Stephen J. Redmond

Barometric Pressure and Triaxial Accelerometry-Based Falls Event Detection

Falls and fall related injuries are a significant cause of morbidity, disability, and health care utilization, particularly among the age group of 65 years and over. The ability to detect falls events in an unsupervised manner would lead to improved prognoses for falls victims. Several wearable accelerometry and gyroscope-based falls detection devices have been described in the literature; however, they all suffer from unacceptable false positive rates. This paper investigates the augmentation of such systems with a barometric pressure sensor, as a surrogate measure of altitude, to assist in discriminating real fall events from normal activities of daily living. The acceleration and air pressure data are recorded using a wearable device attached to the subjects waist and analyzed offline. The study incorporates several protocols including simulated falls onto a mattress and simulated activities of daily living, in a cohort of 20 young healthy volunteers (12 male and 8 female; age: 23.7 ±3.0 years). A heuristically trained decision tree classifier is used to label suspected falls. The proposed system demonstrated considerable improvements in comparison to an existing accelerometry-based technique; showing an accuracy, sensitivity and specificity of 96.9%, 97.5%, and 96.5%, respectively, in the indoor environment, with no false positives generated during extended testing during activities of daily living. This is compared to 85.3%, 75%, and 91.5% for the same measures, respectively, when using accelerometry alone. The increased specificity of this system may enhance the usage of falls detectors among the elderly population.

Sensors-Based Wearable Systems for Monitoring of Human Movement and Falls

The rapid aging of the worlds population, along with an increase in the prevalence of chronic illnesses and obesity, requires adaption and modification of current healthcare models. One such approach involves telehealth applications, many of which are based on sensor technologies for unobtrusive monitoring. Recent technological advances, in particular, involving microelectromechnical systems, have resulted in miniaturized wearable devices that can be used for a range of applications. One of the leading areas for utilization of body-fixed sensors is the monitoring of human movement. An overview of common ambulatory sensors is presented, followed by a summary of the developments in this field, with an emphasis on the clinical applications of falls detection, falls risk assessment, and energy expenditure. The importance of these applications is considerable in light of the global demographic trends and the resultant rise in the occurrence of injurious falls and the decrease of physical activity. The potential of using such monitors in an unsupervised manner for community-dwelling individuals is immense, but entails an array of challenges with regards to design c onsiderations, implementation protocols, and signal analysis processes. Some limitations of the research to date and suggestions for future research are also discussed.

Cardiorespiratory-based sleep staging in subjects with obstructive sleep apnea

A cardiorespiratory-based automatic sleep staging system for subjects with sleep-disordered breathing is described. A simplified three-state system is used: Wakefulness (W), rapid eye movement (REM) sleep (R), and non-REM sleep (S). The system scores the sleep stages in standard 30-s epochs. A number of features associated with the epoch RR-intervals, an inductance plethysmography estimate of rib cage respiratory effort, and an electrocardiogram-derived respiration (EDR) signal were investigated. A subject-specific quadratic discriminant classifier was trained, randomly choosing 20% of the subjects epochs (in appropriate proportions of W, S and R) as the training data. The remaining 80% of epochs were presented to the classifier for testing. An estimated classification accuracy of 79% (Cohens /spl kappa/ value of 0.56) was achieved. When a similar subject-independent classifier was trained, using epochs from all other subjects as the training data, a drop in classification accuracy to 67% (/spl kappa/=0.32) was observed. The subjects were further broken in groups of low apnoea-hypopnea index (AHI) and high AHI and the experiments repeated. The subject-specific classifier performed better on subjects with low AHI than high AHI; the performance of the subject-independent classifier is not correlated with AHI. For comparison an electroencephalograms (EEGs)-based classifier was trained utilizing several standard EEG features. The subject-specific classifier yielded an accuracy of 87% (/spl kappa/=0.75), and an accuracy of 84% (/spl kappa/=0.68) was obtained for the subject-independent classifier, indicating that EEG features are quite robust across subjects. We conclude that the cardiorespiratory signals provide moderate sleep-staging accuracy, however, features exhibit significant subject dependence which presents potential limits to the use of these signals in a general subject-independent sleep staging system.

