Research on the construction of biosensor-assisted mental health monitoring system and talent training strategy

In recent years, mental health monitoring has become increasingly crucial due to the rising awareness of mental health issues and the demand for effective interventions. In this field, a biosensor-assisted mental health monitoring system is an important development that utilizes technology to distribute real-time information on physiological reactions connected to psychological and emotional conditions. The study intends to increase a talent training strategy and a biosensor-assisted mental health monitoring system using deep learning (DL) techniques. This investigation contains 453 participants enrolled in talent training programs that incorporate problem-solving games and theoretical understanding. After the training programs, the data is gathered from biosensors to monitor mental health. The sensor data is preprocessed using bandpass filtering to eliminate noise from the obtained data. The preprocessed data features are extracted using a Convolutional Neural Network (CNN). This study proposed an innovative Refined Prairie Dog-Optimized Poly-Kernel Support Vector Machine (RPDO-PSVM) model to predict mental health after talent training programs. RPDO optimizes the features selected from data, and PSVM predicts mental health. In a comparative analysis, the research determines the different evaluation metrics like accuracy (96%), precision (93.8%), recall (92.1%), and F1-score (94.4%). The conclusion indicates that the suggested method performs better than the forecast for monitoring mental health. The research highlights that the combination of advanced biosensor technology and strategic training offers a promising pathway for improving mental health outcomes.