Xiao-Rong Ding

Unobtrusive Sensing and Wearable Devices for Health Informatics

The aging population, prevalence of chronic diseases, and outbreaks of infectious diseases are some of the major challenges of our present-day society. To address these unmet healthcare needs, especially for the early prediction and treatment of major diseases, health informatics, which deals with the acquisition, transmission, processing, storage, retrieval, and use of health information, has emerged as an active area of interdisciplinary research. In particular, acquisition of health-related information by unobtrusive sensing and wearable technologies is considered as a cornerstone in health informatics. Sensors can be weaved or integrated into clothing, accessories, and the living environment, such that health information can be acquired seamlessly and pervasively in daily living. Sensors can even be designed as stick-on electronic tattoos or directly printed onto human skin to enable long-term health monitoring. This paper aims to provide an overview of four emerging unobtrusive and wearable technologies, which are essential to the realization of pervasive health information acquisition, including: 1) unobtrusive sensing methods, 2) smart textile technology, 3) flexible-stretchable-printable electronics, and 4) sensor fusion, and then to identify some future directions of research.

Continuous Cuffless Blood Pressure Estimation Using Pulse Transit Time and Photoplethysmogram Intensity Ratio

Pulse transit time (PTT) has attracted much interest for cuffless blood pressure (BP) measurement. However, its limited accuracy is one of the main problems preventing its widespread acceptance. Arterial BP oscillates mainly at high frequency (HF) because of respiratory activity, and at low frequency (LF) because of vasomotor tone. Prior studies suggested that PTT can track BP variation in HF range, but was inadequate to follow the LF variation, which is probably the main reason for its unsatisfactory accuracy. This paper presents a new indicator, the photoplethysmogram intensity ratio (PIR), which can be affected by changes in the arterial diameter, and, thus, trace the LF variation of BP. Spectral analysis of BP, PTT, PIR, and respiratory signal confirmed that PTT was related to BP in HF at the respiratory frequency, while PIR was associated with BP in LF range. We, therefore, develop a novel BP estimation algorithm by using both PTT and PIR. The proposed algorithm was validated on 27 healthy subjects with continuous Finapres BP as reference. The results showed that the mean ± standard deviation (SD) for the estimated systolic, diastolic, and mean BP with the proposed method against reference were -0.37 ± 5.21, -0.08 ± 4.06, -0.18 ± 4.13 mmHg, and mean absolute difference (MAD) were 4.09, 3.18, 3.18 mmHg, respectively. Furthermore, the proposed method outperformed the two most cited PTT algorithms for about 2 mmHg in SD and MAD. These results demonstrated that the proposed BP model using PIR and PTT can estimate continuous BP with improved accuracy.

Continuous Blood Pressure Measurement From Invasive to Unobtrusive: Celebration of 200th Birth Anniversary of Carl Ludwig

The year 2016 marks the 200th birth anniversary of Carl Friedrich Wilhelm Ludwig (1816-1895). As one of the most remarkable scientists, Ludwig invented the kymograph, which for the first time enabled the recording of continuous blood pressure (BP), opening the door to the modern study of physiology. Almost a century later, intraarterial BP monitoring through an arterial line has been used clinically. Subsequently, arterial tonometry and volume clamp method were developed and applied in continuous BP measurement in a noninvasive way. In the last two decades, additional efforts have been made to transform the method of unobtrusive continuous BP monitoring without the use of a cuff. This review summarizes the key milestones in continuous BP measurement; that is, kymograph, intraarterial BP monitoring, arterial tonometry, volume clamp method, and cuffless BP technologies. Our emphasis is on recent studies of unobtrusive BP measurements as well as on challenges and future directions.The year 2016 marks the 200th birth anniversary of Carl Friedrich Wilhelm Ludwig (1816-1895). As one of the most remarkable scientists, Ludwig invented the kymograph, which for the first time enabled the recording of continuous blood pressure (BP), opening the door to the modern study of physiology. Almost a century later, intraarterial BP monitoring through an arterial line has been used clinically. Subsequently, arterial tonometry and volume clamp method were developed and applied in continuous BP measurement in a noninvasive way. In the last two decades, additional efforts have been made to transform the method of unobtrusive continuous BP monitoring without the use of a cuff. This review summarizes the key milestones in continuous BP measurement; that is, kymograph, intraarterial BP monitoring, arterial tonometry, volume clamp method, and cuffless BP technologies. Our emphasis is on recent studies of unobtrusive BP measurements as well as on challenges and future directions.