Longitudinal Falls-Risk Estimation Using Triaxial Accelerometry

Falls among the elderly population are a major cause of morbidity and injury-particularly among the over 65 years age group. Validated clinical tests and associated models, built upon assessment of functional ability, have been devised to estimate an individuals risk of falling in the near future. Those identified as at-risk of falling may be targeted for interventative treatment. The migration of these clinical models estimating falls risk to a surrogate technique, for use in the unsupervised environment, might broaden the reach of falls-risk screening beyond the clinical arena. This study details an approach that characterizes the movements of 68 elderly subjects performing a directed routine of unsupervised physical tasks. The movement characterization is achieved through the use of a triaxial accelerometer. A number of fall-related features, extracted from the accelerometry signals, combined with a linear least squares model, maps to a clinically validated measure of falls risk with a correlation of ¿ = 0.81(p < 0.001).

Signal quality measures for pulse oximetry through waveform morphology analysis

Pulse oximetry has been extensively used to estimate oxygen saturation in blood, a vital physiological parameter commonly used when monitoring a subjects health status. However, accurate estimation of this parameter is difficult to achieve when the fundamental signal from which it is derived, the photoplethysmograph (PPG), is contaminated with noise artifact induced by movement of the subject or the measurement apparatus. This study presents a novel method for automatic rejection of artifact contaminated pulse oximetry waveforms, based on waveform morphology analysis. The performance of the proposed algorithm is compared to a manually annotated gold standard. The creation of the gold standard involved two experts identifying sections of the PPG signal containing good quality PPG pulses and/or noise, in 104 fingertip PPG signals, using a simultaneous electrocardiograph (ECG) signal as a reference signal. The fingertip PPG signals were each 1 min in duration and were acquired from 13 healthy subjects (10 males and 3 females). Each signal contained approximately 20 s of purposely induced artifact noise from a variety of activities involving subject movement. Some unique waveform morphology features were extracted from the PPG signals, which were believed to be correlated with signal quality. A simple decision-tree classifier was employed to arrive at a classification decision, at a pulse-by-pulse resolution, of whether a pulse was of acceptable quality for use or not. The performance of the algorithm was assessed using Cohens kappa coefficient (κ), sensitivity, specificity and accuracy measures. A mean κ of 0.64 ± 0.22 was obtained, while the mean sensitivity, specificity and accuracy were 89 ± 10%, 77 ± 19% and 83 ± 11%, respectively. Furthermore, a heart rate estimate, extracted from uncontaminated sections of PPG, as identified by the algorithm, was compared with the heart rate derived from an uncontaminated simultaneous ECG signal. The mean error between both heart rate readings was 0.49 ± 0.66 beats per minute (BPM), in comparison to an error value observed without using the artifact detection algorithm of 7.23 ± 5.78 BPM. These results demonstrate that automated identification of signal artifact in the PPG signal through waveform morphology analysis is achievable. In addition, a clear improvement in the accuracy of the derived heart rate is also evident when such methods are employed.

Tracking the Evolution of Smartphone Sensing for Monitoring Human Movement.

Advances in mobile technology have led to the emergence of the “smartphone”, a new class of device with more advanced connectivity features that have quickly made it a constant presence in our lives. Smartphones are equipped with comparatively advanced computing capabilities, a global positioning system (GPS) receivers, and sensing capabilities (i.e., an inertial measurement unit (IMU) and more recently magnetometer and barometer) which can be found in wearable ambulatory monitors (WAMs). As a result, algorithms initially developed for WAMs that “count” steps (i.e., pedometers); gauge physical activity levels; indirectly estimate energy expenditure and monitor human movement can be utilised on the smartphone. These algorithms may enable clinicians to “close the loop” by prescribing timely interventions to improve or maintain wellbeing in populations who are at risk of falling or suffer from a chronic disease whose progression is linked to a reduction in movement and mobility. The ubiquitous nature of smartphone technology makes it the ideal platform from which human movement can be remotely monitored without the expense of purchasing, and inconvenience of using, a dedicated WAM. In this paper, an overview of the sensors that can be found in the smartphone are presented, followed by a summary of the developments in this field with an emphasis on the evolution of algorithms used to classify human movement. The limitations identified in the literature will be discussed, as well as suggestions about future research directions.

Design of a Decision-Support Architecture for Management of Remotely Monitored Patients

Telehealth is the provision of health services at a distance. Typically, this occurs in unsupervised or remote environments, such as a patients home. We describe one such telehealth system and the integration of extracted clinical measurement parameters with a decision-support system (DSS). An enterprise application-server framework, combined with a rules engine and statistical analysis tools, is used to analyze the acquired telehealth data, searching for trends and shifts in parameter values, as well as identifying individual measurements that exceed predetermined or adaptive thresholds. An overarching business process engine is used to manage the core DSS knowledge base and coordinate workflow outputs of the DSS. The primary role for such a DSS is to provide an effective means to reduce the data overload and to provide a means of health risk stratification to allow appropriate targeting of clinical resources to best manage the health of the patient. In this way, the system may ultimately influence changes in workflow by targeting scarce clinical resources to patients of most need. A single case study extracted from an initial pilot trial of the system, in patients with chronic obstructive pulmonary disease and chronic heart failure, will be reviewed to illustrate the potential benefit of integrating telehealth and decision support in the management of both acute and chronic disease.

