A decision support architecture for telecare patient management of chronic and complex disease.

A major challenge facing designers of telecare systems today is providing decision support to enhance the health carer's review of remotely acquired monitoring data and to support clinical decision-making for the management of chronic and complex disease in this setting. We are implementing a decision support framework to analyze clinical information generated from subjects at their place of residence (home, residential care settings) and from other clinical environments. The telecare information generated from these environments is both substantial and multi-modal (physiological, questionnaire, medication data, etc.). Using the JBoss Application Server, a rules engine is used to analyze these data. The health carer will be alerted to any deterioration in the health status of a patient by way of a Web page that will stratify a clinical data summary into high, medium and low risk groups. In this way, outputs from the decision support system can be used to assist in the efficient review and risk stratification of multiple patient records, and ultimately influence changes in work flow by targeting scarce human resources to patients of most need.

Characterization of PIR detector for monitoring occupancy patterns and functional health status of elderly people living alone at home.

Changes in the behavioral activity of elderly people living alone at home may be an indicator for the assessment of their functional health status. In this paper, we evaluate the characteristics of a pyro-electric infrared detector (PIRD) to identify any section of the room where detector will fail to respond and assess the number of detectors required to identify the movements of the occupant reliably. The experimental results suggest that the PIRD output decreases very sharply from 1 V at 3 m to 0.2 V at 5 m and is below 0.5 V for hand movements at detection angle of <20 degrees and >50 degrees . Hence, spatial sensitivity of PIRD is non-uniform at different distances for different walking speeds. If a detector is used in a room of dimension half the maximum range of the PIRD, there exist dead points where detection sensitivity is low giving rise to false alarms. Hence, a single detector installed in a room fails to respond to small movements in the extended range. Four detectors at each corner of the room help to increase the sensitivity of the detection whereby even small movements such as reading a book, telephonic conversation etc. can be identified.

Time constant of heart rate recovery after low level exercise as a useful measure of cardiovascular fitness.

In this study we aimed to establish the usefulness of the time constant of heart rate recovery (Tr) in the evaluation of cardiovascular fitness. 15 male subjects exercised on recumbent bicycle at three different workloads (75 W, 100 W 125 W) where R-R intervals were monitored to determine Tr. In order to find the maximal oxygen uptake (VO2max) of each subject, oxygen consumption rate (VO2) was recorded throughout the treadmill exercise (10 km/h). Based on VO2max, we classified the subjects into two groups: the "fit" group and the "unfit" group. We found a significant difference in Tr between these two groups only existed when the workload was 75 W (

Analysis of orientation error of triaxial accelerometers on the assessment of energy expenditure.

This paper investigates the effects of orientation error in the positioning of triaxial accelerometers on the assessment of energy expenditure. Four subjects walked on a treadmill at varying velocities ranging from 4km.h-1to 5km.h-1. During each test, a triaxial accelerometer attached to the lower back at arbitrary orientations to record body accelerations. Energy expenditure was estimated by the sum of the integrals of the absolute value of accelerometer output from all the three measurement directions. Based on theoretical analysis and experimental observations, it is concluded that small orientation errors (< 3°) have no distinguishable effects on the estimation of energy expenditure. We propose an efficient method to compensate for larger orientation errors. The experimental results verified the effectiveness of this proposed compensation method.

Modeling of a gas concentration measurement system.

Energy expenditure can be calculated via measurement of oxygen consumption and carbon dioxide production. Precise measurement of expired gas concentrations and volume is required for this determination. For a given gas concentration measurement system, the establishment of a model is a good way to effectively use the equipments and achieve more accurate energy expenditure calculations. This paper proposes a simple but effective approach for the modeling of a gas concentration measurement system.

A methodology to monitor the changing trends in health status of an elderly person by developing a markov model.

In this paper we are proposing a statistical testing methodology to monitor changing trends in the health status of elderly people. The occupancy pattern of elderly people can be modeled using a Markov chain, estimating transition probabilities of the chain and test hypotheses about them. The profile of the person for a given period can be stored as a transition matrix of a discrete, regular, ergodic Markov chain. The observation of the occupancy pattern for a given test period can be established as a test Markov chain using information from sensors such as infrared sensors, magnetic switches etc. In the absence of real time data, we have used uniformly distributed transition probabilities to define the profile of the Markov chain and then generated test Markov chain based on this model. The transition probabilities are extracted for the test and profile Markov chain using Maximum Likelihood Estimates (MLE). The statistical testing of occupancy monitoring establishes a basis for statistical inference about the system performance without generating any real time statistics for the occupancy pattern. Chi square test and likelihood ratio tests ensure that the sequences generated from the two Markov chains are statistically same. Any difference in profile Markov chain and test Markov chain could indicate a changed health status of the elderly person.

