With the rapid development of technology, wearable devices have become an indispensable part of people’s daily lives. These small and lightweight devices not only provide a convenient way to obtain information but also show great potential in the field of health monitoring. This article mainly discusses the monitoring principles of wearable devices in aspects such as blood pressure detection, blood glucose detection, stress detection, and sleep quality detection, as well as related sensor technologies.
/01 Main Monitoring Signals in Health Monitoring Equipment/
Image source: Design of High-Accuracy and Low-Power Heart Rate Monitoring Module Based on PPG by Qiu Hui. Typical principle of reflective PPG sensor: Image source: https://support.coros.com/hc/en-us/articles/4406604073108-How-do-COROS-watches-measure-heart-rate.
2. Oxygen saturation monitoring. Many friends were first exposed to oxygen saturation monitoring during the epidemic stage. The principle of oxygen saturation monitoring can be simply summarized in two sentences: 1) Hemoglobin (Hb) absorbs light. The amount of light absorbed is proportional to the concentration of Hb in the blood. 2) Oxygenated hemoglobin and deoxygenated hemoglobin absorb different wavelengths of light differently. The realization of oxygen saturation monitoring can also be achieved through PPG sensors. During the heart rate monitoring process by PPG sensors, the oxygen saturation signal also pulsates synchronously with the beating of the heart. Therefore, by analyzing and extracting the blood pulse component (AC signal) through signal processing and ignoring all non-blood signals (DC signal), combined with external circuits and algorithm processing, oxygen saturation information can be obtained. Image source: Wearable heart rate and oxygen saturation monitoring system based on the Internet of Things. 3. Blood pressure monitoring. There are certain technical difficulties in implementing small consumer-grade wearable blood pressure measurement devices such as watches and rings. Compared with traditional cuff blood pressure measurement devices, the accuracy of measurement signals is a major performance challenge. The measurement principle of traditional cuff blood pressure measurement devices is all ‘oscillometry method’ (also known as oscillation method). During the deflation process of the cuff, the corresponding pressure signal is found through the amplitude relationship of the pulse wave, and the corresponding pressure value is the blood pressure value to be measured. Blood pressure monitoring based on watches can also be measured in a similar way to the oscillometry method. Pressure is measured by relying on the micro-airbag (cuff) on the wrist. The measurement mode and principle are the same as those of the upper arm cuff oscillometry method. However, due to the limitations of its product form itself, the measurement accuracy is lower than that of traditional arm-type measurement devices. Image source: Chinese Expert Consensus on the Application of Smart Wearable Devices in Blood Pressure Management of Young and Middle-Aged People 10.16439/j. issn.1673-7245.2022.08.005. There is another type of wearable blood pressure measurement device, which is a mobile device for cuffless blood pressure measurement. It mainly obtains blood pressure values through principles such as photoplethysmography, electrocardiogram estimation of blood pressure values, finger infrared sensing, ultrasound, etc. The measurement accuracy and reliability need to be verified.Image source: The measurement of blood pressure by standardized smart wearable devices is based on various principles of cuffless blood pressure measurement devices. One implementation plan combines electrocardiogram (ECG) and photoplethysmogram (PPG) sensors using the pulse wave transmission time (PTT) method. The time difference between the pulse wave transmitted from the heart to the PPG signal testing point is referred to as PTT in blood pressure measurement models.
