EP2109398A1

EP2109398A1 - Wireless temporal-spatial mobility and electrocardiogram analyzer system and method

Info

Publication number: EP2109398A1
Application number: EP08713012A
Authority: EP
Inventors: Alex Kalpaxis
Original assignee: Individual
Current assignee: Individual
Priority date: 2007-01-04
Filing date: 2008-01-04
Publication date: 2009-10-21
Also published as: WO2008085923A1; CA2674500A1

Abstract

Disclosed is a method and system using an apparatus for wireless autonomous temporal-spatial mobility and electrocardiogram (EKG) simultaneous data capture for the purposes of critical and non-critical cardiac event detection, alarming, heuristic archiving, and correlation analysis. A wireless device simultaneously measures acceleration data and EKG data and transmitting the measured data. A wireless collector analyzer receives and processes the measured data, wherein the collector analyzer will output the results of the processing to a storage device or user.

Description

WIRELESS TEMPORAL-SPATIAL MOBILITY AND ELECTROCARDIOGRAM ANALYZER SYSTEM AND METHOD

FIELD OF THE DISCLOSURE

[0001] Exemplary embodiments relate to a method and apparatus for wireless autonomous temporal-spatial mobility and electrocardiogram (EKG) monitoring using simultaneous data capture devices for the purposes of critical and non-critical cardiac event detection, alarming, heuristic archiving, and correlation analysis. More specifically, the exemplary embodiments relate to a method and apparatus for detecting, monitoring and profiling/correlating EKG signals and acceleration data as a result of induced mobility stress, for instance, running.

BACKGROUND

[0002] ' Induced mobility stress can result from activities such as running, stair climbing, sports activities, heavy weight lifting, heavy physical work and the like. Specific induced mobility stress cardiac events will require critical event processing such as arrhythmias, myocardial ischemia, angina and acute myocardial infarctions. Cardiac events such as arrhythmias, myocardial ischemia, angina and acute myocardial infarctions will also be captured even if there is no mobility and the user is in a resting state. [0003] Wireless relay of information regarding these critical series of events to one or many monitor/alarm facilities for immediate notification to potential responders can be improved. In addition, wireless relay of information regarding these critical events to a collector facility that is in an exemplary configuration, attached directly to a medical managed service provider or care giver would be beneficial.

SUMMARY

[0004] It is desirable to provide a method and apparatus for wirelessly monitoring and recording a user's (i.e., wearer of the disclosed device) mobility, which directly correlates to energy output. Simultaneously, the user's electrocardiogram (EKG) is monitored and recorded in real-time. Alarm events are generated autonomously without requiring user intervention. The device accomplishes this by monitoring in real-time, motion in three dimensions, while simultaneously capturing and recording the user's EKG for cardiac event processing, and mobility versus EKG trace correlation analysis. The system has two main components: the device worn by the user, called a Wireless Mobility EKG device, and a collection monitor and analyzer back-end facility, called a Wireless Mobility EKG collector analyzer. The system comprising the device and collection monitor is called the Wireless Mobility EKG system.

[0005] The device, which monitors in real-time, motion in three dimensions, while simultaneously capturing and recording the user's EKG for cardiac event processing, is preferably a low-powered wireless portable device monitor which is small, lightweight and is concealable. The device comprises a Micro-controller Processor Unit (MPU), a Micro Electro Mechanical System (MEMS) based 3-dimensional accelerometer, an electrocardiogram (EKG) capturing signal processor and a wireless sensor network transceiver to communicate 3-dimensional accelerometer motion and simultaneously captured EKG signal data to a collection node, and optionally to a securely attached Internet-enabled PC.

[0006] In an exemplary embodiment, the Wireless Mobility EKG system provides a wearable, wireless, real-time mobility sensing and EKG device for self-monitoring and recording mobility in three dimensions, while simultaneously self-monitoring and recording the wearer's (i.e., user's) EKG. The device, preferably smaller than a cell phone, allows for real-time activity and EKG trace correlation and profiling using the data it relays wirelessly to the collector analyzer.

