Optical sensors represent the most common type of biosensor. This two-part series provides a technical background on how optical technology can be used in bioanalytical applications. This paper summarizes how to use reflection method to pulse plethysmograph (PPG) waveform, and describes the physical and physiological principles in the work. The second part introduces the common noise and error sources affecting optical sensors in mobile and wearable applications, including the influence of confounding factors caused by the environment captured in the measurement and the physiological changes between user groups. It also outlines the current functions of wearable biosensors and the future direction of optical biosensor applications.
Overview: optical biosensors
Optical method is one of the most commonly used biosensors for plants and animals. For example, satellite remote sensing equipment usually uses reflection quality to determine the greenness and pressure of vegetation. The nurse clips the pulse oximeter to the patient’s fingertip, which is a routine step to take vital signs before seeing a doctor.
A wide range of applications reflects the versatility of light sensing, light is coherent or incoherent, and it will interact with the material it passes through, and be absorbed, reflected, scattered or changed in other ways. Scientists can check the size and shape of the light pulse, its spectral content and polarization to obtain information about the analyte in the medium through which the light pulse passes.
This article focuses on the second example we mentioned earlier, in which photoplethysmography (PPG) is used to monitor blood flow in real time.
Real time monitoring of blood flow by PPG
Plethysmograms are volume measurements. With the flow of blood, the cardiovascular pulse wave will be sent out from the heart and spread in the human body, so that the arteries and arterioles in the subcutaneous tissue expand periodically. PPG uses light to interrogate the organization. Because the blood in the tissue absorbs more light than the surrounding tissue, the reduction of blood volume will lead to an increase in the intensity of reflected or scattered light.
According to the relative position of light source and photodetector, PPG may have two configurations: transmission absorption and reflection. In the transmission configuration, the light source and the sensor are on the directly opposite side of the tissue. In the reflection arrangement, they can be on the same side. The reflection configuration takes advantage of the light scattering effect of human tissue, which will be discussed in the next section.
Although PPG is usually performed with finger clips in clinics and hospitals, it is possible to obtain effective PPG signals in many other body locations as long as vascular rich tissues are easily accessible, especially when using reflex configuration. For example, areas around the forehead, the external auditory canal, biceps or calf muscles, and even the wrist. Some of these alternative locations allow successful integration of PPG sensors into sports equipment and wearable devices, which will be explained in the second part of this series.
Optical measurement through tissue
Visible light (NIR) can penetrate several millimeters into human tissues. Penetration is limited because light is absorbed by blood, melanin, fat and water. In this range, light is mainly scattered and therefore diffuses rapidly. Light tissue interaction depends on different tissue components and the wavelength of light used. These changes make PPG signal can convey a lot of information.
The backscattered light is modulated by pulsating arterial blood volume, while the absorption from other tissue components remains constant. From a simplified point of view, these will cause the AC component in the photodetector output to be synchronized and proportional to the subject’s plethysmography signal, while the DC signal is a function of the constant absorption of the light source and tissue in the light path (Fig. 1).
Figure 1 The backscattered light received by PPG photodetector. This figure shows the output of a digital photodetector using four different light sources (red, green, blue and infrared) of the same intensity. The maximum PPG analog front end has processed the signal to eliminate the noise.
Different wavelengths of light are absorbed to different degrees by the human body. For example, many PPG devices contain one or more green LEDs. Green light is easily absorbed by our body, so focusing on green light can reduce the pollution of reflected light in the environment. However, due to its strong absorption capacity and limited penetration depth, it is only suitable for areas with massive blood perfusion.
Hemoglobin also strongly absorbs green light, making it difficult for it to penetrate deeper into the tissue. The medical application of PPG for pulse oxygen saturation is using NIR light source. Red light penetrates the human body and provides abundant physiological signal sources.
Common PPG signal artifacts
PPG signal usually contains not only backscattered light through the tissue, but also artifacts caused by ambient light and poor connection with the tissue. Other factors also affect the quality of PPG signal. These included skin structure, skin pigment, and even skin temperature. Advanced PPG integrated circuits, such as max30112, include special signal processing and sampling schemes to deal with some of the artifacts while saving power.
