesthesia. Specifically, A0, which previously showed a correlation with cognitive decline severity and cognitive load, exhibited the most prominent difference between anesthesia types. Together, this study suggests that measuring brain activity during anesthesia using sensitive features, enables revealing that different anesthesia types may affect brain activity differently, which could affect the recovery from anesthesia, and consequently reduce post-operative cognitive decline

Working Memory (WM) load is an important cognitive feature that is highly correlated with mental effort. Several neurological biomarkers such as theta power and mid-frontal activity show increased activity with increasing WM load. Such correlations often break down in cognitively impaired individuals, making WM load biomarkers a valuable tool for the detection of cognitive impairment. However, most studies have used a multi-channel EEG or an fMRI, which are not massively accessible. In the present study, we evaluate the ability of novel features extracted from a single-channel EEG located on the forehead, to serve as markers of WM load. We employed the widely used n-back task to manipulate WM load. Fourteen participants performed the n-back task while their brain activity was recorded with the Neurosteer inc Aurora EEG device. The results showed that the activity of the newly introduced features increased with WM load, similar to the theta band, but exhibited higher sensitivity to finer WM load changes. These more sensitive biomarkers of WM load are a promising tool for mass screening of mild cognitive impairment., Comment: 25 pages, 6 figures, 3 tables, 1 appendix