[0007] The mobility patterns recorded are three dimensional acceleration vectors that directly correlate to user energy output over time. This allows for active EKG and/or mobility monitoring for not only high risk target groups, but concerned individuals with low risk incidents in their family history. In addition to the aging population, a growing portion of the child/adult population is experiencing dramatic increases in body weight, which directly affects cardiac health. [0008] The Wireless Mobility EKG system can also monitor high risk athletes for cardiac failure due to exercise stress and/or dietary supplement and/or drug usage. The device preferably permits cardiac "pre-scan" in a no-stress/no-cost environment like a home, compared to a clinical or hospital setting. The device also comprises a mode for electrode, conduction contact, and cabling integrity analysis for use with other EKG devices. [0009] The capability to do an EKG trace correlation with mobility simultaneously and in real-time is a unique function of the Wireless Mobility EKG system. Currently, the closest possibility is a Holter device (not very portable) for EKG tracing but provides no mobility capture and relies on the wearer of the Holter device to remember (which is ineffective except in major cases of cardiac malfunction) what they were doing when cardiac events occur, that is, if they are aware of them, which is not always the case. Another aspect of the Wireless Mobility EKG system is that the device that can be attached to the wearer via EKG electrodes, is battery powered preferably at 4 volts DC, and is wirelessly coupled (no physical main power line electrical connections) to the collector analyzer back-end, which is attached to a PC or monitor device.

[0010] The Wireless Mobility EKG system can be used as a screening tool in the evaluation of users who are curious, or suspect, or have existing symptoms of various forms of potential heart related issues, or in situations where there is known cardiac pathology in the absence of symptoms. Symptoms such as palpitations, lightheadedness or fainting may be caused by disturbances in the electrical signals that control the heart muscle contractions. These disturbances can be random, spontaneous, sleep-related, or emotion- or stress-induced. [0011] EKG monitoring is an ideal test because the user performs normal, daily activities, increasing the likelihood that the user will experience the precise situations that can trigger symptoms or cardiac events. This allows correlation of any rhythm problems or abnormalities with activities and/or symptoms. In the opposite sense, EKG monitoring can also be used to "rule out" cardiac causes of user symptoms. [0012] Users will wear the Wireless Mobility EKG system's device monitor, which is portable, longer and more often, increasing the likelihood of catching cardiac events and trends, because of its small size and limited weight. The Wireless Mobility EKG system's portable device monitor is small, lightweight and easily concealable. Variable EKG and mobility record lengths are possible for a user's different needs.

[0013] The Wireless Mobility EKG system supports a television interface (when desired) for use with a standard TV set as a display monitor. This feature addresses the households where PCs or laptops are not available for the Wireless Mobility EKG system computer application. There are situations where the home PC hardware or laptop is used for other applications, and is not readily available or does not exist, which is the case in many geriatric or assisted-living homes. Television (TV) sets are usually available especially older ones that could be used with the Wireless Mobility EKG system. The TV interface used by the Wireless Mobility EKG system can support TV sets as old as 40 years.

[0014] The Wireless Mobility EKG system PQRST signal profile analysis routines differentiate electronic noise and muscular artifact from real cardiac events. The Wireless Mobility EKG system algorithms analyze initial impedance values for electrode skin contact points. This is very critical for proper EKG signal capture and accurate EKG tracing. This also allows for testing to determine which brand of stress electrode works best for each particular user. There are many electrode manufacturers with widely varying configurations and electro-conductive gels. The capability, using testing hardware to automatically determine electrode and a cable impedance provides a customized device for each particular user.

[0015] User skin types and preparation requirements with the various electrode implementations creates a large set of permutations that frustrates the electrode preparation and attachment process. Another important aspect to electrode characterization is that in many instances, EKG electrode positions are changed over the course of EKG tracing session, and the site preparation process must be restarted. There are no guarantees that new electrode site preparation will be the same as the prior electrode site preparation. The Wireless Mobility EKG system helps create a relatively consistent electrode site preparation process because of the immediate EKG electrode/site impedance measurement feedback it provides the user. The Wireless Mobility EKG system will also archive this electrode, cable and preparation characterizations for future reference. The Wireless Mobility EKG system algorithms and automated EKG electrode/cable impedance testing hardware not only provides an important characterization feature to the Wireless Mobility EKG system, but can be used with any of the EKG monitoring/recording devices in use in the marketplace.