Pulse Transit Time Based Continuous Cuffless Blood Pressure Estimation: A New Extension and A Comprehensive Evaluation

Cuffless technique enables continuous blood pressure (BP) measurement in an unobtrusive manner, and thus has the potential to revolutionize the conventional cuff-based approaches. This study extends the pulse transit time (PTT) based cuffless BP measurement method by introducing a new indicator – the photoplethysmogram (PPG) intensity ratio (PIR). The performance of the models with PTT and PIR was comprehensively evaluated in comparison with six models that are based on sole PTT. The validation conducted on 33 subjects with and without hypertension, at rest and under various maneuvers with induced BP changes, and over an extended calibration interval, respectively. The results showed that, comparing to the PTT models, the proposed methods achieved better accuracy on each subject group at rest state and over 24 hours calibration interval. Although the BP estimation errors under dynamic maneuvers and over extended calibration interval were significantly increased for all methods, the proposed methods still outperformed the compared methods in the latter situation. These findings suggest that additional BP-related indicator other than PTT has added value for improving the accuracy of cuffless BP measurement. This study also offers insights into future research in cuffless BP measurement for tracking dynamic BP changes and over extended periods of time.

Photoplethysmogram intensity ratio: A potential indicator for improving the accuracy of PTT-based cuffless blood pressure estimation.

The most commonly used method for cuffless blood pressure (BP) measurement is using pulse transit time (PTT), which is based on Moens-Korteweg (M-K) equation underlying the assumption that arterial geometries such as the arterial diameter keep unchanged. However, the arterial diameter is dynamic which varies over the cardiac cycle, and it is regulated through the contraction or relaxation of the vascular smooth muscle innervated primarily by the sympathetic nervous system. This may be one of the main reasons that impair the BP estimation accuracy. In this paper, we propose a novel indicator, the photoplethysmogram (PPG) intensity ratio (PIR), to evaluate the arterial diameter change. The deep breathing (DB) maneuver and Valsalva maneuver (VM) were performed on five healthy subjects for assessing parasympathetic and sympathetic nervous activities, respectively. Heart rate (HR), PTT, PIR and BP were measured from the simultaneously recorded electrocardiogram (ECG), PPG, and continuous BP. It was found that PIR increased significantly from inspiration to expiration during DB, whilst BP dipped correspondingly. Nevertheless, PIR changed positively with BP during VM. In addition, the spectral analysis revealed that the dominant frequency component of PIR, HR and SBP, shifted significantly from high frequency (HF) to low frequency (LF), but not obvious in that of PTT. These results demonstrated that PIR can be potentially used to evaluate the smooth muscle tone which modulates arterial BP in the LF range. The PTT-based BP measurement that take into account the PIR could therefore improve its estimation accuracy.

Impact of heart disease and calibration interval on accuracy of pulse transit time-based blood pressure estimation.

Continuous blood pressure (BP) measurement without a cuff is advantageous for the early detection and prevention of hypertension. The pulse transit time (PTT) method has proven to be promising for continuous cuffless BP measurement. However, the problem of accuracy is one of the most challenging aspects before the large-scale clinical application of this method. Since PTT-based BP estimation relies primarily on the relationship between PTT and BP under certain assumptions, estimation accuracy will be affected by cardiovascular disorders that impair this relationship and by the calibration frequency, which may violate these assumptions. This study sought to examine the impact of heart disease and the calibration interval on the accuracy of PTT-based BP estimation. The accuracy of a PTT-BP algorithm was investigated in 37 healthy subjects and 48 patients with heart disease at different calibration intervals, namely 15 min, 2 weeks, and 1 month after initial calibration. The results showed that the overall accuracy of systolic BP estimation was significantly lower in subjects with heart disease than in healthy subjects, but diastolic BP estimation was more accurate in patients than in healthy subjects. The accuracy of systolic and diastolic BP estimation becomes less reliable with longer calibration intervals. These findings demonstrate that both heart disease and the calibration interval can influence the accuracy of PTT-based BP estimation and should be taken into consideration to improve estimation accuracy.

