Over the Horizon: The Present and Future of Endovascular Neural Recording and Stimulation

The past decade has witnessed an explosion in applications for neural recording and stimulation in the treatment of clinical disorders. Neuromodulatory approaches are now a mainstay of care for essential tremor and Parkinson's disease, and are expanding rapidly into a wide range of other neurological and psychiatric diseases. In parallel, advancements in endovascular approaches to cerebrovascular diseases have resulted in minimally invasive techniques that deliver devices to neural tissue in the central and peripheral nervous systems, with significantly improved safety and efficacy. In this review, we discuss the history of endovascular neural recording and stimulation, its current progress, and applications for neurological disease.

A Platform for Cognitive Monitoring of Neurosurgical Patients During Hospitalization

Intracranial recordings in epilepsy patients are increasingly utilized to gain insight into the electrophysiological mechanisms of human cognition. There are currently several practical limitations to conducting research with these patients, including patient and researcher availability and the cognitive abilities of patients, which limit the amount of task-related data that can be collected. Prior studies have synchronized clinical audio, video, and neural recordings to understand naturalistic behaviors, but these recordings are centered on the patient to understand their seizure semiology and thus do not capture and synchronize audiovisual stimuli experienced by patients. Here, we describe a platform for cognitive monitoring of neurosurgical patients during their hospitalization that benefits both patients and researchers. We provide the full specifications for this system and describe some example use cases in perception, memory, and sleep research. We provide results obtained from a patient passively watching TV as proof-of-principle for the naturalistic study of cognition. Our system opens up new avenues to collect more data per patient using real-world behaviors, affording new possibilities to conduct longitudinal studies of the electrophysiological basis of human cognition under naturalistic conditions.

Non-linear Embedding Methods for Identifying Similar Brain Activity in 1 Million iEEG Records Captured From 256 RNS System Patients

Finding electrophysiological features that are similar across patients with epilepsy may facilitate identifying treatment options for one patient that worked in patients with similar brain activity patterns. Three non-linear iEEG (intracranial electroencephalogram) embedding methods of finding similar cross-patient iEEG records in a large iEEG dataset were developed and compared. About 1 million iEEG records from 256 patients with drug-resistant focal onset seizures who were treated in prospective trials of the RNS System were used for analyses. Data from 200, 25, and 31 patients were randomly selected to be in the train, validation, and test datasets. In method 1, ResNet50 convolutional neural network (CNN) model pre-trained on the ImageNet dataset was used for extracting feature maps from spectrogram images (ImageNet-ResNet) of iEEG records. In method 2, ResNet50 custom trained on an iEEG classification task using ~138,000 manually labeled iEEG records was used as the feature extractor (ESC-ResNet). Feature maps were passed through dimensionality reduction and k nearest neighbors were found in the reduced feature space. In method 3, a 256 dimensional iEEG embedding space was learned via contrastive learning by training a ResNet50 model with triplet training sets generated using within-patient iEEG clustering (CL-ResNet). All three methods had comparable performance when identifying iEEG records from the search dataset similar to test iEEG records of baseline (non-seizure) and interictal spiking activity. Epileptic interictal spikes are represented by vertical (broadband) edges in spectrogram images, and hence even generic features extracted using models trained on everyday images appear to be sufficient to represent iEEG records with similar levels of interictal spiking activity in close proximity. In the case of electrographic seizures, however, the ESC-ResNet model, identified cross-patient iEEG records with electrographic seizure morphology features that were most similar to the test iEEG records. For nuanced electrographic seizure iEEG representation learning, domain specific model training with manually generated labels had the advantage. Finally, representative iEEG records were selected from every patient using an unsupervised clustering method which effectively reduced the number of iEEG records in the search dataset from ~750,000 to 2,148, thus substantially reducing the time required for finding similar cross-patient iEEG records.