Typically, the ECG is used as the starting point for PTT, and the PPG signal recorded at the fingertips is used as the endpoint. By identifying the main peaks of the ECG and PPG signals, the time interval between the two main peaks is obtained. A linear regression equation with correction coefficients is used to calculate systolic pressure. Since the model needs to be calibrated continuously with human activity, the performance in calculating diastolic pressure is relatively poor. Additionally, there is the pulse wave characteristic parameter method, which derives blood pressure characteristic parameter equations by establishing a relationship between pulse wave characteristic parameters and blood pressure to achieve blood pressure measurement. This method requires a high degree of completeness of PPG signals. Image source: https://conicear.best/product_details/66481975.html Glucose monitoring in the field of smart wearables has always been a focus for major manufacturers. Apple has made breakthrough progress in non-invasive glucose monitoring technology and will equip this feature on the Apple Watch. It is understood that the project, known as E5, aims to measure the glucose content in human blood without puncturing the skin to draw blood. Apple’s non-invasive glucose-related patents, image source: https://appleinsider.com/articles/18/08/23/apple-patent-suggests-work-on-non-invasive-glucose-monitoring-tech Insiders say that the glucose monitoring system will rely on silicon photonic chips and sensors designed by Apple. Apple’s developed silicon photonic chips determine the glucose concentration in the body by collecting the optical absorption spectrum returned after laser irradiation to the skin. Oxygen saturation detection can be divided into invasive, non-invasive, and minimally invasive measurements according to the type of puncture. Image source: ‘Optical Non-Invasive Measurement of Blood Glucose New Technology’. Traditional invasive blood glucose monitoring is based on the electrochemical principle of invasive blood glucose measurement, which is based on the principles of ammeters and voltmeters. In addition, measurements can be made based on the principle of optical sensors, where optical glucose sensors use a lectin substance and a fluorescent reagent to detect solutions of different concentrations. These types of detections all require finger pricking, resulting in poor patient user experience, and thus minimally invasive measurement technology has gradually emerged.Image source: ‘New Technology of Optical Non-Invasive Glucose Measurement’. Minimally invasive blood glucose measurement includes needle puncture type and laser blood collection type. The needle puncture type adopts the principle of subcutaneous sensor. The laser type uses laser to replace penetrating microneedles to achieve blood collection, reducing the occurrence of cross-infection caused by needle contact and having less pain.
Image source: https://innoacademy.engg.hku.hk/glucose-biosensor/ In recent years, non-invasive blood glucose measurement technology has also been continuously broken through. Non-invasive blood glucose can be divided into direct measurement and indirect measurement. The former is based on the molecular structure of glucose, and the latter monitors the impact of glucose on the second process such as pH. However, since indirect measurement is essentially inferring blood glucose content through body fluid glucose content in implementation, it cannot establish a good one-to-one correspondence, resulting in detection results sometimes not accurately reflecting pathological changes. Therefore, direct measurement is truly of pathological significance. In the book ‘New Technology of Optical Non-Invasive Glucose Measurement’, the author introduces the current mainstream and emerging non-invasive blood glucose measurements, including electrical impedance method, absorption spectroscopy method, polarimetry method, and Raman spectroscopy method. Image source: ‘New Technology of Optical Non-Invasive Glucose Measurement’. 5. Sleep quality detection. Image source: https://www.unityhealthnetwork.org/news/ultimate-sleep-tracker-guide-navigating-sleep-health-best-tools. Sleep quality detection is a process of multi-sensor and algorithm fusion. Wearable devices monitor the user’s sleep state through built-in acceleration sensors and photoelectric sensors. The acceleration sensor can obtain the user’s body movement characteristics and determine whether the user is in a sleep state. The photoelectric sensor analyzes sleep quality through heart rate and electrocardiogram signals. Different detectors perform filtering processing and algorithm analysis on the collected electrocardiogram signals and body movement signals respectively to obtain parameters such as sleep duration, body movement times, and heart rate changes. Then, through the analysis of the sleep staging algorithm, the user’s sleep quality report is obtained. 6. Stress detection. Stress detection is an important indicator for evaluating the user’s psychological state and life stress. Wearable devices usually use heart rate variability (HRV) and pressure sensors to achieve stress detection.HRV sensors assess stress levels by analyzing variations in a user’s heart rate, while pressure sensors directly measure changes in pressure on the skin’s surface. Studies have shown that wearable devices combining these two sensor technologies have high accuracy and sensitivity in stress detection. Monitoring psychological states, including stress, is highly significant, especially for individuals with mental disorders.
