Multibeam Doppler Sensor-Based Non-Contact Heartbeat Detection Using Beam Diversity

Kitagawa, Tsukiko; Yamamoto, Kohei; Endo, Koji; Ohtsuki, Tomoaki

doi:10.1109/access.2022.3148426

Cited by 6 publications

(2 citation statements)

References 50 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Notably, the variation in R-R Interval (RRI) has been identified as a valuable parameter for assessing both cardiovascular and non-cardiovascular conditions in a subject (Aram & Setarehdan, 2013;W. Hu et al, 2014;Kitagawa et al, 2022;Shi et al, 2023;Xiang et al, 2023;Yamamoto & Ohtsuki, 2021;Ye & Ohtsuki, 2021;Zar & Tun, 2022). Traditional approaches for RRI estimation involve the use of established heartbeat monitoring systems such as Electrocardiogram (ECG) or Photoplethysmography (PPG) (Tun, 2021).…”

Section: Introductionmentioning

confidence: 99%

Innovative non-contact r-r intervals estimation using viterbi algorithm with Squared Branch Metric (VSBM)

Zar,

Tun,

Win

et al. 2024

JPTK

View full text Add to dashboard Cite

Non-contact heartbeat detection by using Doppler sensor is a critical component of remote health monitoring systems, enabling continuous and unobtrusive monitoring of an individual’s cardiovascular health. In this paper, we reported an innovative approach for non-contact heartbeat detection using the Viterbi algorithm, leveraging the distribution of the difference of two adjacent R-R intervals (RRIs). RRIs represent the time between successive peaks in the electrocardiogram (ECG) signal and are fundamental in analyzing heart rate variability, mental stress conditions and heart diseases. Numerous non-contact Doppler sensor-based methods have been proposed for heartbeat detection, leveraging the evaluation of RRIs without physical device attachment. However, challenges arise from unwanted peaks introduced by respiration and slight body movements, even when the subject remains motionless with normal breathing. This study presented an innovative approach for selecting heartbeat peaks utilizing the Viterbi algorithm with the squared difference of two adjacent RRIs as the Branch metric (BM). Our preliminary experiments reveal that the difference between two adjacent RRIs closely follows a Gaussian distribution. Building upon this observation, we considered the Viterbi algorithm with Squared Branch Metric (VSBM) to estimate the heartbeat accurately. To assess the accuracy of our peak selection method, we conducted experiments comparing it with two existing peak detection methods: (i) Doppler output after Low-Pass Filter (LPF)-based method and (ii) Spectrogram-based method. Our results demonstrate that the proposed VSBM method is effective to detect the heartbeat accurately for each peak detection method. Furthermore, we compared the performance of “Spectrogram + VSBM” outperforms the “Doppler output after LPF + VSBM” method by the Root-Mean-Square Error (RMSE) of RRIs.

show abstract

Section: Introductionmentioning

confidence: 99%

Innovative non-contact r-r intervals estimation using viterbi algorithm with Squared Branch Metric (VSBM)

Zar,

Tun,

Win

et al. 2024

JPTK

View full text Add to dashboard Cite

show abstract

“…The most common methods for monitoring heart rate are electrocardiography and photoplethysmography. These use contacttype sensors that can cause discomfort and restrict the wearer's activities [5] and are therefore unsuitable for long-term monitoring applications. In contrast, radar-based methods that measure body surface displacement caused by heart motion and pulse waves have attracted attention because they enable long-term heartbeat monitoring without sensor attachment [6]- [10].…”

Section: Introductionmentioning

confidence: 99%

Complex Number Assignment in the Topology Method for Heartbeat Interval Estimation Using Millimeter-Wave Radar

Tanaka,

Sumi,

Iwata

et al. 2024

IEEE Sens. Lett.

View full text Add to dashboard Cite

The topology method is an algorithm for accurate estimation of instantaneous heartbeat intervals using millimeter-wave radar signals. In this model, feature points are extracted from the skin displacement waveforms generated by heartbeats and a complex number is assigned to each feature point. However, these numbers have been assigned empirically and without solid justification. This study used a simplified model of displacement waveforms to predict the optimal choice of the complex number assignments to feature points corresponding to inflection points, and the validity of these numbers was confirmed using analysis of a publicly available dataset.

show abstract

An RNN-Bi LSTM Based Multi Decision GAN Approach for the Recognition of Cardiovascular Disease (CVD) From Heart Beat Sound: A Feature Optimization Process

Vinay,

Vidyasagar,

Rohith

et al. 2024

IEEE Access

View full text Add to dashboard Cite

The cardiovascular system is responsible for carrying the blood along with nutrition and oxygen throughout the body. This system consists of heart, blood, and blood vessels. The experts, or doctors called as cardiologists, analyze the sounds of heart's (lub-dub) beat and flow of blood to diagnose Cardio Vascular Disease (CVD) using a traditional stethoscope and phonological cardiogram technique. Through the stethoscope, the cardiologist will listen to vibration produced by heart beat and heart beat sound and murmur sound are popularly known as phonocardiogram (PCG) signals, which are being recorded for medical diagnosis purposes using a stethoscope. The development of a technique for the automatic recognition of HVD's assists the experts in recognizing the CVD effectively in the initial stage itself from PCG signals.There are many tools available to help doctors in a clinical setting for the accurate diagnose the CVD in a less time. The main aim of this proposed work is to provide an Artificial Intelligence (AI) based PCG signal analysis for the automatic and early detection of various cardiac conditions using supervised and unsupervised Recurrent Neural Network (RNN) based Bidirectional Long Short-Term Memory (Bi-LSTM) Machine Learning (ML) algorithm. Along with this algorithm, Generative Adversarial Networks (GAN's) is considered because they can create fresh, high-quality, pseudo-real data that resembles their training set which has been demonstrated by using their two unique networks: Discriminator Network (DN) and the Generator Network (GN). The proposed method is tested using heart sound signals from the well-known, freely accessible PhysioNet and Kaggle datasets. The Experimental results are validated based accuracy, precision, F1-score, sensitivity, and specificity.

show abstract

Multibeam Doppler Sensor-Based Non-Contact Heartbeat Detection Using Beam Diversity

Cited by 6 publications

References 50 publications

Innovative non-contact r-r intervals estimation using viterbi algorithm with Squared Branch Metric (VSBM)

Innovative non-contact r-r intervals estimation using viterbi algorithm with Squared Branch Metric (VSBM)

Complex Number Assignment in the Topology Method for Heartbeat Interval Estimation Using Millimeter-Wave Radar

An RNN-Bi LSTM Based Multi Decision GAN Approach for the Recognition of Cardiovascular Disease (CVD) From Heart Beat Sound: A Feature Optimization Process

Contact Info

Product

Resources

About