Brain Oscillatory Modes as a Proxy of Stroke Recovery.

The study did not include a sham stimulation condition, so auditory or somatosensory effects from the TMS pulses cannot be entirely ruled out as confounding factors. The patient cohort was predominantly mildly to moderately impaired, so results may not apply to those with severe motor deficits. Attrition was significant: only 27 patients completed all three time points, which limits the statistical power of the longitudinal analysis. The analysis was at the group level. It cannot currently predict individual patient recovery or tell a clinician exactly how much a specific patient will improve based on their brain signals. The exact timing of the 'disinhibition' phase remains unclear, and the study cannot determine if the brain signal changes cause recovery or are merely a byproduct of it.

Dr Hummel serves on the board of the Novartis Foundation. Dr Blanke is a co-founder/shareholder of Metaphysiks Engineering and a board member/shareholder of Mindmaze. Funding came from ETH Domain, Defitech Foundation, SNSF, and the Wyss Center.

Do not use these brain oscillation patterns to make individual predictions about patient recovery. The analysis was done on group averages, and the authors explicitly state that the method prevents direct inference of physiological changes at the individual level.