Decoding Rehabilitation: Neural Markers for Assessing Exercise Intervention Effectiveness in Drug Addicts

To objectively evaluate the intervention effectiveness of exercise rehabilitation on drug addicts, this study proposes a neural-assessment method for measuring rehabilitation efficacy. Taking subjects from isolation rehabilitation centers as the research objects, they were divided into an experimental group (receiving exercise rehabilitation training), a control group (not receiving exercise rehabilitation training), and a newly admitted group (just admitted to the rehabilitation center). Brain activities of the three groups under resting state and audio-stimulated state were evaluated, and power features and nonpower features were extracted. A comparative study between groups was conducted by combining statistical analysis and machine learning models. The results show that the power index of the FPz channel is the most sensitive for distinguishing whether exercise intervention is received, and the nonpower features of the FP1 and FP2 channels are the core basis for identifying different withdrawal stages. Various machine learning models have achieved effective identification of subjects in different intervention states and withdrawal stages, confirming the reliability and potential of our method for evaluating exercise rehabilitation effectiveness. This study provides technical support and theoretical basis for optimizing exercise-assisted rehabilitation strategies and improving drug control governance effectiveness.