Heart rate measurement based on a time-lapse image
Introduction
Various technologies using non-contact, non-invasiveness, and ambulatory methods have been studied and developed for biomedical measurement. X-ray computed tomography and magnetic resonance imaging are typical examples of the non-contact method for collecting anatomical information. Research studies using non-contact measurement have continued for some years, such as measurement of brain activity by near-infrared radiation [1] and detection and evaluation of a wound and ischemic tissue by non-contact electric impedance measurement [2]. Chen et al. also measured the breathing and heartbeat signals of human subjects buried under earthquake rubble by using a microwave beam [3]. Optical methods are also widely used for non-contact measurement. For example, Sinko et al. evaluated the spine shape by using Moiré topography [4]. Matsui et al. monitored heart rate and respiration rate using laser irradiation by fast Fourier transform [5], [6], and Augousti et al. measured respiratory and cardiac function by using optical fibers [7].
Because measurement of circulatory or musculoskeletal parameters can be principally invasive, attempts have been made to develop bloodless, painless, and less invasive methods by using ultrasound techniques such as diagnosis of fetal anemia [8] or arterial occlusive disease [9] by the Doppler ultrasound method, measurement of large sudden pressure drops seen in hypovolemia based on pulse wave transit time [10], measurement of bone deformation by echo tracking [11], and tendon force measurement by analyzing the transmission of ultrasonic waves [12]. In addition to these studies using ultrasound techniques, electrocardiography and photoplethysmography can be used for the non-invasive evaluation of blood pressure [13] and the detection of vascular disease [14], respectively.
For measurement of normal activities, various ambulatory methods have been developed by minimizing the monitoring devices. Typical examples are the Holter monitor and the pedometer [15]. de Lusignan et al. developed a wireless data collection system by using small sensors placed on the chest and recorded heart and breath rate, blood pressure, the electrocardiogram, and body temperature [16]. Gramse et al. mounted dry ECG electrodes as well as capacity-based elastic strain gauges on infant pajamas and monitored cardiopulmonary parameters [17]. Chen et al. also installed an air-free water-filled tube sensor under the pillow and monitored heart and breath rate during sleep by measuring the pressure under the near- and far-neck occiput [18]. Regarding the measurement of gait parameters, Veltink et al. proposed the use of two six-degrees-of-freedom force and moment sensors mounted in the shoe sole [19]. Since worldwide ageing of societies has been rapidly going on for some years, the ambulatory methods above could be useful for the health care of the elderly.
In recent years, there has been a trend in the development of non-contact instrumentation by combining a handy imager, or the CCD, and highly advanced PC technologies. A typical example is kinematics measurement by using the VICON system [20]. In addition, motion analysis of neonates in incubators by motion tracking [21], analysis of visual-gestural language using video cameras [22], and automatic recognition of facial expression [23] have been conducted. Furthermore, trials have been carried out to extract physiological parameters from the obtained kinematics data. Nakajima et al. [24] recorded the real-time image sequence of a subject in bed and measured the respiratory rate based on postural change. Cala et al. estimated lung volume using chest wall motion data acquired by reflective markers places on the body surface [25]. We also applied these image technologies and developed a new non-contact measuring method, which could collect heart and respiratory rate simultaneously.
