Page 89 - 2024F

82 UEC Int’l Mini-Conference No.53

The UEC International mini-Conference

Non-Contact Heart Rate Monitoring of Anesthetized Laboratory Animals Using Doppler Radar
Technology
Ta Hoai Nam, Nguyen Huu Son
The University of Electro-Communications, Japan; tahoainam@gl.cc.uec.ac.jp.
I. INTRODUCTION
DOPPLER radar was first used for contactless human vital signs detection in 1975.
Monitoring vital signs in small laboratory animals like Cat is crucial for physiological
studies, therapy development, and disease understanding. Traditional invasive
methods, such as telemetry sensor implants, provide accurate data but cause
discomfort and limit mobility. Non-invasive contact methods like ECG and PPG
avoid these issues but still require. attaching sensors, which can distress the
animals.
Recent advancements in non-contact techniques, including CW-Doppler, FMCW,
and UWB radar, allow for accurate monitoring of vital signs without disturbing the
animals’ natural behavior. These innovations offer a more humane and effective
approach to vital sign monitoring in small animals. Radar can detect vital signs
thanks to the modulation effect caused by a cat’s chest movements associated with
respiration and heart muscle contractions.
II. DESCRIPTION

In practice, the “noise” respiration signal may mask the heartbeat signal, i.e., the
heartbeat frequency component is not strong enough in amplitude to be recognized. Fig. 1: Laboratory cat vital sign monitoring experiment
In this situation, the heartbeat cannot always be obtained accurately. This is and anesthesia concentration procedure.
a typical scenario of trying to receive a weak signal, situated near a higher
magnitude interfering signal in the frequency domain. The proposed dual-band
system with the analog respiration cancellation method is useful to increase
heartbeat detection accuracy. The concept of this method is shown in Fig. 2. In this
illustration, both channels work well with respiration detection. However, for
heartbeat detection, the one with higher frequency works better than the lower
frequency one. The signal from the lower frequency channel can be regarded as
a reference for respiration cancellation. This takes place after aligning both the
amplitude and the phase of the two received and demodulated signals from two
channels of the radar system. Then, the respiration components in two frequency
channels become the same, while the heartbeat components are different, due to
different heartbeat detection abilities at two frequency channels.
1. Research’s target
Propose a non-contact, time-domain signal processing algorithm for heart rate (HR) Fig 2. Block diagram of the dual-band radar system.
monitoring in laboratory cats using continuous wave (CW) Doppler radar. The study
focuses on two main objectives:
• Employing a non-contact method with two 24GHz and 60GHz CW-Doppler
Radars .
• Providing high-accuracy in average Heartrate estimation.
2. Experiment Setup
• The experiment was conducted on anesthetized cats, using radar to record their
vital signs, resulting in two output channel: I-channel & Q-channel, showing in
fig. 1 and fig.2.
• ECG sensor: provide Heartbeat reference to compare
• All experiments were conducted under the guidelines established by Fig 3. Pre-processing Stage.
Physiological Society of Japan and were approved by the University of
Electro-Communications Institutional Animal Care (#A44)
3. Method
The fig. 4 demonstrate proposed method consists of processes:
• Pre-processing Stage: Utilizing Fast Fourier Transform (FFT) and peak-finding
algorithms to calculate the temporary heart rate.
• Respiratory Component Removal and Heart Rate Extraction: The experiment
al data collected by the oscilloscope were imported to MATLAB for processing,
and the signal processing flowchart to obtain the heartbeat by canceling the
respiration is shown in Fig. 8. From the algorithm diagram, it is shown that the
processing consists of two FFT operations, one estimation, and normalizations.
• Heartbeat Detection: Implementing a second-stage bandpass filter and a
zero-crossing algorithm to detect HR.

Fig 4. Respiration cancellation technology in the frequency domain
Flowchart and Algorithm diagram.
References
•[1] N. H. Son, H. T. Yen, G. Sun, and K. Ishibashi, “High-Accuracy Heart Rate Estimation By Half/Double BBI Moving Average and Data Recovery Algorithm
of 24GHz CW-Doppler Radar,” in 2022 International Conference on Advanced Technologies for Communications (ATC), Ha Noi, Vietnam: IEEE, Oct. 2022, p
p. 360–363. doi: 10.1109/ATC55345.2022.9943010.

ASP-DAC 2025