An artificial neural-network approach to identify motor hotspot for upper-limb based on electroencephalography: a proof-of-concept study.
Ga-Young Choi, Chang-Hee Han, Hyung-Tak Lee and 3 others
PMID 34930380WHAT IT FOUND
An AI model using EEG data from healthy adults identified motor hotspots with a mean error of 0.22 cm compared to TMS.
The method works with as few as nine electrodes, but has not yet been tested in patients with motor impairment.
Key findings
01The artificial neural network estimated motor hotspot locations with a minimum mean error distance of 0.22 cm when using gamma-band EEG features from 63 channels.
02Accuracy decreased as the number of EEG channels was reduced, but a mean error distance of approximately 1.32 cm was still achieved using only nine channels.
03The study was conducted exclusively on healthy, right-handed adults, so the method's performance in patients with neurological disorders remains unverified.
STILL TO COME
How it was doneWhat they foundWhat it means for PTs
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
The study included only healthy adults, not patients with stroke or other neurological conditions. Brain activity and anatomy in patients differ significantly from healthy individuals, so the model's accuracy in clinical populations is unknown. All participants were right-handed. The study did not test whether the algorithm performs differently for left-handed individuals. The algorithm still requires an initial TMS session to 'register' the patient's specific motor hotspot before the EEG method can be used for subsequent sessions. The study did not test the efficacy of tES treatment itself, only the accuracy of locating the stimulation target.
Declared interests
Funding was provided by the Ministry of Communication and Information Technology, the National Research Foundation of Korea, and the Seoul National University Bundang Hospital Research Fund. No commercial conflicts of interest were declared.
The easy way to misread this
Do not assume this method is ready for clinical use. The high accuracy reported was achieved in healthy, right-handed adults using a complex setup. It has not been tested in patients with brain injuries, and it still requires a baseline TMS measurement to work.