[0016] Therefore, it is an object of the exemplary embodiments to provide a method and apparatus to autonomous and simultaneously, capture in real-time, temporal-spatial mobility and electrocardiogram (EKG) data for the purposes of critical and non-critical cardiac event detection, alarming generation, heuristic archiving, and correlation and relaying these data/events in a wireless manner to collection computer for further alarm generation and event processing. [0017] It is a further object of the exemplary embodiments to provide a method and apparatus for real-time profiling and correlating human temporal-spatial mobility and electrocardiogram (EKG) events with stored templates to determine critical vs. normal human mobility/EKG events and/or temporal-spatial cardiac behaviors.

[0018] It is a further object of the exemplary embodiments to provide a method and apparatus for real-time/heuristic information gathering to allow for the measuring and detection of temporal-spatial mobility and EKG related events correlated with specific disease progression.

[0019] It is a further object of the exemplary embodiments to provide a method and apparatus for real-time/heuristic information gathering to allow for the measuring and detection of temporal-spatial mobility and EKG related events correlated with specific medication scheduling and dosing. [0020] It is a further object of the exemplary embodiments to provide a method and apparatus for detecting in real-time, various states of motion such as static, rollover, free-fall, impact, shaking, and complex linear/angular motion generated by the wearer of the device, while simultaneously capturing the EKG data from the wearer of the device, and relaying the data to the collection monitor and analyzer back-end facility (e.g., collector analyzer server).

[0021] It is a further object of the exemplary embodiments to provide a method and apparatus for detecting and analyzing motion generated by the wearer of the device, which is then used as input to calculate the differential acceleration time derivatives ([d(A_x)/dt]² + [d(A_y)/dt]² + [d(A_z)/dt]²), which then is used to calculate the energy expenditure by the wearer of the device. This is done by calculating d(Sχyz)/dt = (([d(A_x)/dt]² + [d(A_y)/dt]² + [d(A_z)/dt]²))^'Λ, and an overall work function (user energy expenditure) Work(x,y,z,t) ~ M_h(A_x∑JJA_xdt + A_yΣjJA_ydt + A_zjJA_zdt. [0022] Another object of the exemplary embodiments is to provide a method and apparatus for generating alarms and alerts based on pre-determined rules on temporal-spatial mobility and EKG events that have been analyzed by the collector analyzer server from data it has received wirelessly from the device. These alarms, alerts, and spatial-temporal data will be sent via a securely attached Internet-enabled PC or laptop to medical service providers, or to individuals identified as responders (neighbors, friends/family, emergency service providers such as paramedics, fire or ambulance). [0023] A further object of the exemplary embodiments is to provide a method and apparatus for detecting and monitoring an individual's degree of inactivity as it relates to the individual's EKG. The collector analyzer server can profile the individual's inactivity that is correlated with the individual's EKG against predetermined rules. If there are excessive inactivity episodes and/or critical EKG events detected within a selected time period, notification will be generated and appropriate alarms and alerts will be generated and forwarded via predefined personalized call-lists. Physical conditions, such as stroke, congestive heart disease, coronary artery disease, arthritis, macular degeneration, paralysis, neuromuscular disease (such as Parkinson's, Multiple Sclerosis, Cerebral Palsy), amputation and osteoporosis, which greatly limit an individual's mobility thus producing periods of inactivity and/or critical EKG events as a function of disease progression can be profiled and monitored to determine if alarms are necessary. [0024] It is a further object of the exemplary embodiments to provide a method and apparatus for determining the stages of mobility stress induced cardiac electrical signal irregularities that control the heart muscle contractions such as palpitations, lightheadedness or fainting. The system captures and archives mobility and EKG data together which allows for temporal-spatial correlation of significant motion events with EKG data that contain cardiac electrical signal irregularities. The system will be able to detect the per-cursor stages to these cardiac electrical signal irregularities and will provide alerting/reporting when desired. These cardiac events will also be profiled over specified periods of time to allow for analysis such as heuristic cardiac disease progression.

