Leveraging Imperfect Medical Data: A Manifold-Consistent Spatio-Temporal Network for Sensor-based Human Activity Recognition

TL;DR

This paper introduces a robust spatio-temporal neural network for sensor-based human activity recognition that effectively handles real-world sensor imperfections like noise and missing data.

Contribution

It proposes a dual-level corruption modeling mechanism and a dual-stream architecture to learn stable, corruption-invariant representations for improved HAR performance.

Findings

Achieves competitive results on PAMAP2, Opportunity, and WISDM datasets.

Demonstrates robustness under sensor noise and missing data conditions.

Outperforms existing methods in imperfect sensing scenarios.

Abstract

Sensor-based Human Activity Recognition (HAR) has attracted increasing attention in medical and healthcare monitoring, particularly with the growth of Internet of Medical Things (IoMT). However, in real-world wearable sensing scenarios, IoMT signals are often corrupted by missing measurements, sensor failures, and environmental noise, which significantly degrade the performance of conventional deep learning models that assume clean and complete inputs. To address this challenge, we propose a Manifold-Consistent Spatio-Temporal Network (MCSTN) for robust HAR under imperfect sensing conditions. The proposed framework introduces a dual-level corruption modeling mechanism that simulates realistic sensor imperfections through both physical-level corruption and diffusion-driven continuous corruption. By enforcing representation consistency across multiple corrupted views, the model learns stable and corruption-invariant semantic representations. Furthermore, we design a dual-stream spatio-temporal architecture that explicitly decouples temporal dynamics modeling and spatial correlation learning. The temporal stream captures long-term activity dynamics, while the spatial stream models inter-sensor relationships, enabling more effective spatio-temporal representation learning. Extensive experiments on three widely used HAR benchmark datasets, PAMAP2, Opportunity, and WISDM, demonstrate that the proposed MCSTN achieves competitive performance compared with existing state-of-the-art methods, particularly under imperfect sensing conditions. These results validate the effectiveness and robustness of the proposed framework for real-world wearable IoMT sensing applications.