PPGFlowECG：用潜在整流流与跨模态编码实现 PPG 引导的 ECG 生成与心血管疾病检测
PPGFlowECG: Latent Rectified Flow with Cross-Modal Encoding for PPG-Guided ECG Generation and Cardiovascular Disease Detection
PPGFlowECG 是一个两阶段框架，先用 CardioAlign Encoder 将 PPG 与 ECG 对齐到共享潜在空间，再用潜在整流流合成 ECG，论文形式化分析表明该编码器是稳定且语义一致生成的必要条件。在四个数据集上的实验显示其合成保真度与下游诊断效用均有提升，有望在无法获取标准 ECG 时支持可穿戴设备优先的心血管疾病筛查。该工作已被 IEEE ICDM 2026 接收。
Authors:Xiaocheng Fang, Jiarui Jin, Haoyu Wang, Che Liu, Jieyi Cai, Yujie Xiao, Guangkun Nie, Bo Liu, Shun Huang, Hongyan Li, Shenda Hong
Abstract:Electrocardiography (ECG) is the clinical gold standard for cardiovascular disease (CVD) assessment, yet continuous monitoring is constrained by the need for dedicated hardware and trained personnel. Photoplethysmography (PPG) is ubiquitous in wearable devices and readily scalable, but it lacks electrophysiological specificity, limiting diagnostic reliability. While generative methods aim to translate PPG into clinically useful ECG signals, existing approaches are limited by the misalignment of physiological semantics in generative models and the complexity of modeling in high-dimensional signals. To address these limitations, we propose PPGFlowECG, a two-stage framework that aligns PPG and ECG in a shared latent space using the CardioAlign Encoder and then synthesizes ECGs with latent rectified flow. We further provide a formal analysis of this coupling, showing that the CardioAlign Encoder is necessary to guarantee stable and semantically consistent ECG synthesis under our formulation. Extensive experiments on four datasets demonstrate improved synthesis fidelity and downstream diagnostic utility. These results indicate that PPGFlowECG supports scalable, wearable-first CVD screening when standard ECG acquisition is unavailable.
| Comments: | Accepted by IEEE ICDM 2026 |
| Subjects: | Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Signal Processing (eess.SP) |
| Cite as: | arXiv:2509.19774 [cs.LG] |
| (or arXiv:2509.19774v3 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2509.19774 arXiv-issued DOI via DataCite |
Submission history
From: Xiaocheng Fang [view email]
[v1]
Wed, 24 Sep 2025 05:54:33 UTC (8,667 KB)
[v2]
Wed, 21 Jan 2026 15:58:20 UTC (5,706 KB)
[v3]
Tue, 13 Oct 2026 00:57:05 UTC (4,630 KB)
来源：arXiv cs.AI · arxiv.org