A comparison of activity classification in younger and older cohorts using a smartphone.

Automatic recognition of human activity is useful as a means of estimating energy expenditure and has potential for use in fall detection and prediction. The emergence of the smartphone as a ubiquitous device presents an opportunity to utilize its embedded sensors, computational power and data connectivity as a platform for continuous health monitoring. In the study described herein, 37 older people (83.9  ±  3.4 years) performed a series of activities of daily living (ADLs) while a smartphone (containing a triaxial accelerometer, triaxial gyroscope and barometric pressure sensor) was placed in the front pocket of their trousers. These results are compared to a similar trial conducted previously in which 20 young people (21.9  ±  1.65 years) were asked to perform the same ADLs using the same smartphone (again in the front pocket of their trousers).In each trial, the participants were asked to perform several activities (standing, sitting, lying, walking on level ground, up and down staircases, and riding an elevator up and down) in a free-living environment. During each acquisition session, the internal sensor signals were recorded and subsequently used to develop activity classifiers based on a decision tree algorithm that classified ADL in epochs of ~1.25 s. When training and testing with the younger cohort, using a leave-one-out cross validation procedure, a total classification sensitivity of 80.9% ± 9.57% ([Formula: see text] = 0.75  ±  0.12) was obtained. Retraining and testing on the older cohort, again using cross validation, gives a comparable total class sensitivity of 82.0% ± 8.88% ([Formula: see text] =0.74  ±  0.12).When trained with the younger group and tested on the older group, a total class sensitivity of 69.2% ± 24.8% (95% confidence interval [69.6%, 70.6%]) and [Formula: see text] = 0.60  ±  0.27 (95% confidence interval [0.58, 0.59]) was obtained. When trained on the older group and tested on the younger group, a total class sensitivity of 80.5% ± 6.80% (95% confidence interval [79.0%, 80.6%]) and [Formula: see text] = 0.74  ±  0.08 (95% confidence interval [0.73, 0.75]) was obtained.An instance of the decision tree classifier developed was implemented on the smartphone as a software application. It was capable of performing real-time activity classification for a period of 17 h on a single battery charge, illustrating that smartphone technology provides a viable platform on which to perform long-term activity monitoring.

ECG quality measures in telecare monitoring

We analyze the use of unsupervised ECG acquisition in the home environment. An algorithm for automatically marking ECG recordings for sections of obvious artifact is described. The algorithm was validated against a set of 150 records randomly chosen from a database of ECGs and manually annotated to identify sections of artifact. Using this algorithm 4751 single lead-I ECG recordings from 24 home-dwelling patients were examined. The ECGs were collected using a remote home monitoring system. The participant ages (N=24) ranged from 54–92 years and were suffering either chronic obstructive pulmonary disease and/or congestive heart failure. Percentages of amplifier saturation, high frequency artifact, low signal power and the maximum continuous section of useable ECG are quoted. 1344 records were found to contain no artifact, while 3506 records contained 10 seconds or more of uninterrupted ECG (including the 1344 with no artifact). The results show that in the majority of cases, the capture of ECG in an unsupervised home environment is achievable.

A wearable triaxial accelerometry system for longitudinal assessment of falls risk

Falls-related injuries in the elderly population are a major cause of morbidity and represent one of the most significant contributors to hospitalizations and rising health care expense in developed countries. Many laboratory-based studies have described falls detection systems using wearable accelerometry. However, only a limited number of reports have tried to address the difficult issues of falls detection and falls prevention in unsupervised or free-living environments. We describe a waist-mounted triaxial accelerometry (Triax) system with a remote data collection capability to provide unsupervised monitoring of the elderly. The basis of the monitoring is a self-administered directed-routine (DR) comprising three separate tests measured by way of the Triax. We present an initial evaluation of the DR results in 36 patients to detect early changes in functional ability and facilitate falls risk stratification. Extracted features considered alone show a correlation with falls risk of approximately ρ=0.5. Estimation of falls risk using a linear least squares model provides a root-mean-squared error of 0.69 (ρ=0.58, p<0.0002).