Section snippets
System configuration
Analogue images from a CCD camera (Network Handycam IP, Sony Corp., Japan) were acquired through a picture board (PX500, Image Nation, USA) and stored in a PC (GP7-500, Gateway Corp., Windows NT). Using the timer function of Visual Basic (VB), the time-lapse image of a part of the subject's skin was consecutively captured, and the changes in the average image brightness of the region of interest (ROI) were measured using image-processing software (X Caliper, Optimas Corp., USA), which is
Results
Fig. 2 shows a typical example of the heart rate measurement by time-lapse imaging. Fig. 2(a) shows the changes in the average image brightness in the ROI regulated with spline interpolation. Two kinds of waves, which are considered to be caused by respiration and heart beat, were observed. Fig. 2(b)–(d) shows the results of the process of the first-order differentiation, low pass filter, and AR spectral analysis, respectively. As shown in Fig. 2(d), two peaks could be clearly detected at
Discussion
In this study, we developed a new device by combining a time-lapse image from a handy video camera and image processing on a PC, and found that it could measure the 30-s average heart and respiratory rates based on the changes in the brightness of the ROI set around the cheek of the unrestricted subject. Regarding the mechanism of the respiratory measurement, the cheek skin moved due to the respiratory motion of the upper thorax, which varied the brightness of the ROI. The same mechanism might
Conclusion
In this study, we found that the simultaneous measurement of both respiratory and pulse rate was possible by acquiring the time-lapse image of a subject's face for 30 s and examining the image brightness in the ROI by AR spectral analysis. Through calibration experiments with a thermistor and a pulse oximeter, correlation coefficients of 0.93 and 0.90 were obtained for respiratory and pulse rate measurements, respectively. With the attractive features of ambulatory and non-contact monitoring,
Acknowledgements
This work was supported by a grant from Konica Minolta Imaging Science Foundation. The author (CT) is deeply grateful to Yushi ODASHIMA, Emeritus Professor of Tokyo University, for review of the manuscript and a lot of encouragement.
References (35)
- et al.
Dysgnathia, orthognathic surgery and spinal posture
Int J Oral Maxillofac Surg
(2006) - et al.
A comparison of the wavelet and short-time Fourier transforms for Doppler spectral analysis
Med Eng Phys
(2003) - et al.
A new method for measurement of bone deformation by echo tracking
Med Eng Phys
(2006) - et al.
A non-invasive method of tendon force measurement
J Biomech
(2005) - et al.
A model-based image-matching technique for three-dimensional reconstruction of human motion from uncalibrated video sequences
J Biomech
(2005) - et al.
Fully automated health monitoring system in the home
Med Eng Phys
(1998) - et al.
Multi-sensors acquisition, data fusion, knowledge mining and alarm triggering in health smart homes for elderly people
C R Biol
(2002) - et al.
Measurement of brain activity by near-infrared light
J Biomed Opt
(2005) - et al.
Electrical conductivity imaging via contactless measurements: an experimental study
IEEE Trans Med Imaging
(2003) - et al.
Microwave life-detection systems for searching human subjects under earthquake rubble or behind barrier
IEEE Trans Biomed Eng
(2000)
The non-contact monitoring of heart and respiratory rates using laser irradiation: an experimental simultaneous monitoring with and without clothes during biochemical hazards
J Med Eng Technol
A novel method to prevent secondary exposure of medical and rescue personnel to toxic materials under biochemical hazard conditions using microwave radar and infrared thermography
IEEE Trans Biomed Eng
Improved fibre optic respiratory monitoring using a figure-of-eight coil
Physiol Meas
Evaluating suspected fetal anemia with Doppler ultrasound
J Reprod Med
Noninvasive investigation of blood pressure changes using the pulse wave transit time: a novel approach in the monitoring of hemodialysis patients
J Artif Organs
Determining the relationship of heart rate and blood pressure using voluntary cardio-respiratory synchronization (VCRS)
Physiol Meas
Age-related changes in the characteristics of the photoplethysmographic pulse shape at various body sites
Physiol Meas
Cited by (320)
Hand-over-face occlusion and distance adaptive heart rate detection based on imaging photoplethysmography and pixel distance in online learning
2023, Biomedical Signal Processing and ControlMeasurements of pulse rate using facial video cameras from smart devices in patients diagnosed with atrial fibrillation
2023, Journal of ElectrocardiologyAssessment of facial video-based detection of atrial fibrillation across human complexion
2022, Cardiovascular Digital Health JournalSmart mattress integrated with pressure sensor and IoT functions for sleep apnea detection
2022, Measurement: SensorsCitation Excerpt :The power supply can be given through the cable. RTC and LCD pins are connected to the IoT [17,18]. Real-Time Clock (RTC) board enables to use of a real-time clock and calendar in applications.