Advancing time series completion via RFAMoE and MDFF

TL;DR

This paper introduces a novel MoE-based diffusion framework for medical time series completion, improving accuracy and efficiency by adaptively selecting receptive fields and generating fused signals in a single inference.

Contribution

The paper proposes RFAMoE and MDFF modules that enhance diffusion models for medical time series imputation, reducing computational costs and improving reconstruction quality.

Findings

Outperforms state-of-the-art diffusion methods on multiple datasets

Reduces computational cost and latency in signal reconstruction

Achieves superior accuracy in medical time series completion

Abstract

Recent studies show that using diffusion models for time series signal reconstruction holds great promise. However, such approaches remain largely unexplored in the domain of medical time series. The unique characteristics of the physiological time series signals, such as multivariate, high temporal variability, highly noisy, and artifact-prone, make deep learning-based approaches still challenging for tasks such as imputation. Hence, we propose a novel Mixture of Experts (MoE)-based noise estimator within a score-based diffusion framework. Specifically, the Receptive Field Adaptive MoE (RFAMoE) module is designed to enable each channel to adaptively select desired receptive fields throughout the diffusion process. Moreover, recent literature has found that when generating a physiological signal, performing multiple inferences and averaging the reconstructed signals can effectively reduce reconstruction errors, but at the cost of significant computational and latency overhead. We design a Fusion MoE module and innovatively leverage the nature of MoE module to generate K noise signals in parallel, fuse them using a routing mechanism, and complete signal reconstruction in a single inference step. This design not only improves performance over previous methods but also eliminates the substantial computational cost and latency associated with multiple inference processes. Extensive results demonstrate that our proposed framework consistently outperforms diffusion-based SOTA works on different tasks and datasets.