A Dataset and Toolkit for Multiparameter Cardiovascular Physiology Sensing on Rings

TL;DR

τ-Ring introduces the first open, multi-parameter ring-based cardiovascular dataset and accompanying RingTool benchmarking suite to close gaps in data accessibility and reproducible evaluation. The dataset captures synchronized infrared and red PPG, ACC data from reflective and transmissive ring paths across 34 subjects, 28.21 hours, and 7 activities, with ground-truth HR, RR, SpO2, and BP. RingTool combines physics-based signal processing with four deep learning backbones (ResNet, InceptionTime, Transformer, Mamba) to benchmark HR, RR, SpO2, and BP estimation, achieving best overall HR MAE of 5.18 BPM (Ring 1) and RR MAE of 2.98 BPM (Ring 1), among other metrics, and showing supervised methods often outperform physics-based baselines, especially for stationary conditions. The work demonstrates the potential of ring-based sensing, highlights motion-related challenges, and provides a foundation for community-driven advances through open data and tools, with future work focusing on dataset expansion and on-device deployment optimization using model compression and adaptive fusion.

Abstract

Smart rings offer a convenient way to continuously and unobtrusively monitor cardiovascular physiological signals. However, a gap remains between the ring hardware and reliable methods for estimating cardiovascular parameters, partly due to the lack of publicly available datasets and standardized analysis tools. In this work, we present $τ$-Ring, the first open-source ring-based dataset designed for cardiovascular physiological sensing. The dataset comprises photoplethysmography signals (infrared and red channels) and 3-axis accelerometer data collected from two rings (reflective and transmissive optical paths), with 28.21 hours of raw data from 34 subjects across seven activities. $τ$-Ring encompasses both stationary and motion scenarios, as well as stimulus-evoked abnormal physiological states, annotated with four ground-truth labels: heart rate, respiratory rate, oxygen saturation, and blood pressure. Using our proposed RingTool toolkit, we evaluated three widely-used physics-based methods and four cutting-edge deep learning approaches. Our results show superior performance compared to commercial rings, achieving best MAE values of 5.18 BPM for heart rate, 2.98 BPM for respiratory rate, 3.22\% for oxygen saturation, and 13.33/7.56 mmHg for systolic/diastolic blood pressure estimation. The open-sourced dataset and toolkit aim to foster further research and community-driven advances in ring-based cardiovascular health sensing.

Figures (12)

• Figure 1: Ring design and sensor layout. This figure shows the design and sensor layout of our rings. (A) displays the physical structure of our ring. (D) shows the internal components including the battery and chip. (B) and (E) illustrate the sensor configurations for Ring 1 (reflective), showing the placement of LEDs and PDs. (C) and (F) illustrate the sensor configurations for Ring 2 (transmissive).
• Figure 4: Data collection procedure across stimulus-evoked physiological states. The protocol consists of three main activities: (1) A 10-minute seated resting, (2) A 9-minute supervised low-oxygen simulation, and (3) Two 2-minute sessions of deep squat exercises. Blood pressure measurements were taken before and after each activity, while physiological data was continuously recorded by our custom rings and periodically measured by commercial rings for comparison.
• Figure 5: Data collection procedure across daily activities. The protocol consists of five activity segments: (1) A 30-minute seated resting, (2) 5-minute sitting and talking, (3) 5-minute head movement, (4) 5-minute standing, and (5) 5-minute walking in place. Participants wore the oximeter, Ring 1 (reflective), Ring 2 (transmissive), and respiratory band throughout all activities.
• Figure 6: Ground Truth Distribution of HR, RR, SpO2, and BP.
• Figure 7: Overview of Our RingTool Structure. RingTool offers various tools for researchers to choose, including: (A) Data Preprocessing including standardization, filtering, difference-normalization and spectral analysis. (B) Supervised neural methods. (C) Physics-based methods. (D) Systematic metrics evaluations.(E) Editable configurations for dataset and training set.