In fact, psychological stress presents differently based on various pressures and types, making it difficult for a single signal to conduct comprehensive monitoring, classification, and judgment. Therefore, a variety of sensor types are required. Although current products have relatively simple sensor components and monitoring patterns, the future trend will definitely integrate more sensors and information. A research team at the University of California, Los Angeles (UCLA), has developed a smartwatch capable of accurately, non-invasively, and in real-time evaluating cortisol levels in sweat. Cortisol, a steroid hormone often referred to as the ‘stress hormone,’ plays a key role in the body’s response to stress. When the body or mind is under stress, whether acute or chronic, the adrenal glands release cortisol to help the body cope with stress. Cortisol levels can provide information on stress and other biochemical indicators for the wearer. Cortisol is well-suited for measurement through sweat; by tracking cortisol in sweat, changes can be monitored in a wearable form, thus analyzing the user’s mental state. From the discussion above, it is evident that while many current wearable health monitoring products come in various forms (wristbands, watches, rings, headphones, etc.) and can monitor a variety of biological signals, the number and types of key sensors directly obtaining these signals are relatively limited. As a result, the accuracy and credibility of the detected signals are significantly lower than those of professional medical devices. Therefore, the information they provide can only be used for daily health monitoring and not for scenarios with strong medical purposes. Current products’ heart rate monitoring, blood oxygen detection, and blood pressure detection are all achieved with the help of PPG and ECG sensors. Skin temperature can be achieved with temperature sensors, and physical movement and exercise monitoring can be achieved through accelerometers. 1. PPG Sensor The PPG sensor, full name Photoplethysmogram sensor, is a type of biosensor that uses optical principles to monitor changes in blood volume. Integrated PPG sensors consist of multiple light-emitting diodes (LEDs) and a photoelectric detector (PD). The working principle is that the photoelectric detector measures changes in reflected light from the skin surface to form PPG signals.Under light source illumination, there are three situations for the blood volume in the skin: a certain amount of light is absorbed, a certain amount of light penetrates, and a certain amount of light is reflected. The intensity of reflected light and the blood volume at the collection site (such as the wrist or fingertip) will change with the change of heart rate. The measured signal can be divided into two types, namely the direct current component with unchanged absorbance from parts such as skin pigmentation, fat, muscle, and bone, and the alternating current component related to the change in blood volume generated by the heart.
The curve diagram of detecting the relationship between light intensity and time is called PPG signal. The time period of each pulse of PPG signal is affected by the heart rate, and the amplitude is affected by the concentration of different components of arterial blood. PPG is mainly used in the following signal detections: Heart rate monitoring: PPG sensors can monitor heart rate in real time, which is very helpful for sports training and health management. Heart rate variability (HRV) analysis: By measuring the change in consecutive heartbeat intervals, the activity of the autonomic nervous system can be evaluated. Oxygen saturation (SpO2) measurement: PPG sensors combined with red and infrared light can estimate the saturation of oxygen in the blood. Sleep monitoring: By monitoring the changes in heart rate and heart rate variability, combined with the results of body movement detection, sleep quality and sleep stages can be evaluated. Interestingly, combined with multispectral technology, multispectral PPG can also be designed. Multiwavelength photoplethysmography (MW-PPG) sensing technology is considered superior to single-wavelength photoplethysmography (SW-PPG) sensing technology. However, due to the availability limitations of sensing detectors, many previous studies can only use traditional bulky and expensive spectrometers as detectors, so MW-PPG technology cannot be applied to daily life. The team from National Taipei University of Technology has developed a chip-level multiwavelength photoplethysmography (MW-PPG) sensor using an innovative chip spectrometer, aiming at wearable applications. Combined with signal processing methods, this device can be used to stably extract PPG signals, and the signal-to-noise ratio (S/N) is increased by up to 50%, enabling the measurement of multiple parameters such as oxygen saturation and blood pressure. Picture source: MW-PPG Sensor: An on-Chip Spectrometer Approach. The process and material selection of PPG sensors have great diversity and can be realized without being based on traditional silicon-based processes, so they have advantages such as low cost and multiple design freedoms. Picture source for the main material system involved in PPG sensors: Systematic Review on Fabrication, Properties, and Applications of Advanced Materials in Wearable Photoplethysmography Sensors. The global photoplethysmography (PPG) biosensor market size was valued at 4 in 2022.The global market is valued at 16.8 billion dollars, with an expected compound annual growth rate of 11.6% from 2023 to 2030.
