Electroencephalography-based endogenous brain-computer interface for online communication with a completely locked-in patient.
Chang-Hee Han, Yong-Wook Kim, Do Yeon Kim and 3 others
PMID 30700310WHAT IT FOUND
In a single patient with completely locked-in ALS, an EEG-based mental-imagery BCI classified yes/no answers online with 87.5% average accuracy across four runs, but this case cannot prove the method works for other patients.
Key findings
01Online classification of left-hand motor imagery and mental subtraction averaged 87.5% accuracy across four runs.
02A classifier trained on data recorded about 6 months earlier still reached 85% accuracy, and combining old and new training data reached 90% accuracy.
03Four months after the online experiment, no evidence of consciousness or alertness was detected.
STILL TO COME
How it was doneWhat they foundWhat it means for SLPs
Read the rest of this summary
You get three full summaries a month, free, and we do not ask for a card. Search, the TL;DRs and your library stay unlimited either way.
What it does not show
Only a single patient was studied, so the result cannot show whether the BCI works for other patients. The patient could not take part in follow-up experiments after the online test, and her status worsened after acute pneumonia. The four online run accuracies were 100%, 90%, 80%, and 80%, and mental fatigue was not quantitatively measured. Four months later, no evidence of consciousness or alertness was detected, which limits any claim about lasting communication.
Declared interests
The supplied text does not include a funding or conflicts-of-interest declaration. The publication types list Research Support, Non-U.S. Gov't.
The easy way to misread this
Do not read this as proof that EEG-based BCI can reliably restore communication for patients with completely locked-in ALS. The result comes from a single patient, and four months after the online experiment no evidence of consciousness or alertness was detected.