Effects of cuff inflation and deflation on pulse transit time measured from ECG and multi-wavelength PPG.

Pulse transit time (PTT), which refers to the time it takes a pulse wave to travel between two arterial sites is a promising index for cuff-less blood pressure (BP) estimation, as well as non-invasive assessment of arterial functions. However, it has not been investigated whether PTTs measured from ECG and different wavelength PPG are equally affected by the arterial status. Furthermore, comparison between the changes of different PTTs can provide enlightenment on the hardware implementation of the PTT-based BP estimation method. This work mainly studied the changes of PTTs calculated from electrocardiogram (ECG) and multi-wavelength photoplethysmogram (PPG) after exerting cuff pressure on the upper arm. A four-channel PPG acquisition system was developed to collect the multi-wavelength PPG signals of red, yellow, green and blue light at the fingertip simultaneously. Ten subjects participated in the experiment and their PTTs measured from different PPG and ECG signals before and after exerting cuff pressure were compared. This study found that within one minute after the four-minute cuff inflation and deflation process, the PTT measured from ECG and yellow PPG experienced a significant increase (p<;0.05) while the PTT from ECG and blue PPG had no statistical difference (p>0.9) compared with that before exerting cuff pressure. This indicates that PTTs calculated from different wavelength PPG have different recoverability from smooth muscle relaxation. Another interesting finding is that the PTT calculated from ECG and yellow PPG had a strong correlation (|r|>0.7) with the time difference between yellow PPG and other PPG signals, which implies the potential of the time difference between yellow PPG and other PPGs as a complementary to PTT-based model for blood pressure estimation.

A preliminary study on multi-wavelength PPG based pulse transit time detection for cuffless blood pressure measurement

Pulse transit time (PTT) has been widely studied as an index of blood pressure (BP) changes. In recent years, some prototypes of PTT-based wearable BP measurement devices have been developed, which can relieve users from the discomfort caused by the inflating cuff used in auscultatory and oscillometric BP measurement techniques. However, in the common practice for PTT detection, multi-site sensor implementation on human body is required, making it difficult for the integration of wearable devices. Since multi-wavelength (MW) photoplethysmogram (PPG) signals carry blood pulsation information of different blood vessels embedded in different skin depths, the time difference between different wavelength PPG signals collected at the same body site can be treated as a special PTT on a short length of blood vessels beneath the skin. In this work, the time difference between MW PPG, denoted as PTT_MW, was explored to track BP changes as a substitute of infrared (IR) PTT_EP. (PTT_EP is the time interval between electrocardiogram (ECG) and IR PPG.) Ten healthy adult subjects participated in the experiment, and their continuous BP, ECG and fingertip MW PPG signals generated from blue, green, yellow and IR light were recorded after 2-minute static handgrip exercise at 33% maximal voluntary contraction. The results showed that the correlation between Systolic BP (SBP) and IR-Blue PTT_MW (|r|= 0.52) was comparable to the correlation between SBP and IR PTT_EP (|r|= 0.59). Moreover, we optimized the wavelength combination of PTT_MWs for each subject and found the average value of optimal correlation between SBP and PTT_MW reached 0.76, which was significantly (p<;0.01) higher than the correlation between IR PTT_EP and SBP. This study reveals that the time difference between MW PPG can be potentially used as PTT for cuffless BP measurement with its unique advantage in simple sensor implementation at only one body site.Pulse transit time (PTT) has been widely studied as an index of blood pressure (BP) changes. In recent years, some prototypes of PTT-based wearable BP measurement devices have been developed, which can relieve users from the discomfort caused by the inflating cuff used in auscultatory and oscillometric BP measurement techniques. However, in the common practice for PTT detection, multi-site sensor implementation on human body is required, making it difficult for the integration of wearable devices. Since multi-wavelength (MW) photoplethysmogram (PPG) signals carry blood pulsation information of different blood vessels embedded in different skin depths, the time difference between different wavelength PPG signals collected at the same body site can be treated as a special PTT on a short length of blood vessels beneath the skin. In this work, the time difference between MW PPG, denoted as PTT_MW, was explored to track BP changes as a substitute of infrared (IR) PTT_EP. (PTT_EP is the time interval between electrocardiogram (ECG) and IR PPG.) Ten healthy adult subjects participated in the experiment, and their continuous BP, ECG and fingertip MW PPG signals generated from blue, green, yellow and IR light were recorded after 2-minute static handgrip exercise at 33% maximal voluntary contraction. The results showed that the correlation between Systolic BP (SBP) and IR-Blue PTT_MW (|r|= 0.52) was comparable to the correlation between SBP and IR PTT_EP (|r|= 0.59). Moreover, we optimized the wavelength combination of PTT_MWs for each subject and found the average value of optimal correlation between SBP and PTT_MW reached 0.76, which was significantly (p<;0.01) higher than the correlation between IR PTT_EP and SBP. This study reveals that the time difference between MW PPG can be potentially used as PTT for cuffless BP measurement with its unique advantage in simple sensor implementation at only one body site.

