With the significant advancement in the Internet of Things (IoT), Streaming Federated Learning (SFL) as a novel distributed learning approach can deal with time-varying streaming data among multiple sources. Standard SFL protocol is a collaborative training framework that enables many clients bounded with different online data sources to participate in a continuous training task. However, existing works ignore the cold-start problem and insufficient training data obstacle. Besides, due to the client heterogeneity and forgetting problem, the global model faces performance degradation during the time-series streaming data. In our work, we propose a digital twin-enabled SFL, a novel federated learning system with digital twin support to augment training data on demand. Instead of adopting an asynchronous federated learning protocol or buffer technique to wait for clients to have enough data, Generative adversarial network-based digital twins are introduced to construct a virtual replica for each federated learning client to generate a synthetic dataset based on the real data stream. We conduct the experiments using real-world datasets to evaluate the proposed SFL framework. The results under multiple data stream scenarios and various client behaviors demonstrate that our work outperforms the state-of-the-art baseline.

Accurate and efficient bacterial detection is essential for public health and medical diagnostics. However, traditional detection methods are constrained by limited dataset size, complex bacterial morphology, and diverse detection environments, hindering their effectiveness. In this study, we present EagleEyeNet, a novel multi-scale information fusion model designed to address these challenges. EagleEyeNet leverages large models as teacher networks in a knowledge distillation framework, significantly improving detection performance. Additionally, a newly designed feature fusion architecture, integrating Transformer modules, is proposed to enable the efficient fusion of global and multi-scale features, overcoming the bottlenecks posed by Feature Pyramid Networks (FPN) structures, which in turn reduces information transmission loss between feature layers. To improve the model’s adaptability for different scenarios, we create our own QingDao Bacteria Detection (QDBD) dataset as a comprehensive evaluation benchmark for bacterial detection. Experimental results demonstrate that EagleEyeNet achieves remarkable performance improvements, with mAP50 increases of 3.1% on the QDBD dataset and 4.9% on the AGRA dataset, outperforming the State-Of-The-Art (SOTA) methods in detection accuracy. These findings underscore the transformative potential of integrating large models and deep learning for advancing bacterial detection technologies.