[0025] Another object of the exemplary embodiments is to provide a method and apparatus for capturing mobility with EKG data simultaneously, and within this mobility/EKG data, the exemplary embodiments can identify unstable angina, ST- elevation/non-ST-elevation myocardial infarction, dynamic coronary occlusive process, plaque rupture, platelet stimulation, coronary vasospasm, thrombus formation and the like, ST-segment analysis can determine the occurrence/frequency and severity of ischemic episodes.

[0026] A further object of the exemplary embodiments is to provide an apparatus which is compact and small enough for monitoring that allows the user to assume normal, daily activities, increasing the likelihood that the user will experience the precise mobility stress induced situations that can trigger symptoms or cardiac events. This allows correlation of any rhythm problems or abnormalities with activities and/or symptoms. In the opposite sense, system monitoring can also be used to "rule out" cardiac causes of user symptoms. [0027] Another object of the exemplary embodiments is to provide a method and apparatus for correlating medication schedules and dosing that may contribute to cardiac issues as a result of the captured EKG cardiac event data from the user whether they are young or old. The risk of serious cardiac issues may increase when a user is taking four or more medications, when the user is placed on a new medication and/or when the dose of user's current medication is increased for medical reasons. Medications are not cleared as easily from the body as user gets older, and as a result, side effects can be stronger. Side effects of these medications that put the user at risk include decrease in blood pressure when trying to stand up, dizziness, drop in blood sugar, slowing of the heart rate, loss of balance and/or dehydration. The system monitoring will be used to correlate drastic movement events with the recorded EKG data over any time-period with medication schedules and/or dosing.

[0028] Another object of the exemplary embodiments is to provide a method and apparatus for directed automated induced mobility stress testing such as competitive sports activities. The Wireless Mobility EKG system can provide temporal-spatial measurements of mobility and EKG data and archive in real-time these measurements. Examples of induced mobility stress are sprint running, repetitive weight lifting, sports play and rapid distance walking. The Wireless Mobility EKG system can archive performance of these stresses as a baseline, and this can be repeated periodically to identify any changes in specific EKG data over a selected time-period.

[0029] It is a further object of the exemplary embodiments to provide a method and apparatus for calculating the mean and standard deviation of the heart rate over any time intervals as direct function of mobility. The Wireless Mobility EKG system will automatically detect and report values of PQRST onset, amplitude and intervals either during (online) or after (offline) recording of an EKG trace with the correlated mobility trace. The Wireless Mobility EKG system can also provide the user with graphical plots of EKG data such as QT segment interval versus time, QT segment interval versus RR interval, RR interval with correlated mobility data such as significant energy expenditures such running/stressful workout versus time and ST segment interval versus time as the EKG/mobility trace data is being generated and analyzed. [0030] These and other objects of the Disclosure are achieved by the method and apparatus herein described and are readily apparent to those skilled in the art upon review of the following drawings, detailed description and claims.

BRIEF DESCRIPTION OF THE DRAWINGS [0031] In order to show the manner that the above recited and other advantages and objects of the dislcosure are obtained, a more particular description of the preferred embodiments of the disclosure, which are illustrated in the appended drawing, is described as follows. The reader should understand that the drawings depict only preferred embodiments of the disclosure, and are not to be considered as limiting in scope.

[0032] A brief description of the drawings is as follows:

[0033] Figure 1 is an exemplary process flow chart showing the information flow and processing steps of the wireless mobility and EKG data capture, collection, analysis and archiving; [0034] Figure 2 is time-series plot of captured mobility that represent differential acceleration time derivatives ([d(A_x)/dt]² + [d(A_y)/dt]² + [d(A_z)/dt]²) for three dimensional mobility detection which is a result of the user wearing the device sending three dimensional acceleration (A_x, A_y, A_z) and EKG data to the collector analyzer server; [0035] Figure 3 is an exemplary time-series plot of the simultaneously captured EKG data as it relates to Figure 2 which is a result of a user wearing the device sending three dimensional acceleration (A_x, A_y, A_z) and EKG data to the collector analyzer server; [0036] Figure 4 is an exemplary expanded time-series plot of the simultaneously captured EKG data which is a result of user wearing the device which sends mobility and EKG data to the collector analyzer server;