Heart rate control during treadmill exercise.

A computer-controlled treadmill and related data collection and processing systems have been developed for the control of heart rate during treadmill exercise. Minimizing deviations of heart rate from a preset profile is achieved by controlling the speed and/or the gradient of the treadmill. A simple and practical heart rate measurement algorithm has been developed to robustly measure the variations of heart rate. Both conventional Proportional-Integral-Derivative (PID) control and fuzzy Proportional-Integral (PI) control approaches have been employed for the controller design. The fuzzy Proportional-Integral algorithm achieved better heart rate tracking performance. Finally, a heart rate based exercising protocol was successfully implemented on the newly designed exercise system.

Classification of basic daily movements using a triaxial accelerometer.

A generic framework for the automated classification of human movements using an accelerometry monitoring system is introduced. The framework was structured around a binary decision tree in which movements were divided into classes and subclasses at different hierarchical levels. General distinctions between movements were applied in the top levels, and successively more detailed subclassifications were made in the lower levels of the tree. The structure was modular and flexible: parts of the tree could be reordered, pruned or extended, without the remainder of the tree being affected. This framework was used to develop a classifier to identify basic movements from the signals obtained from a single, waist-mounted triaxial accelerometer. The movements were first divided into activity and rest. The activities were classified as falls, walking, transition between postural orientations, or other movement. The postural orientations during rest were classified as sitting, standing or lying. In controlled laboratory studies in which 26 normal, healthy subjects carried out a set of basic movements, the sensitivity of every classification exceeded 87%, and the specificity exceeded 94%; the overall accuracy of the system, measured as the number of correct classifications across all levels of the hierarchy, was a sensitivity of 97.7% and a specificity of 98.7% over a data set of 1309 movements.

A home health monitoring system including intelligent reporting and alerts.

We describe the design and implementation of an intelligent reporting and alerts system that has been designed with a specific goal to address the needs of managing chronic and complex disease through the use of home telecare technology. Our approach has been to develop these tools using as far as possible, open standards. Clinical measurement data gathered using home telecare and stored in a relational database in XML format is extracted and converted into a Clinical Document Architecture (CDA) as defined by the Health Level 7 (HL7) organization. Data trends are presented to the clinician as simple graphs and summary statistics (means, standard deviations) over time for an individual patient. Clinicians may receive this data by display through a Web-interface or by email or faxed reports. A Ripple Down Rules (RDR) knowledge base supports more complex decision-making provided in the Alerts module. The RDR output is incorporated into the output reports as a textual statement, and/or a graphical highlighting of key parameters in the trends images and tables. Rule development and validation is part of ongoing research.

Detection of daily physical activities using a triaxial accelerometer.

Triaxial accelerometers have been employed to monitor human movements in a variety of circumstances. The study considered the use of data from a single waist-mounted triaxial accelerometer to distinguish between activity states and rest A method using acceleration magnitude was applied to data collected from 26 normal subjects performing sit-to-stand and stand-to-sit transitions and walking. The effects of three parameters were investigated: the length n of a smoothing median filter, the width w of the averaging window used to process the signal and the value of the acceleration magnitude threshold th. These were found to be inter-related, and sets of parameters that resulted in accurate discrimination were determined by the relationship between th and the product of w and n, and by the relationship between n and w. The subjects were randomly divided into control (N = 13) and test (N = 13) groups. Optimum parameter sets were determined using the control group. Eleven sets of parameters yielded the same optimum results of a sensitivity of 1.0 and a specificity of 0.96 in the control group. Upon application to the test group, using these parameters, the system successfully distinguished between activity and rest, giving sensitivities greater than 0.98 and specificities between 0.88 and 0.94.

Phase response of model sinoatrial node cells.

