轻量级医疗数据分类框架，边缘设备上能效提升350倍
HD3C: Efficient Medical Data Classification for Edge Devices
- 用高维超向量编码数据，聚类生成原型，通过相似性搜索分类
- 心音分类任务中能效比贝叶斯残差网络高350倍，准确率仅低1%
- 对噪声、数据少、硬件错误都鲁棒，适合真实边缘部署
高效医疗数据分类对现代疾病筛查至关重要，尤其在资源受限环境下，其功耗预算和计算能力均受限制。我们提出HD3C，一种专为低功耗边缘设备设计的轻量级分类框架。该框架将数据编码为高维超向量，聚合生成多个聚类原型，并通过超空间中的相似性搜索实现分类。我们在三个医疗分类任务上评估了HD3C；在心音分类任务中，其能效比贝叶斯残差网络（Bayesian ResNet）高350倍，且准确率差异小于1%。此外，理论分析与实证结果均表明，HD3C对噪声、训练数据有限及硬件错误具有优异鲁棒性，展现出在真实场景中可靠部署的潜力。
原文摘要 · Abstract (English)
Efficient medical data classification is essential for modern disease screening, particularly in resource-constrained environments where power budgets and computing capabilities are limited. We present HD3C, a lightweight classification framework designed for low-power edge devices. HD3C encodes data into high-dimensional hypervectors, aggregates them into multiple cluster prototypes, and performs classification through similarity search in hyperspace. We evaluate HD3C across three medical classification tasks; on heart sound classification, HD3C is 350x more energy-efficient than Bayesian ResNet with less than 1% difference in accuracy. Moreover, HD3C demonstrates exceptional robustness to noise, limited training data, and hardware error, supported by both theoretical analysis and empirical results, highlighting its potential for reliable deployment in real-world settings.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。