Multi-Task Temporal Shift Attention Networks for On-Device Contactless Vitals Measurement

TL;DR

This paper introduces a novel multi-task temporal shift convolutional attention network for real-time, contactless vital sign measurement on mobile devices, achieving high accuracy and efficiency suitable for telehealth applications.

Contribution

The paper presents a new on-device neural network architecture that enables simultaneous cardiovascular and respiratory vital sign measurement in real-time, with significant accuracy improvements.

Findings

Achieves over 150 fps on ARM CPU for real-time processing

Reduces measurement error by 20-50% compared to previous methods

Generalizes well across multiple benchmark datasets

Abstract

Telehealth and remote health monitoring have become increasingly important during the SARS-CoV-2 pandemic and it is widely expected that this will have a lasting impact on healthcare practices. These tools can help reduce the risk of exposing patients and medical staff to infection, make healthcare services more accessible, and allow providers to see more patients. However, objective measurement of vital signs is challenging without direct contact with a patient. We present a video-based and on-device optical cardiopulmonary vital sign measurement approach. It leverages a novel multi-task temporal shift convolutional attention network (MTTS-CAN) and enables real-time cardiovascular and respiratory measurements on mobile platforms. We evaluate our system on an Advanced RISC Machine (ARM) CPU and achieve state-of-the-art accuracy while running at over 150 frames per second which enables real-time applications. Systematic experimentation on large benchmark datasets reveals that our approach leads to substantial (20%-50%) reductions in error and generalizes well across datasets.