When a brief current pulse is incident on excitable cells in cardiac and other nervous tissue, a change in phase of the cell's response is usually observed. In cardiac tissue, the cells are exposed to external stimulation of mainly positive currents, which depolarize the cells. We performed a systematic study of the effect of depolarizing stimuli, covering timing, magnitude, and duration, and demonstrated that all of these parameters influence the phase response of the cell. The phase response of our model cell compares favorably with measurements on isolated sinoatrial node cells. We investigated the phase response to single depolarizing stimuli as a function of the stimulus parameters (phase response curves), and then studied cell responses to the combined effect of a pulse train (entrainment phenomena). The range of magnitudes and durations for the stimuli were 0.01-5 nA and 0.01-50 ms. Comparisons of the entrainment properties of the model with experimental results show good agreement with similar modes and different entrainment ratios occurring for similar basic cycle lengths (as functions of the unperturbed cell period). Our results demonstrate that any combination of parameters that provide the same charge transfer to the cell causes a similar phase response, independent of the specific magnitude and duration for the entire range of stimuli investigated.

Inhomogeneity of action potential waveshape assists frequency entrainment of cardiac pacemaker cells.

In this paper, we have employed ionic models of sinoatrial node cells to investigate the synchronization of a pair of coupled cardiac pacemaker cells from central and peripheral regions of the sinoatrial node. The free-running cycle length of the cell models was perturbed using two independent techniques and the minimum coupling conductance required to achieve frequency entrainment was used to assess the relative ease with which various cell pairs achieve entrainment. The factors effecting entrainment were further investigated using single-cell models paced with an artificial biphasic coupling current. Our simulation results suggest that dissimilar cell types, those with largely different upstroke velocities entrain more easily, that is, they require less coupling conductance to achieve 1:1 frequency entrainment. We, therefore, propose that regional variation in action-potential waveshape within the sinoatrial node assists frequency synchronization in vivo.

Vagal entrainment of heart rate is simulated by an integrator with feedback.

Paradoxical stable entrainment of heart rate to inhibitory vagal impulses can be simulated with two distinct mathematical models; a complex ionic current model of sinoatrial node pacemaker activity, as well as a simple integrator with non-linear feedback. We show that both models exhibit similar entrainment characteristics to repetitive vagal stimuli. By applying a sharp disturbance to each model whilst entrained, the subsequent path of cycle length recovery can be described by dynamic phase response curves and phase-phase plots, the properties of which dictate whether stable entrainment is possible.

Information technology in primary health care.

Appropriate application of information technology in primary health care will extend traditional diagnosis and patient management beyond the doctor's clinic into the everyday living environment. We describe a model of information management in primary health care, and place special emphasis on the emerging areas of clinical decision support, computerised clinical measurements, patient education and network connectivity. Briefly discussed is the design of innovative home monitoring techniques and a telemedicine based communications infrastructure that should improve access to high quality primary health care for all citizens, irrespective of their distance from major urban centres. A preliminary design for a telemedicine-assisted primary health care network is presented, based on this model of information management. The premise is that improvements in health care services and reductions in health care costs can be effected by establishing a continuum of patient care from the patient's home, to the doctor's surgery, to speciality services in hospitals and to other service providers in the health care sector. While, the proposal focuses on new opportunities arising from the imminent introduction of broad band interactive fibre optic networks throughout Australia, the technology and projected data transfers could easily be handled in the short-term using modem access to the standard telephone network. A simple connectivity scheme for system integration is also presented.

A web-based approach for electrocardiogram monitoring in the home.

A Web-based electrocardiogram (ECG) monitoring service in which a longitudinal clinical record is used for management of patients, is described. The Web application is used to collect clinical data from the patient's home. A database on the server acts as a central repository where this clinical information is stored. A Web browser provides access to the patient's records and ECG data. We discuss the technologies used to automate the retrieval and storage of clinical data from a patient database, and the recording and reviewing of clinical measurement data. On the client's Web browser, ActiveX controls embedded in the Web pages provide a link between the various components including the Web server, Web page, the specialised client side ECG review and acquisition software, and the local file system. The ActiveX controls also implement FTP functions to retrieve and submit clinical data to and from the server. An intelligent software agent on the server is activated whenever new ECG data is sent from the home. The agent compares historical data with newly acquired data. Using this method, an optimum patient care strategy can be evaluated, a summarised report along with reminders and suggestions for action is sent to the doctor and patient by email.

The potential impact of home telecare on clinical practice.

Home telecare, in which the health status of patients at home is monitored remotely, has the potential to improve care and reduce costs. Its widespread implementation would require fundamental changes in the healthcare system.

