Continuous electrocardiography (ECG) monitoring could surface rhythm abnormalities before they escalate into cardiovascular events. However, a deployable system must satisfy three requirements simultaneously: legal-grade privacy (GDPR, HIPAA), real-time inference on constrained edge hardware, and detection quality under non-IID cross-hospital data. We design and evaluate an end-to-end federated system addressing all three for unsupervised 12-lead ECG anomaly detection on PTB-XL dataset, combining three autoencoder families (VanillaAE, ConvAE, VAE), Flower-based federated averaging (FedAvg) across ten simulated hospitals, client-side differentially private SGD (DP-SGD) with a Rényi-DP accountant, and 8-bit integer (INT8) post-training quantization with Raspberry Pi 4 benchmarking. Our main contributions are: an empirical characterization of how these mechanisms compose, practical DP-specific recommendations, and technical and security insights for a clinically sensitive setting. Federated learning matches or exceeds the centralized baseline across all architectures (ConvAE federated area under the ROC curve, AUROC, $0.782$), and an $\varepsilon$ sweep identifies $\varepsilon=4$ as the recommended clinical operating point. INT8 quantization roughly halves model size and cuts Pi 4 latency by up to $44%$ with $<0.12%$ AUROC loss. Crucially, DP and quantization penalties are empirically independent, so practitioners need not trade a strong privacy guarantee for a compact edge footprint. To our knowledge, this is the first system combining federated learning, formal $(\varepsilon,δ)$-DP, unsupervised reconstruction-based detection, and quantized AArch64 deployment.

翻译：连续心电监测有助于在心血管事件恶化前发现心律异常。然而，可部署系统必须同时满足三项要求：法定级隐私保护（GDPR、HIPAA）、受限边缘硬件上的实时推理能力，以及跨医院非独立同分布数据下的检测质量。我们设计并评估了一个端到端联邦系统，在PTB-XL数据集上解决12导联无监督心电异常检测的所有上述需求，该系统结合三种自编码器家族（VanillaAE、ConvAE、VAE）、基于Flower的跨十家模拟医院的联邦平均算法、具备Rényi差分隐私计数器的客户端差分隐私随机梯度下降，以及基于树莓派4基准测试的8位整数后训练量化。主要贡献包括：这些机制组合效果的实证表征、实用的差分隐私专用建议，以及针对临床敏感场景的技术与安全见解。联邦学习在所有架构上均达到或超越集中式基线（ConvAE联邦ROC曲线下面积AUROC为0.782），且通过ε扫描确定ε=4为推荐临床操作点。INT8量化可将模型体积缩小约一半，树莓派4延迟降低高达44%，而AUROC损失低于0.12%。关键发现是差分隐私与量化的性能损失在经验上相互独立，因此从业者无需为获得紧凑的边缘部署而牺牲强隐私保证。据我们所知，这是首个将联邦学习、形式化(ε,δ)-差分隐私、无监督重构检测与量化AArch64部署相结合的系统。