Home Monitoring of Blood Pressure: Short-Term Changes During Serial Measurements for 56398 Subjects

User engagement with home blood pressure monitoring: a multinational cohort using real-world data collected with a connected device

OBJECTIVE

Connected blood pressure (BP) monitors provide reliable data when used properly. Our objective was to analyse the engagement of real-world users with self-measurements.

METHODS

We included adult first-time users of a connected BP monitor from July 2019 to March 2021. They were categorized as persistent users if they continued to use the device between 311 and 400 days after inclusion. We defined a criterion to analyse the timing of self-measurements: at least 12 measurements performed within three consecutive days, at least once every 90 days. Persistent users were clustered by state sequence analysis according to the consistency of their BP monitor measurement timing with this criterion during 1 year of follow-up.

RESULTS

Among the 22 177 included users, 11 869 (54%) were persistent during the first year. Their use was consistent with the timing criterion 25% (median) of this time (first and third quartiles: 0%, 50%) and four patterns of use were identified by clustering: 5215 persistent users (44%) only performed occasional sparse measurements, 4054 (34%) complied at the start of follow-up up to eight cumulated months, 1113 (9%) complied at least once during later follow-up up to eight cumulated months, and the remaining 1487 (13%) complied nine or more cumulated months of follow-up.

CONCLUSION

Although connected BP monitors can collect a high volume of data, the real-life timing of self-measurements is far from recommended schedules. We must promote the use of BP monitors as recommended by guidelines and/or learn to analyse more occasional and sparse measurements.

Standardized home blood pressure monitoring: Rationale behind the 722 protocol

Home blood pressure (HBP) has been recognized as a prognostic predictor for cardiovascular events, and integrated into the diagnosis and management of hypertension. With increasing accessibility of oscillometric blood pressure devices, HBP monitoring is easy to perform, more likely to obtain reliable estimation of blood pressures, and feasible to document long-term blood pressure variations, compared to office and ambulatory blood pressures. To obtain reliable HBP estimates, a standardized HBP monitoring protocol is essential. A consensus regarding the optimal duration and frequency of HBP monitoring is yet to be established. Based on the current evidence, the "722" protocol, which stands for two measurements on one occasion, two occasions a day (morning and evening), and over a consecutive of 7 days, is most commonly used in clinical studies and recommended in relevant guidelines and consensus documents. HBP monitoring based on the "722" protocol fulfills the minimal requirement of blood pressure measurements to achieve agreement of blood pressure classifications defined by office blood pressures and to predict cardiovascular risks. In the Taiwan HBP consensus, the frequency of repeating the "722" protocol of HBP monitoring according to different scenarios of hypertension management, from every 2 weeks to 3 months, is recommended. It is reasonable to conclude that the "722" protocol for HBP monitoring is clinically justified and can serve as a basis for standardized HBP monitoring.

Adaptive Multimodal Fusion Framework for Activity Monitoring of People with Mobility Disability

The development of activity recognition based on multi-modal data makes it possible to reduce human intervention in the process of monitoring. This paper proposes an efficient and cost-effective multi-modal sensing framework for activity monitoring, it can automatically identify human activities based on multi-modal data, and provide help to patients with moderate disabilities. The multi-modal sensing framework for activity monitoring relies on parallel processing of videos and inertial data. A new supervised adaptive multi-modal fusion method (AMFM) is used to process multi-modal human activity data. Spatio-temporal graph convolution network with adaptive loss function (ALSTGCN) is proposed to extract skeleton sequence features, and long short-term memory fully convolutional network (LSTM-FCN) module with adaptive loss function is adapted to extract inertial data features. An adaptive learning method is proposed at the decision level to learn the contribution of the two modalities to the classification results. The effectiveness of the algorithm is demonstrated on two public multi-modal datasets (UTD-MHAD and C-MHAD) and a new multi-modal dataset H-MHAD collected from our laboratory. The results show that the performance of the AMFM approach on three datasets is better than the performance of the video or the inertial-based single-modality model. The class-balanced cross-entropy loss function further improves the model performance based on the H-MHAD dataset. The accuracy of action recognition is 91.18%, and the recall rate of falling activity is 100%. The results illustrate that using multiple heterogeneous sensors to realize automatic process monitoring is a feasible alternative to the manual response.

