RESUMEN

Spike sorting - the process of separating spikes from different neurons - is often the first and most critical step in the neural data analysis pipeline. Spike-sorting techniques isolate a single neuron's activity from background electrical noise based on the shapes of the waveforms obtained from extracellular recordings. Despite several advancements in this area, an important remaining challenge in neuroscience is online spike sorting, which has the potential to significantly advance basic neuroscience research and the clinical setting by providing the means to produce real-time perturbations of neurons via closed-loop control. Current approaches to online spike sorting are not fully automated, are computationally expensive and are often outperformed by offline approaches. In this paper, we present a novel algorithm for fast and robust online classification of single neuron activity. This algorithm is based on a deep contractive autoencoder (CAE) architecture. CAEs are neural networks that can learn a latent state representation of their inputs. The main advantage of CAE-based approaches is that they are less sensitive to noise (i.e., small perturbations in their inputs). We therefore reasoned that they can form the basis for robust online spike sorting algorithms. Overall, our deep CAE-based online spike sorting algorithm achieves over 90% accuracy in sorting unseen spike waveforms, outperforming existing models and maintaining a performance close to the offline case. In the offline scenario, our method substantially outperforms the existing models, providing an average improvement of 40% in accuracy over different datasets.

Asunto(s)

Modelos Neurológicos , Procesamiento de Señales Asistido por Computador , Potenciales de Acción/fisiología , Algoritmos , Neuronas/fisiología

RESUMEN

OBJECTIVES: Psychogenic non-epileptic seizures (PNES) have been hypothesized to emerge in the context of neural networks instability. To explore this hypothesis in children, we applied a graph theory approach to examine connectivity in neural networks in the resting-state EEG in 35 children with PNES, 31 children with other functional neurological symptoms (but no PNES), and 75 healthy controls. METHODS: The networks were extracted from Laplacian-transformed time series by a coherence connectivity estimation method. RESULTS: Children with PNES (vs. controls) showed widespread changes in network metrics: increased global efficiency (gamma and beta bands), increased local efficiency (gamma band), and increased modularity (gamma and alpha bands). Compared to controls, they also had higher levels of autonomic arousal (e.g., lower heart variability); more anxiety, depression, and stress on the Depression Anxiety and Stress Scales; and more adverse childhood experiences on the Early Life Stress Questionnaire. Increases in network metrics correlated with arousal. Children with other functional neurological symptoms (but no PNES) showed scattered and less pronounced changes in network metrics. CONCLUSION: The results indicate that children with PNES present with increased activation of neural networks coupled with increased physiological arousal. While this shift in functional organization may confer a short-term adaptive advantage-one that facilitates neural communication and the child's capacity to respond self-protectively in the face of stressful life events-it may also have a significant biological cost. It may predispose the child's neural networks to periods of instability-presenting clinically as PNES-when the neural networks are faced with perturbations in energy flow or with additional demands.