This article explores the user-centered theory in computational smart healthcare service systems (SHSS) and proposes a comprehensive SHSS solution based on digital twin (DT) and decentralized federated learning (DFL). To clarify the interaction between users, products, and services in SHSS, a hybrid framework called user-centered data-driven product and smart healthcare service system (UDP-SHSS) is proposed. The key challenges related to medical data privacy protection, real-time user state monitoring and data updates, and the integration of resources across decentralized medical institutions are identified through the analysis of this framework. To address these challenges, the DT-empowered DFL Framework is designed, where DT maps physical hospital entities to the virtual decentralized network and DFL offers a decentralized model interaction scheme to the decentralized network. To construct the dynamic real-time and privacy-preserving objectives of UDP-SHSS, the DT-based decentralized online federated learning (DT-DOFL) algorithm is developed, which models the framework with graph theory to account for the real-world communication scenarios between clients and incorporates differential privacy (DP) mechanisms to ensure user privacy. Given the black-box nature of DFL models, the one-point bandit feedback (OPBF) mechanism is employed to address the online convex optimization problem in UDP-SHSS. Rigorous proofs demonstrating that the algorithm achievesϵ-DP and sublinear convergence are given. Finally, several experiments on various diseases in SHSS are conducted to verify the effectiveness of our framework and algorithm.

During the age of the Industrial Internet of Things (IIoT), extensive sensors are deployed on steel production lines to construct intelligent monitoring systems. A fan is a crucial piece of machinery in steel production lines, making its fault diagnosis imperative to prevent air pollution and casualties. DNN (Deep Neural Network) with powerful real-time IIoT data analysis has achieved outstanding performance in recognizing faults. Due to the black-box nature of DNNs, these models cannot provide reasonable explanations for their diagnostic decisions. It is still challenging for experts to make reliable and trustworthy conclusions. To address the issue, this paper introduces a new neural-symbolic model, termed Confidence and Classification DBN (CC-DBN), where confidence and classification rules are extracted from a Deep Belief Network (DBN) to provide an explainable representation of DBN feature learning and reasoning. In order to extract confidence rules, this paper develops a new clustering logic Restricted Boltzmann Machine (C-LRBM). Confidence rules can generate latent features of the raw vibration data of the fan and simultaneously explain the hierarchical reasoning of stacked RBM. Besides, to make trustworthy fan diagnosis decisions, classification rules are extracted to provide an explainable symbolic representation between input and output feature spaces. The experiment is performed on an industrial fan dataset from a leading steel production line in Shanghai. The results demonstrate that the proposed CC-DBN can effectively discover knowledge for fan diagnostic decisions and simultaneously achieve superior fault discrimination over typical classifiers and DBNs.