[0037] Figure 5 is an exemplary time domain plot showing the distance traversed by the user wearing the device which is sending three dimensional acceleration (A_x,

A_y, A_z) and EKG data to the collector analyzer server, which is calculating the distance traversed by a wearer of the device using normalized position vectors for this figure;

[0038] Figure 6 is an exemplary block diagram illustrating the device and the collection analyzer server;

[0039] Figure 7 illustrates exemplary non-interrupt routines that are system initializations;

[0040] Figure 8 illustrates an exemplary block diagram of interrupt handlers;

[0041] Figure 9 is an exemplary sequence diagram of successful transmission of acceleration vectors and EKG data;

[0042] Figure 10 illustrates exemplary internal subsystems of the collector analyzer server;

[0043] Figure 11 illustrates an exemplary block diagram of an eleventh-order filter; and [0044] Figure 12 illustrates an exemplary block diagram of this n^th-order filter.

DETAILED DESCRIPTION

[0045] Reference will now be made in detail to the present preferred embodiment of the disclosure, examples of which are illustrated in the accompanying drawings. Figure 1 is an exemplary system diagram showing the end-to-end processing steps of the device and the collector analyzer server to a medical managed service provider as well as the data flow between the processing steps. [0046] The device 110-112 is worn by a user, preferably different users, to be monitored, and it preferably comprises three accelerometers, one for each dimension X (A_x), Y (A_y) and Z (A_z) used to measure motion (mobility) and an EKG signal trace capture system. Each device 110-112 preferably comprises a Micro-controller Processor Unit (MPU), a Micro Electro Mechanical System (MEMS) based 3- dimensional accelerometer, electrocardiogram (EKG) capturing signal processor, and a wireless sensor network transceiver to communicate 3-dimensional accelerometer motion, and simultaneously captured EKG signal data to the collection node and optionally attached to a securely attached Internet-enabled PC, laptop, personal computing device, personal digital assistant or the like. [0047] The device 110-112 can be enclosed in a portable cell phone-type package of less than 1 inch in thickness and weighing about 3 ounces. The device 110-112 is preferably worn on a belt like a cell phone, although other locations are suitable. The design uses micro-electro machine systems (MEMS) based three dimensional motion detectors and extremely low power, self-healing, reliable IEEE 802.15.4 "ZigBee" mesh network wireless technology. Of course, other suitable wireless technology can be used to implement the exemplary embodiments. [0048] The EKG analog front-end uses highly compact and very high-precision instrumentation amplifiers. The EKG signals are filtered and processed using several digital signal processing algorithms. The device 110-112 sends the data wirelessly to the collection analyzer 120 which is typically the collection analyzer front-end (USB) connected to a broadband connected PC that can optionally be connected in an instant-on mode to a medical service provider (MSP) 130. The collection analyzer 120 front-end preferably has a small form factor of 4 inches x 3 inches x 1 inch, for example, and has the device's 110-112 complement wireless interface for receiving mobility and EKG data and transmitting commands/control data to the respective device 110-112.

[0049] Various states of motion, such as static, rollover, free-fall, impact, shaking, complex linear and angular motion, are detected. The exemplary method can be implemented using a differential acceleration time derivative algorithm with heuristic functionality. The output of each acceleration axis and the corresponding real-time EKG signal are sampled and digitized with a 10-bit Analog Digital Converter (ADC), for example. The exemplary 10-bit ADC is contained in the device 110-112 micro controller, which integrates the sampled data and feeds it to the device 110-112 core processor. [0050] The device 110-112 measures, for example, five acceleration vectors per second for the three dimensions of possible movement and the EKG signal is digitized at a minimum rate, for example, of 200 samples per second. The number of acceleration vectors and sample rate are adjustable parameters. The acceleration vectors and EKG data are sent via a wireless IEEE 802.15.4 link to the collection analyzer server 120. The acceleration vectors and EKG data are signal averaged using weighted and/or not-weighted dynamically sized moving average convolution filters, and can be used to determine distances traversed. Further analytics can be performed by the collector analyzer server 120 to determine various states of motion such as walking, running, climbing (such as stairs), static, rollover, free-fall, impact, shaking, complex linear and angular motion "groups." The various state determinations are used as input to calculate the differential acceleration time derivatives ([d(A_x)/dt]² + [d(A_y)/dt]² + [d(A_z)/dt]²) algorithm contained within collection analyzer server 120. The time derivatives can be used to calculate the energy expenditure of the wearer of the device 110-112 by calculating d(Sχyz)/dt = (([d(A_x)/dt]² + [d(A_y)/dt]² + [d(A_z)/dt]²))^'Λ, where S_xyz represents a scalar quantity that describes an aggregate 3-D acceleration variation which implies varying degrees of motion (e.g, running versus walking or jumping). The collector analyzer server 120 also functions as a large data collection node.