Multi-wavelength photoplethysmography method for skin arterial pulse extraction

In this work, we present a multi-wavelength (MW) PPG method exploiting the wavelength dependence of light penetration in skin tissue to provide depth resolution of skin blood pulsation. The MW PPG system requires two to three light sources in different wavelengths and extracts the arterial blood pulsation through a multi-wavelength multi-layer light-skin interaction model, which removes the capillary pulsation (determined from the short-wavelength PPG signal) from the long-wavelength PPG signal using absorption weighting factors that are quasi-analytically calibrated. The extracted pulsations are used to calculate blood pressure (BP) through pulse transit time (PTT), and the results are compared with those obtained from the single wavelength PPG method. The comparative study is clinically performed on 20 subjects including 10 patients diagnosed with cardiovascular diseases and 10 healthy subjects. The result demonstrates that the MW PPG method significantly improves the measurement accuracy of systolic BP (SBP), reducing the mean absolute difference between the reference and the estimated SBP values from 5.7 mmHg (for single-wavelength PPG) to 2.9 mmHg (for three-wavelength PPG).

A new modeling methodology for continuous cuffless blood pressure measurement

Continuous blood pressure (BP) monitoring in a noninvasive and unobtrusive way can significantly improve the awareness, control and treatment rate of prevalent hypertension. Pulse transit time (PTT) has become increasingly popular in recent years for continuous BP measurement without a cuff. However, the accuracy issue of PTT-based method remains to be solved for clinical application. Some previous studies have attempted to estimate BP with only PTT by using linear regression, which is susceptible to arterial regulation and may not reflect the actual relationship between PTT and BP. Furthermore, PTT does not contain all the information of BP variation, thereby resulting in unsatisfactory accuracy. In this paper we establish a cuffless BP estimation model from a physiological perspective by utilizing PTT and photoplethysmogram (PPG) intensity ratio (PIR), an indicator we have recently proposed for evaluation of the change in arterial diameter and the low frequency variation of BP, with the consideration that PIR can track changes in mean BP (MBP) and arterial diameter change. The performance of the proposed BP model was evaluated by comparing the estimated BP with Finapres BP as reference on 10 healthy subjects. The results showed that the mean ± standard deviation (SD) of the estimation error for systolic and diastolic BP were -0.41 ± 5.15 and -0.84 ± 4.05 mmHg, and mean absolute difference (MAD) were 4.18 and 3.43 mmHg, respectively. Furthermore, the proposed modeling method was superior to one contrast PTT-based method, demonstrating the proposed model would be promising for reliable continuous cuffless BP measurement.