Remote smartphone monitoring of Parkinson's disease and individual response to therapy

Remote health assessments that gather real-world data (RWD) outside clinic settings require a clear understanding of appropriate methods for data collection, quality assessment, analysis and interpretation. Here we examine the performance and limitations of smartphones in collecting RWD in the remote mPower observational study of Parkinson's disease (PD). Within the first 6 months of study commencement, 960 participants had enrolled and performed at least five self-administered active PD symptom assessments (speeded tapping, gait/balance, phonation or memory). Task performance, especially speeded tapping, was predictive of self-reported PD status (area under the receiver operating characteristic curve (AUC) = 0.8) and correlated with in-clinic evaluation of disease severity (r = 0.71; P < 1.8 × 10^-6) when compared with motor Movement Disorder Society-Unified Parkinson's Disease Rating Scale (MDS-UPDRS). Although remote assessment requires careful consideration for accurate interpretation of RWD, our results support the use of smartphones and wearables in objective and personalized disease assessments.

Cuff-less Blood Pressure Estimation from Photoplethysmography via Visibility Graph and Transfer Learning

This paper presents a new solution that enables the use of transfer learning for cuff-less blood pressure (BP) monitoring via short duration of photoplethysmogram (PPG). The proposed method estimates BP with low computational budget by 1) creating images from segments of PPG via visibility graph (VG) that preserves the temporal information of the PPG waveform, 2) using pre-trained deep convolutional neural network (CNN) to extract feature vectors from VG images, and 3) solving for the weights and bias between the feature vectors and the reference BPs with ridge regression. Using the University of California Irvine (UCI) database consisting of 348 records, the proposed method achieves a best error performance of 0.008.46 mmHg for systolic blood pressure (SBP), and -0.045.36 mmHg for diastolic blood pressure (DBP), respectively, in terms of the mean error (ME) and the standard deviation (SD) of error, ranking grade B for SBP and grade A for DBP under the British Hypertension Society (BHS) protocol. Our novel data-driven method offers a computationally-efficient end-to-end solution for rapid and user-friendly cuff-less PPG-based BP estimation.

The Impact of Measurement Methods on Office Blood Pressure and Management of Hypertension in General Practice

INTRODUCTION

The use of unattended automated office blood pressure (uAutoOBP) versus attended automated (aAutoOBP) and manual auscultatory office blood pressure (AuscOBP) measurements is a topic of current controversy.

AIM

To evaluate the differences between OBP measurements methods in the general practice (GP) setting.

METHODS

We first compared aAutoOBP and uAutoOBP in 42 consecutive patients with hypertension (group 1). Secondly, we compared AuscOBP to uAutoOBP measurements in 133 consecutive patients with hypertension (group 2). In addition, we analyzed the achieved OBP targets as recommended in the 2018 European Society of Cardiology (ESC) and the European Society of Hypertension (ESH) guidelines in group 2.

RESULTS

The mean age of patients in group 1 was 71 years (range 34-89 years, 54.8% females). The aAutoOBP and uAutoOBP systolic (131.7 and 131.6 mmHg) and diastolic (83.4 and 82.4 mmHg) mean values were not significantly different. The patient characteristics in group 2 were similar to group 1. We observed a significant difference between AuscOBP and uAutoOBP measurement for both systolic (149.4 versus 129.5 mm Hg) and diastolic (85.4 versus 81.6 mm Hg, p < 0.0001, respectively). Accordingly, 20.3% and 45.9% of patients reached the overall 2018 ESC/ESH systolic and diastolic OBP targets of < 140/80 mmHg according to AuscOBP and uAutoOBP (p < 0.0001).

CONCLUSION

The attended versus unattended status of automated OBP measurements had no impact on OBP values in GP. However, significantly higher OBP values and lower rates of achieved target OBP were observed by using AuscOBP measurements by physicians in comparison to automated OBP recordings.