SSVEP-based BCI for a DMD Patient – A Case Study

The present paper presents the adaptation made to yield a working 5-class steady-state visual evoked potential (SSVEP) based brain-computer interface (BCI) system for a Duchenne Muscular Dystrophy (DMD) patient. The patient was unable to use the "standard" BCI system which made use of a liquid crystal display (LCD) of a laptop screen as the visual stimulator. A detailed assessment of the patient’s electroencephalography (EEG) spectrogram suggested that he had an inherently active alpha rhythm that manifested itself regularly even when he was focusing on other stimulation frequencies. Besides, he also exhibited an extraordinarily noisy EEG spectrum which was probably caused by his uncontrollable jerking movements. To improve his BCI performance, we substituted the LCD stimulator with LEDs programmed to flicker at different frequencies and mounted on the two sides of the laptop using a custom aluminium frame. We also conducted another pre-test to identify five stimulation frequencies that the patient could use to evoke useful SSVEP responses. While there is still room for improvement, such adaptations had rendered him an acceptable level of BCI control with >60% classification accuracy. The present study can serve as a reference for researchers or clinicians that encounter similar problem especially in a clinical setting where patients showed undesirable performance dynamics.