Paralyzed man flies virtual drone with thought-controlled finger movements

In a demonstration of “mind over matter,” a 69-year-old man with C4 AIS C spinal cord injury — whose remaining movement was largely restricted to low-amplitude muscle twitching — has piloted a virtual quadcopter merely by thinking about moving his paralyzed fingers. This achievement stems from an intracortical brain-computer interface (iBCI) primarily developed and tested at Stanford University through the BrainGate2 clinical trials program. One of the lead authors, Matthew S. Willsey, Ph.D., is now an assistant professor of neurosurgery and biomedical engineering at the University of Michigan, as noted in a recent press release.

In 2016, two 96-channel microelectrode arrays were placed in the anatomically identified “hand knob” area of his left precentral gyrus, enabling real-time decoding of his neural signals to drive the drone’s maneuvers.

These electrodes record neural signals when T5 attempts to move distinct groups of fingers. Using the 4D decoder, initially, “the mean acquisition time was 1.98 ± 0.05 s, target acquisition rate was 64 ± 4 targets/min, and 98.7% of trials were successfully completed.” After becoming more accustomed to the task in the final 4 blocks, performance improved to “1.58 ± 0.06 s (a target acquisition rate of 76 ± 2 targets/min), and 100% of trials were completed.”

To navigate the virtual quadcopter, each finger movement was mapped to a degree of freedom, with the system directly linking finger positions to velocity control. As Matthew S. Willsey explained in a press release,

“It takes the signals created in the motor cortex that occur simply when the participant tries to move their fingers and uses an artificial neural network to interpret what the intentions are to control virtual fingers in the simulation. Then we send a signal to control a virtual quadcopter.”

Addressing unmet needs and enhancing social inclusion

“More than 5 million people in the United States live with severe motor impairments.” “Although many basic needs of people with paralysis are being met, unmet needs for peer support, leisure activities, and sporting activities are reported, respectively, by 79%, 50%, and 63% of surveyed people with paralysis from spinal cord injury,” according to
the bioRxiv preprint.

For quadcopter performance comparison, the paper states: “Compared with a state-of-the-art electroencephalographic (EEG)-controlled quadcopter that navigated through 3.1 rings in 4 minutes, our system allowed navigation through or around 18 rings – at peak performance – in less than 3 minutes on a similar flight path, a more than sixfold increase in performance.”

From gaming to telepresence

“This approach shows promise for controlling multiple-DOF end-effectors, such as robotic fingers or digital interfaces for work, entertainment, and socialization.” By restoring fine motor control at the finger level, users might eventually type, play musical instruments, or operate sophisticated game controllers with greater dexterity. Researchers suggest it could also extend to social networking or remote work, offering new avenues for independence and engagement in daily life.

Toward clinical adoption

“Although high performance was achieved using this decoding system, potential improvements to increase the likelihood of clinical adoption include reducing the calibration times and increasing the robustness to neural instabilities.” Increasing electrode channel counts could further boost decoding accuracy — benefiting current applications like robotic arms or virtual finger controllers, and paving the way for new and more advanced use cases.

Ultimately, the feasibility of precise iBCI control may address significant unmet needs in leisure and social engagement for people with paralysis, underscoring its promise for real-world tasks like gaming, virtual reality, and online collaboration.