Design and characterization of a low-cost, open-source transcranial stimulation device (TES) for brain neuromodulation

Nowadays, there are several studies suggesting that neural activity in brain cortex could be modulated by noninvasive transcranial electrical stimulation (TES). These studies have shown potential benefits of TES inducing changes in cortical excitability by applying weak electrical currents to the scalp. TES is currently studied in a wide range of conditions that range from psychiatric diseases to chronic pain. In this work, we designed and manufactured an open source TES device using integrated low-cost components. The TES device hardware is composed of four different stages: Signal generation, signal amplification-adaptation, impedance isolation and safety stage. For the signal generation stage, we use a 32-bit ARM core microcontroller (Atmel SAM3X8E) that has his own DAC module. For the remaining stages, we use low-cost integrated components, including OPAMS, instrumentation amplifiers and a digital potentiometer (X9C104P). On the other hand, our TES device is programmed with C++ language and has two different open source algorithms: calibration and stimulation. The calibration scrip indirectly senses the current that is passing through the electrodes and amplifies the signal if it is necessary until the set current value is reached. Additionally, stimulation script generates and sends a stimulus signal when the microcontroller receives the instruction by serial port. Electrical tests were performed, and the results shows that our device can reliably generate several positive waveforms including: sine, pulse trains, ramp, sawtooth, random noise and square signals between frequencies of 0.1 and 100 Hz. Regarding the impedance isolation stage, we found that the device can maintain a current value of 200μA for resistances up to 70KOhm which is desirable considering the scalp resistance. Finally, with respect to safety stage results, we found that microcontroller stops stimulation 20ms after the detection of a not set value of current. Results validated that our device reach common TES parameters found in the literature. We think that this device could be used in up to date research or clinical studies (like close loop stimulation) with the added value that it is a lower-cost open-source device.