The most common problem faced by wearable device designers is environmental light pollution. Not only the ambient light is constantly changing, but the indoor light usually contains flicker, which tracks the frequency of the power line (i.e. 50 or 60 Hz, depending on the position), which will blur the information carried in the AC component of the PPG signal.
Motion artifacts may be caused by intermittent poor contact with tissues and photodetectors. In wearable devices that cannot restrict the movement of objects, motion artifacts may make it difficult to measure some slow physiological changes.
Another problem with PPG is that the detectable signal strength is affected by the concentration of melanin in the skin or skin pigment. By design, melanin reduces the wavelength of the incident light. It is found that there is no blood supply to the epidermis, so the target of PPG equipment is always under the epidermis. Designers can overcome the weak signal caused by melanin deposition by using stronger light source or by selecting different light frequency (such as NIR), where the absorption of melanin is low.
Once the artifacts are eliminated and the PPG waveform is amplified by signal processing, advanced algorithms can be used to extract and interpret its functions. Figure 2 The typical PPG waveform is shown.
The distance between two consecutive systolic peaks (called peak to peak interval) represents the complete cardiac cycle.
Figure 2 Typical PPG waveform
The range of contraction indicates the change of blood volume caused by arterial blood flow, and it is related to moderate air volume, which may indicate vasoconstriction or vasodilation, and also reflects cold sensitivity, response to anesthetics, blood loss, hypothermia, etc. Pulse width and pulse area are also related to systemic vascular resistance, which may reflect drug interaction or blood viscosity.
Dilation index, ratio between diastolic peak and systolic peak are indicators of aortic stiffness. The time delay between the systolic and diastolic peaks decreases with the age of the subjects, and can provide an index of arterial stiffness when the height of the subjects is constant. Both indicators can tell the subjects about their heart health.
We can also compare PPG signals from sources with more than one wavelength, for example, by comparing light with absorption through red and infrared LEDs in the same tissue to determine their oxygen saturation levels. Deoxyhemoglobin absorbs more red light than oxygenated hemoglobin, and oxygenated hemoglobin absorbs more infrared light (thus, the dark blue hue in venous blood compared to arterial blood). Therefore, blood oxygenation is proportional to the ratio of reflected red light to infrared light.
Successful and recommended measurements using PPG include:
- Heart rate and heart rate changes
- Respiratory rate
- Blood oxygen saturation
- Body rehydration
- Severe venous reflux disease (varicose veins)
- Venous function
- Cold sensitivity
- blood pressure
- Cardiac output
PPG was compared with other induction methods
PPG has been used to measure the breadth of physiological conditions. However, its most attractive quality is its ease of use and non-invasive. Due to the progress of special PPG circuit, the optical capacitance product method has the characteristics of portability, low power consumption and easy implementation.
Since both PPG and ECG are used to measure cardiac activity, they are often compared. Many clinical studies have examined the performance of PPG for ECG in different use situations for long-term heart rate variability (HRV) measurement. They often find that PPG provides a practical alternative to ECG. However, it should be noted that if the photodetector is unable to maintain good contact with the tissue (due to misinterpreted motion artifacts), the PPG may provide incorrect readings.
Part 2 of this series will discuss in detail the common mistakes that wearable PPG implementation may encounter, as well as their current and future functions.
Ian Chen is the executive director of the industrial and healthcare business unit of Maxim integrated.
Nakajima K, Tamura T, Miike h. heart rate and respiratory rate were monitored by photoplethysmography using digital filtering technique, Med eng physics 1994; 18（5）：365-72。
Moore D, Maher T, Kingston V, Shanik g, assessment of venous function by photoplethysmography, IR J Med SCI 1982; 151（1）：308-12。
Lees T, Lambert D, patterns of venous return in extremities due to skin changes associated with chronic venous insufficiency, British Journal of surgery. 1993； 80（6）：725-8。
Liu Jian, Yan Bo, Dai Wei, Ding Xin, Zhang y, Zhou Na, extraction of cutaneous arterial pulse by multi wavelength photoplethysmography, Biomedical Optics Express 7 (10): 4313-4326
Weinschenk s, Beise R, Lorenz J, heart rate variability (HRV) in deep breathing tests and 5-minute short-term recordings: agreement between otoplethysmography and ECG measurements in 343 subjects, European Journal of Applied Physiology (2016) 116:1527-1535