Aiming to address critical challenges in piano player finger function health monitoring—such as insufficient real-time capability, limited personalization, and the need for improved early warning accuracy—this study proposes an intelligent risk assessment model that integrates Internet of Things (IoT) technology with deep learning. While most existing research focuses on general occupational health management, there is a lack of dynamic monitoring schemes specifically tailored to the hand activities of pianists. Traditional methods also struggle to capture the high-dimensional nonlinear characteristics of finger movements. Using multidimensional finger function data from 150 pianists sourced from the UCI Machine Learning Repository, this paper develops a health risk early warning model based on a deep feedforward neural network. The study addresses the following research questions: (1) How can high-dimensional finger movement data be collected and processed in real time? (2) How can an early warning mechanism for finger function degeneration be established? (3) How can the enhancement of performance skills be balanced with the protection of finger health? Experimental results show that the proposed model achieves 94.69% accuracy in health risk analysis, which is 15.88% higher than that of the comparison algorithm, and a recall rate of 95.64%, representing an increase of 16.74%. This research offers scientific training guidance and health protection strategies for piano players.
Wei Xiang (Thu,) studied this question.