[0051] Figure 2 is a plot of the device's 110-112 reported time-series acceleration vector data as processed by the collection analyzer server 120. These time-series plots can be archived for further analysis such as profiling, event capture, group correlation of events, and data mining as determined by the computer application and implementation purpose. The Figure 2 plot indicates a fall event (the large signal spike) with the vertical axis depicting acceleration in acceleration of gravity units (g = 9.8 meters/sec²) (128 units on y- axis = 0 g, 255 = +1.5 g, 0 = -1.5 g). [0052] Figure 3 is a plot of the device 110-112 reported time-series EKG trace data that was captured simultaneously while the device 110-112 reported time-series acceleration vector data in Figure 2. Figure 3 is the captured EKG trace data as processed by the collection analyzer server 120. These time-series plots are archived for further analysis such as profiling, event capture, group correlation of events, and data mining as determined by the computer application and implementation. The EKG signal is segmented in various measurements, which measure the time base length of different signals, for example, the Q point (signal start) to the T point (signal end) in milliseconds.

[0053] Figure 4 is a plot of the device 110-112 reported time-series EKG trace data that was captured simultaneously while the device 110-112 reported time-series acceleration vector data. Figure 4 is the captured EKG trace data as processed by the collection analyzer server 120. The EKG trace data is a recording of surface potentials due to the electrical activity associated with myocardium (cardiac muscle) events. It is a non-invasive technique that is used to measure of cardiac function. Specific waveforms within the EKG trace data recording represent the electrical activity associated with cardiac mechanical events such as ventricular contraction (depolarization) and relaxation (redepolarization). Figure 4 depicts some of the initial EKG signal processing performed by the Wireless Mobility EKG collector analyzer server. [0054] These bioelectrical impulses (signals) are very small in amplitude, on the order of one thousand of a volt (one mV), and a high-gain instrumentation operational amplifier with very-low signal interference rejection capability, such as a Texas Instruments model no. INA321EA or the like, is used to capture the EKG signal. Once the EKG signal is amplified and analog-filtered, the signal is converted to a digital signal which is further processed using digital signal processing (DSP) algorithms, such as FIR, windowed-FER. or CLS FIR, to apply a range of filtering options to remove unwanted signal noise artifacts. The DSP algorithms are based on a general class of DSP algorithms, such as those found in Digital Signal Processing, A Practical Guide for Engineers and Scientists, by S. W. Smith, the contents of which are incorporated herein by reference.

[0055] An EKG receives signals from a plurality of electrodes (3, 5, and 12 connections are common) that are placed on a user's skin surface. The EKG monitors voltage signals appearing between selected pairs of the electrodes and performs a temporal vector analysis of the resultant signal pairs to prepare selected two-dimensional voltage-time graphs indicative of internal cardiac activity.