Review of ionic models of vagal-cardiac pacemaker control.

Mathematical models of ion currents in pacemaker cells of the heart and their associated modulation by vagal stimulation have provided numerous insights into the ionic mechanisms underlying parasympathetic control of heart rate. In this article, ionic models described in the literature are reviewed and compared, with a view to examining their effectiveness in reproducing known chronotropic responses to vagal stimulation.

Simulations of postvagal tachycardia at the single cell pacemaker level: a new hypothesis.

Simulations performed on a single cell model of rabbit sinoatrial node activity after prolonged vagal stimulation have been able to reproduce the known characteristics of cycle length recovery, including the presence of rapid and slow recovery phases and the transient undershoot phenomenon known as postvagal tachycardia (PVT). In the model, the PVT component has been hypothesized to result from the recovery of background levels of the muscarinic K+ current iK,ACh from desensitization due to prolonged exposure to acetylcholine (ACh) neurotransmitter. Other components of the recovery were found to be due to the inactivation of iK,ACh after the hydrolysis of ACh (rapid phase) and the recovery of the hyperpolarizing-activated current i(f) from its ACh-induced inhibition (slow phase). The magnitudes of both the rapid component and the PVT were found to increase linearly with preceding vagally mediated increase in cycle length, whereas the gain of the slow component was found to saturate, reflecting the limited contribution of i(f) inhibition to cycle prolongation.

Vagal control of sinoatrial rhythm: a mathematical model.

The ionic mechanisms underlying vagal control of the cardiac pacemaker were investigated using a new single cell mathematical model of sinoatrial node electrical activity. The model was formulated from a wide range of electrophysiological data available in the literature, with particular reference to whole cell recordings from enzymatically isolated sinoatrial node cells. Development of the model was prompted by the lack of an existing physiologically accurate formulation of sinoatrial node activity that could reproduce the known complex chronotropic response of the pacemaker to brief-burst vagal stimulation, as observed in whole animal and isolated sinus node preparations. Features of the model include the dynamic modulation of the hyperpolarisation-activated current (i(f)) and the L-type calcium current (iCa,L) by acetylcholine, the improved characterisation of the muscarinic potassium current (iK,ACh), assigning the entire background potassium current (ib,K) to spontaneous openings of its channels, and the utilisation of second order kinetics for acetylcholine within the neuroeffector junction. Simulations performed using brief vagal stimuli elicited a strong hyperpolarisation of the membrane which prolonged the cycle in which it was delivered in a phase-dependent manner. This phase-dependency was presented in the form of a standard phase response curve which was characterised by a positive linear slope region, a breakpoint characteristic and a "no effect" zone in which the vagal pulse could no longer prolong the cycle. The breakpoint was manifested as a discontinuity in the curve which was examined by bracketing this point at the limit of the double precision arithmetic employed. At these boundary points on either side of the breakpoint, the vagal stimulus was able to activate outward iK,ACh in such a manner as to finely balance the increasing inward iCa,L trying to generate phase 0 upstroke. On decay of iK,ACh, the membrane either subsequently repolarised or fired to produce an action potential depending on the precise phase of the stimulus. The positive linear slope portion of the PRC was characterised by a strong resetting type behaviour in which the membrane hyperpolarised to approximately the same value, irrespective of the phase of stimulus delivery. For vagal stimulus bursts applied throughout the "no effect" zone, outward iK,ACh was not sufficiently activated in order to overcome the strong inward drive of iCa,L and could not prevent upstroke occurring. For these vagal stimuli, the subsequent cycle was hyperpolarised and prolonged. The size of the "no effect" zone was directly related to the inherent latency incorporated in the activation characteristic of iK,ACh. In contrast to previous models of vagal pacemaker control, our new model was able to reproduce the classical triphasic chronotropic response to brief vagal stimulation characterised by a primary inhibition response, a postinhibitory rebound and a secondary inhibition response. In particular, the postinhibitory rebound was due to activation of the inward hyperpolarisation-activated current by the vagally-induced membrane hyperpolarisation, whilst the secondary inhibition phase resulted from the inhibition of the hyperpolarisation-activated current by acetylcholine. The model suggests that the complex chronotropic responses of the cardiac pacemaker to brief vagal stimulation arises from inherent ionic mechanisms operating within the sinoatrial node.