ECG sensors, or electrocardiogram sensors, are biosensors designed to monitor and record electrical activity of the heart. They detect electrical signals from the heart and convert these signals into graphical representations for analyzing cardiac health and function. The working principle of an ECG sensor is based on the electrophysiological properties of heart cells. Each heartbeat is caused by the electrical activity of myocardial cells. When these cells depolarize and repolarize, they generate tiny electrical signals. ECG sensors detect the potential difference produced by cardiac electrical activity through monitoring electrodes, filter and amplify the signals, and then analyze them using algorithms to derive physiological information such as heart rate. While both ECG and PPG can obtain cardiac activity signals, their principles differ; ECG relies primarily on detecting cardiac electrical signals, whereas PPG relies on photoelectric signals. ECG and PPG each have their strengths and weaknesses, with PPG excelling in terms of morphological and biometric diversity. ECG is faster than PPG in detecting bioelectrical signals and is less affected by environmental and surface differences, making it more accurate for heart rate monitoring and other precision tests. As sensor technology continues to advance and algorithms are continuously optimized, wearable devices will increasingly find broader applications in the field of health monitoring. From blood pressure detection and blood sugar monitoring to stress assessment and sleep quality analysis, wearable devices provide a comprehensive and multifaceted solution for health monitoring. It is evident that the technologies used in current products are still relatively limited, and the signals collected are also relatively singular, with much room for improvement in product form. From a technical standpoint, the key technologies involved in health monitoring devices include sensor technology, wireless communication technology, and data analysis algorithms. The sensor technology currently used in products is primarily traditional photoelectric technology such as PPG sensors. It is believed that the future trend of wearable sensors will be towards multi-signal, non-invasive, low-energy, high-precision, and flexible stretchable directions. In my opinion, the future development trend of health monitoring devices is mainly in the following areas: diversification of wearable forms. Currently, the main wearable monitoring devices are mostly watches, and in the future, these wearable devices will develop towards more detection positions and more structural forms, such as rings, patches, headphones, and even integrate with some accessories, such as those integrated into the frames of glasses that come into contact with the ears, hair clips for women, tattoo patches, and contact lenses.Flexible and stretchable next-generation wearable Photoplethysmography (PPG) systems are urgently needed to achieve high detection rates, rapid response times, and ultra-thin, flexible, stretchable device module designs, eliminating the need for cumbersome power supplies for self-powered operation, which is also beneficial for immediate detection applications.
Flexible stretchable PPG sensors are a new type of optoelectronic sensor with the following advantages: High comfort in wearing: Flexible materials allow it to conform to the skin, providing comfort and minimizing skin discomfort. Good stretchability: Adaptable to the movement and deformation of human skin, maintaining good measurement accuracy even during exercise or other physical activities. High measurement accuracy: Better adherence to skin and tissue allows for accurate measurement of physiological parameters such as heart rate and blood oxygen saturation. High integration: Directly integrable into various wearable devices, such as smart patches, headphones, rings, etc., facilitating user health monitoring anytime, anywhere. Broad potential applications: In addition to wearable devices, it can also be integrated into surgical instruments for auxiliary monitoring. Wearable body fluid sensors are an emerging medical monitoring device that can monitor and assess an individual’s health status by analyzing body fluids (such as sweat, saliva, tears, etc.). These devices are typically integrated into wearable devices, such as smartwatches, health trackers, smart clothing, etc., providing users with real-time health information. Common body fluids include sweat, interstitial fluid (ISF), saliva, etc. This technology can also be combined with point-of-care testing (POCT) and other developments to further advance wearable devices towards pathological diagnosis and other medical significance. Wearable sweat biosensors can analyze sweat components in real-time, providing insightful information about health status through the analysis of biomarkers in sweat. New sensor technologies are currently diversifying in the field of wearable sensors. When we combine electrochemical sensors, force sensors, body fluid sensors, optoelectronic sensors, etc., we can obtain more information. From a clinical perspective, wearable health monitoring devices can only provide reference information and cannot be used for diagnosis and medical purposes. This is because wearable sensors cannot reach some lesion points inside the body, cannot directly obtain direct information related to diseases, and the health indicators that can be detected on the body surface are easily disturbed. Therefore, ingestible biosensors, such as the illustrated biosensing capsules (IBC), can approach major organs through the gastrointestinal (GI) tract, monitor a wide range of biomarkers, serve as effective clinical diagnostic tools, and even provide targeted surgical and drug treatments.