[0056] Surface EKG signals as sensed by a plurality of electrodes, in a plurality of lead configurations. Surface EKG monitor electrodes are designated as the center (C) electrode, the left arm (LA) electrode, the right arm (RA) electrode, the left leg (LL) electrode, and the right leg (RL) electrode. These EKG electrodes sense and conduct electrical signals from the user's skin surface indicative of the cardiac activity. Lead I is defined as the RA signal as the (negative)-Input and the LA signal as the (positive)+Input. Lead II is defined as the RA signal as the -Input and the LL signal as the +Input. Lead III is defined as the LA signal as the -Input and the LL signal as the +Input. An authoritative description of EKG lead placement and interpretation can be found in Rapid Interpretation of EKG's, 6th Edition, (2000) by Dale Dubin, the contents of which are incorporated herein by reference. [0057] An aVR lead is defined as the (LA+LL)/2 signal as the -Input and the RA signal as the +Input. The aVL lead is defined as the (RA+LL)/2 signal as the -Input and the LA signal as the +Input. The aVF lead is defined as the (RA+LA)/2 signal as the -Input and the LL signal as the +Input. The V lead is defined as the (RA+LA+LL)/3 signal as the -Input and the C signal as the +Input. EKG lead configurations and inputs are well-known, and the disclosed system can be configured to accommodate any of the known configurations.

[0058] The EKG system in an exemplary configuration uses a single channel differential instrumentation operational amplifier analog front end, such as a Texas Instruments model no. INA321EA, that is capable of recording a range of EKG signals which include Lead I, Lead II or Lead III type of EKG traces. These EKG signals are filtered, and then digitized with a high-speed analog-to-digital converter (ADC). The digitized EKG signals and 5 dimensional acceleration data for generated by the user meaning the device are 110-112 simultaneously wirelessly transmitted to the collector analyzer server 120 for further digital signal processed (DSP) to further remove signal noise artifacts, and to determine fundamental EKG wave shape signatures.

[0059] The collector analyzer outputs reduced noise digitized EKG bio-potentials that are archived to form EKG recordings. These traces are now EKG traces that are correlated with three dimensional motion artifacts. The motion artifacts are acceleration vectors that directly reflect the device 110-112 user's instantaneous energy output and energy output integrated over any time period. This information is used to determine the possible cause of cardiac events and trends. [0060] The Wireless Mobility EKG system comprises a clinically accurate autonomous capture event monitor. It is designed to capture based on rate, rhythm, QRS signal profiling, and P-wave analysis for the detection of both symptomatic and asymptomatic arrhythmias. The Wireless Mobility EKG system P-wave analysis process has the ability to autonomously capture atrial arrhythmias. The Wireless Mobility EKG system back-end processing system preferably utilizes algorithms with artificial intelligence that analyze the user's EKG in real time for the entire procedure.

[0061] The Wireless Mobility EKG system preferably automatically detects and reports the onset, amplitude and intervals of the PQRST complex, either online or offline. The raw data is unaltered by any calculations, ensuring that the EKG can be re-analyzed at any time. The Wireless Mobility EKG system preferably allows for EKG heartbeat classification which allows for the selection of EKG waveforms for analysis. With a two-paned window, the user can identify any noise-contaminated waveforms, or those with abnormal QRS shapes or RR intervals. These can be removed from analysis, if desired. Valid EKG heartbeats are preferably automatically displayed in the EKG Averaging View, with indicators, for example, indicating P Start, P Peak, QRS Start, QRS End, T Peak and T End. The time, amplitudes and intervals are inserted into the EKG Table save store. [0062] The Wireless Mobility EKG system preferably uses digital signal processing techniques for signal filtering and analysis. The system's EKG signal morphological processes form the basis for cleaning up noise after background subtraction and threshold adjustment. The EKG signal morphological techniques include EKG signal dilation to expand the signal morphology to enhance detection and analysis of PR segment (atria depolarization), QRS duration (ventricle depolarization), QT/QTc measure (activation/recovery of the ventricular muscle), and ST segment (ventricle red polarization) signal signatures. The EKG signal morphological processes preferably utilize multi-layered neural networks to allow for the "personalization" of the EKG signal traces on a per- individual user level. The EKG signal traces vary greatly across a user base in their detail but provide a great deal of information on a specific user level, once the EKG signal processing algorithms utilizing multi-layered neural networks are "trained." [0063] The Wireless Mobility EKG system leverages advance DSP algorithms in conjunction with a unique multi-layer neural network that embodies a parts-based representation of the inputs. This allows every neuron in the multi-layer neural network to represent a part of the solution. This allows manipulating the individual elements of the multi-layer neural network to achieve much higher throughputs. The multi-layer neural network is an adaptive model that learns from the data and generalizes the features learned. It extracts the numerical characteristics from the numerical data instead of memorizing all of it.

