Zunyi Tang

Automatic sleep monitoring system for home healthcare

In this paper, we present an automatic and real-time sleep monitoring system for home healthcare. The basic system consists of sensor boards, bedside boxes and a group of servers with various functions for analyzing sleep data and reporting sleep qualities of users. The sensor board is installed beneath a pillow to measure pressure change signal during sleep. The bedside box digitizes and transmits the signal to the servers. The system is different from conventional sleep monitoring devices used in a hospital, whose sensor is non-intrusive and easy-to-use that any person can use it at home. Furthermore, it provides a web-based interactive platform for users to analyze, manage and visualize their sleep data. These functions are evaluated by dozens of persons at different age groups over three months, and they are confirmed useful not only for a doctor who is responsible for checking his/her patients daily sleep pattern, but also useful for self-care. It is of great application potential in home healthcare.

Dictionary Learning Based on Nonnegative Matrix Factorization Using Parallel Coordinate Descent

Sparse representation of signals via an overcomplete dictionary has recently received much attention as it has produced promising results in various applications. Since the nonnegativities of the signals and the dictionary are required in some applications, for example, multispectral data analysis, the conventional dictionary learning methods imposed simply with nonnegativity may become inapplicable. In this paper, we propose a novel method for learning a nonnegative, overcomplete dictionary for such a case. This is accomplished by posing the sparse representation of nonnegative signals as a problem of nonnegative matrix factorization (NMF) with a sparsity constraint. By employing the coordinate descent strategy for optimization and extending it to multivariable case for processing in parallel, we develop a so-called parallel coordinate descent dictionary learning (PCDDL) algorithm, which is structured by iteratively solving the two optimal problems, the learning process of the dictionary and the estimating process of the coefficients for constructing the signals. Numerical experiments demonstrate that the proposed algorithm performs better than the conventional nonnegative K-SVD (NN-KSVD) algorithm and several other algorithms for comparison. What is more, its computational consumption is remarkably lower than that of the compared algorithms.

Structural Optimization of a Wearable Deep Body Thermometer: From Theoretical Simulation to Experimental Verification

Deep body temperature (DBT) has yet to be measured continuously in everyday life, even though it is useful in physiological monitoring and chronobiology studies. We tried to address this issue by developing a transcutaneous thermometer based on the dual-heat-flux method (DHFM) invoking the principle of heat transfer, for which measurement error was mitigated by elaborate design. First, a structural modification based on the original design of the DHFM was implemented by the finite element method. Based on the results of the simulations, prototypes were then implemented and tested with an experimental system that mimicked the thermometer being applied to skin. The simulation phase proposed the adoption of an aluminum cover to boost measurement accuracy and suggested that thermometers of different height be chosen according to specified requirements. The results of the mock-up experiments support the modification put forward in the simulation phase: the standard type (15 mm in height) achieved the accuracy with error below 0.3°C while the thin type (9 mm in height) attained accuracy with error less than 0.5°C under normal ambient temperature ranging from 20 to 30°C. Even though the design should also be examined in vivo, it is believed that this study is an important step in developing a practical noninvasive deep body thermometer.

Preliminary study of a new home healthcare monitoring to prevent the recurrence of stroke

We tested a new model that involves monitoring physiological parameters and the use of a rehabilitation training system to improve team-based healthcare. The system consisted of unobtrusive monitoring, a highly efficient database, and interventions by health professionals. This report discusses the core technologies, especially those involved in monitoring, which have been developed. The unobtrusive blood pressure (BP) estimation system, electrocardiogram (ECG), Kinect-based rehabilitation training system, and web-based care system, including its database, were designed and tested separately. BP was estimated with a cuffless BP monitor Estimated BP values were sufficiently accurate compared with noninvasive beat-by-beat BP values. The ECG was monitored with textile electrodes mounted on pajamas, and this allowed accurate calculation of heart rates. The training system was tested with four hemiplegic patients and garnered a high acceptance rate. The core technologies operated well, and the proposed system may prove useful and effective for home-based healthcare. Further studies are needed to evaluate the total care system.

A Chair–Based Unobtrusive Cuffless Blood Pressure Monitoring System Based on Pulse Arrival Time

In this paper, we present an unobtrusive cuffless blood pressure (BP) monitoring system based on pulse arrival time (PAT) for facilitating long-term home BP monitoring. The proposed system consists of an electrocardiograph (ECG), a photoplethysmograph (PPG), and a control circuit with a Bluetooth module, all of which are mounted on a common armchair to measure ECG and PPG signals from users while sitting on the armchair in order to calculate continuous PAT. Considering the good linear correlation of systolic BP (SBP) and the nonlinear correlation of diastolic BP (DBP) with PAT, a new BP estimation method was proposed. Ten subjects underwent BP monitoring experiments involving stationary sitting on a chair, lying on a bed, and pedaling using an ergometer in order to assess the accuracy of the estimated BP. A cuff-type BP monitor was used as reference in the experiments. Results showed that the mean difference of the estimated SBP and DBP was within 0.2 ± 5.8 mmHg (

Dictionary learning by nonnegative matrix factorization with 1/2-norm sparsity constraint

p

Improving dictionary learning using the Itakura-Saito divergence

< 0.00001) and 0.4 ± 5.7 mmHg (

Preliminary study of unobtrusive monitoring to increase safety in daily living

p

A chair for cuffless real-time estimation of systolic blood pressure based on pulse transit time.

< 0.00001), respectively, and the mean absolute difference of the estimated SBP and DBP were 4.4 and 4.6 mmHg, respectively, compared to references. Additionally, five subjects participated in data collections consisting of sitting on a chair twice a day for one month. Compared to the reference, the difference did not obviously increase along with time, even though individualized calibration was executed only once at the beginning. These results suggest that the proposed system has quite the potential for long-term home BP monitoring.

An Invariant Pattern Recognition System Using the Bayesian Inference on Hierarchical Sequences with Pre-processing

In this paper, we propose an overcomplete, nonnegative dictionary learning method for sparse representation of signals, which is based on the nonnegative matrix factorization (NMF) with 1/2-norm as the sparsity constraint. By introducing the 1/2-norm as the sparsity constraint into NMF, we show that the problem can be cast as sequential optimization problems of quadratic functions and quartic functions. The optimization problem of each quadratic function can be solved easily since the problem has closed-form unique solution. The optimization problem of quartic function can also be formulated as solving a cubic equation, which can be efficiently solved by the Cardano formula and selecting one of solutions with a rule. To implement this nonnegative dictionary learning, we develop an algorithm by employing coordinate-wise decent strategy, i.e., coordinate-wise decent based nonnegative dictionary learning (CDNDL). Numerical experiments show that the proposed algorithm performs better than the nonnegative K-SVD (NN-KSVD) and the other two compared algorithms.