By Faisal Khan | 28 July 2026
Medium
Brain is the most important organ of the body — be it lifting your hand to perform an easy task or solving a complex mathematical problem, the neuron pathways in this part of the body process these tasks. Decoding this brain activity can open up a wide array of medical possibilities, which could lead to effective screening against various diseases.
We have already seen a lot of research being conducted on how to utilize technology to treat various health conditions. While nanotechnology is seen as having an enormous role to play in the future of healthcare, we are already seeing some practical applications being tested via Brain-Computer interfaces (BCI).
Last year, I wrote about how a mind-reading exoskeleton let a completely paralyzed man walk again. Then a few months ago, there was news about a BCI, which restored a paralyzed man’s vital senses. Also worth noticing was a groundbreaking study in which researchers used an AI that converted brain activity into text — helping patients to communicate, who are unable to speak or type.
“We see this as trying to build a partnership between two learning systems — brain and computer — that ultimately lets the artificial interface become an extension of the user, like their own hand or arm.”
~ Karunesh Ganguly, Senior Author, Study
Building on all this research, scientists have now developed a Plug-and-Play BCI device that utilizes machine learning techniques to enable the patients in using a prosthetic limb. Researchers at the University of California, San Francisco (UCSF) have come up with a new BCI that allows the sufferers of paralysis to regain control of prosthetic devices and limbs via their brain signals.
⚡ #LongCOVID “brain fog” is beginning to reveal a measurable electrical signature in the brain.
➡️ EEG recordings from people with LongCOVID showed abnormal activity in the sensorimotor cortex, including a weaker post-movement beta rebound and slower alpha rhythms. 1/ pic.twitter.com/KZPoWQLIYy
— Vipin M. Vashishtha (@vipintukur) August 19, 2026
Although BCIs have come a long way, the existing systems need to be reset and recalibrated each day to tap into the brain’s natural learning processes. This is where this process exceeds expectations. Utilizing machine learning algorithms, the system enables a brain & computer ‘partnership’, with the resulting software that translates brain activity into action. Also, this is the first time an artificial learning system has been shown to work smoothly with the brain’s sophisticated long-term learning processes.
The BCI used in the device is called an ECoG array — a pad of electrodes around the size of a Post-it note that is surgically implanted on the surface of the brain (figure above). Contrary to the previous BCI implants, consisting of sharp electrodes that lost signal over time, the ones used for this experiment were found not to require daily calibration. The results also showed that the paralyzed person’s brain optimized its activity to control the BCI.
According to the study, the participant was not only enrolled in a clinical trial designed to test the use of ECoG arrays to allow paralyzed patients to control a prosthetic arm and hand, but also used the implant to control a computer cursor on a screen. Originally, the machine learning algorithm had to be reset every day — gradually learning to match the user’s desired motions with the actual movement of the cursor on the screen. This process, however, took a lot of time.
Eventually, researchers tweaked the algorithm enabling it to continue learning about the user’s brain activity and desired movements, without resetting and starting from scratch each day. This created a bridge between the bridge and the AI — resulting in continuous improvements in performance over many days.
In a Melbourne lab, a volunteer swallowed a psilocybin capsule, slid into an MRI scanner, and began to merge with it.
"Melding with the MRI machine, floor, walls, air," they reported afterward.
Another volunteer described losing all sense of self and becoming "at one with all… pic.twitter.com/kbZgJaTzec
— Paul F. Austin (@PaulAustin3w) August 19, 2026
Even switching off the Plug-n-Play device’s auto-updating feature and the user could simply plugin and start using it each day. The team found out that the performance of the system did not decline, despite a lack of daily calibration over a period of 44-days. Not using the device at all for several days in between only resulted in a minor performance decline.
Reprinted with permission from the author.
Faisal Khan is a prolific Canada-based tech blogger and influencer. He is the founder and editor of the Technicity publication which focuses on technical, scientific and financial knowledge sharing. Follow him on Twitter @fklivestolearn.
Brain-Computer Interface: No Open Brain Surgery Required
The New Generation of Brain-Computer Interface
Mysteries of the Brain: Brain-Computer Interface
Be sure to ‘like’ us on Facebook
