The Portiloop: a deep learning-based open science tool for closed-loop brain stimulation

Abstract

Electroencephalography (EEG) is a method of measuring the brain's electrical activity, using non-invasive scalp electrodes. In this article, we propose the Portiloop, a deep learning-based portable and low-cost device enabling the neuroscience community to capture EEG, process it in real time, detect patterns of interest, and respond with precisely-timed stimulation. The core of the Portiloop is a System on Chip composed of an Analog to Digital Converter (ADC) and a Field-Programmable Gate Array (FPGA). After being converted to digital by the ADC, the EEG signal is processed in the FPGA. The FPGA contains an ad-hoc Artificial Neural Network (ANN) with convolutional and recurrent units, directly implemented in hardware. The output of the ANN is then used to trigger the user-defined feedback. We use the Portiloop to develop a real-time sleep spindle stimulating application, as a case study. Sleep spindles are a specific type of transient oscillation (∼2.5 s, 12-16 Hz) that are observed in EEG recordings, and are related to memory consolidation during sleep. We tested the Portiloop's capacity to detect and stimulate sleep spindles in real time using an existing database of EEG sleep recordings. With 71% for both precision and recall as compared with expert labels, the system is able to stimulate spindles within ∼300 ms of their onset, enabling experimental manipulation of nearly the entire spindle. The Portiloop can be extended to detect and stimulate other neural events in EEG. It is fully available to the research community as an open science project 1 .

Automated real‐time EEG sleep spindle detection for brain‐state‐dependent brain stimulation

“…Removing the criterion of detecting a spindle centre may allow earlier targets within a spindle but will also increase the false-positive rate; whether or not this is acceptable eventually depends on the study goals. This limitation may also be overcome in the future by using additional information from the data (such as power envelope gradients) or by training deep learning approaches to predict spindles during very early phases or even before their initiation (Valenchon et al, 2021).…”

“…Lustenberger et al (2016) used transcranial alternating current, while Antony et al (2018) and Choi et al (2019) used auditory stimulation to target ongoing spindles, but only basic information and no validation procedure was provided regarding the real-time algorithm and its phase specificity. The Portiloop system, a portable system on chip, at least achieved real-time spindle detection with an F1-Score of 71% by training artificial neural networks on a field-programmable gate array (Valenchon et al, 2021).…”

Automated real-time EEG sleep spindle detection for brain state-dependent brain stimulation

“…Removing the criterion of detecting a spindle center may allow earlier targets within a spindle but will also increase the false positive rate; whether or not this is acceptable eventually depends on the study goals. This limitation may also be overcome in the future by using additional information from the data (such as power envelope gradients) or by training deep learning approaches to predict spindles during very early phases or even before their initiation (Valenchon et al, 2021).…”

“…Lustenberger et al (2016) targeted ongoing spindles with transcranial alternating current stimulation (TACS) but provided limited information regarding the real-time algorithm and its phase specificity. The Portiloop system, a portable system on chip, at least achieved real-time spindle detection with an F1-Score of 0.71 by training artificial neural networks on a field-programmable gate array (FPGA) (Valenchon et al, 2021).…”