OBJECTIVE: The aim of the present study is to examine the relationship between EEG measures and functional recovery in right-hemisphere stroke patients. METHODS: Participants with stroke (PS) and neurologically unimpaired controls (UC) were enrolled. At enrolment, all participants were assessed for motor and cognitive functioning with specific scales (motricity index, trunk control test, Level of Cognitive Functioning, and Functional Independence Measure (FIM). Moreover, EEG data were recorded. At discharge, participants were re-tested with the FIM RESULTS: Powers in the delta, theta, alpha, and beta bands and connectivity within the fronto-parietal network were compared between groups. Then, the between-group discriminative EEG measures and the motor/cognitive scales were used to feed a machine learning algorithm to predict FIM scores at discharge and the length of hospitalization (LoH). Higher delta, theta, and beta and impaired connectivity were found in PS compared to UC. Moreover, motor/cognitive functioning, beta power, and fronto-parietal connectivity predicted the FIM score at discharge and the LoH (accuracy=73.2 % and 85.2 % respectively). CONCLUSIONS: Results show that the integration of motor/cognitive scales and EEG measures can reveal the rehabilitative potentials of PS predicting their functional outcome and LoH. SIGNIFICANCE: Synergistic clinical and electrophysiological models can support rehabilitative decision-making.