[0064] The Wireless Mobility EKG system's multi-layer neural network will be given initial setup parameters and a large initial training EKG data set so that it can be taught to differentiate one kind of PQRST signal profile pattern from another. The Wireless Mobility EKG system's multi-layer neural network embodies a parts- based representation of the inputs where every neuron in the network represents a part of the solution. For an example, in the case of object recognition, the requirement of the multi-layer neural network is to detect if a given image has a car in it.

[0065] A traditional neural network can be trained to recognize cars by supplying it hundreds or thousands of examples of car and non-car images. Then, given an image that it has never seen before, it will classify an image as containing a car or non-car. [0066] The Wireless Mobility EKG system's multi-layer neural network is similarly trained in the same way, except that in the process of its training, it develops an internal representation of cars. This is where one neuron will correspond to a headlight, one to the rear door, one to the front fender, and so on. This internal representation is not dictated by the user, but falls out naturally from the properties of the multi-layer neural network itself. If the multi-layer neural network is not working as desired by the user, the user can change the performance of the multi-layer neural network in a predictable, understandable way such as paying more attention to windows and less to headlights. This can be done by increasing the weights of the neurons that represent those elements of the input, for example.

[0067] The Wireless Mobility EKG system's multi-layer neural network is not a black box. It is now clear how the internal elements are transforming the inputs to the outputs, and they can be manipulated appropriately. They can now be considered data rich and theory rich. The internal elements of the neural network have the same applications as before, except they can be "pre-loaded" with known facts about input-output relationships. Other than the output result of the multi-layer neural network itself, examining the internal elements is also a new result, because it reveals the nature of the inter-relationships of the inputs and the relationship of the inputs to the output. This allows for feeding any new relationships that are revealed as the multi-layer neural network start to yield results back into the multi-layer neural network further improving its performance.

[0068] An additional feature of the parts-based representation in the multi-layer neural network is that it allows creating a database of the PQRST signal profile patterns in a parts-centric way, for instance, store the data using an alphabet where every letter of the alphabet corresponds to a part of the input. This allows performing searches in a database in the same way as traditional text searches, only the results are delivered on the basis of the parts-based patterns (e.g., PQRST signal profiles), thus becoming an extremely effective pattern matching tool. This will allow for detecting any number of different arrhythmias signal profiles, and have the capability to extract abnormalities of the heart's EKG signal profiles. At the start, the Wireless Mobility EKG system's multi-layer neural network establishes a normal pattern for the user and continually adapts this pattern in a self-learning mode throughout the recording session as the EKG signal profiles change.

[0069] The Wireless Mobility EKG system's multi-layer neural network analyzes the EKG signal profile data both prospectively and retrospectively to identify, for example, arrhythmias and ischemia conditions. EKG monitoring is the only practical way to detect recurrent or transient ischemic events over time. These cardiac ischemia (reduced coronary blood flow) events can cause the user to have chest pains, referred to as angina, which is indicated with a transient T-wave inversion. The T-wave inversion is captured immediately by the Wireless Mobility EKG system. Specific analysis features supported by the Wireless Mobility EKG system are arrhythmia analysis, ST segment analysis, and heart rate variability analysis and the like.

[0070] The Wireless Mobility EKG system can detect and capture the asymptomatic changes in the ST-segment signal profile in a user with coronary artery disease, which is a specific indication that the user is experiencing transient myocardial ischemia. Myocardial ischemic events detected early via ST-segment analysis benefits the user greatly since this can result in immediate treatment of ischemia. Current EKG unit implementations have technical problems with noise levels, and lack adequate accurate ST-segment analysis techniques, such as the lack of clarity about how information about changes in the ST-segment signal profile, especially in asymptomatic users, indicate various medical conditions. [0071] Unstable angina and ST-elevation or non-ST-elevation myocardial infarction most often occur in users with significant coronary artery disease, who have disruption of an atherosclerotic plaque and a subsequent dynamic coronary occlusive process that involves cycles of plaque rupture, platelet stimulation, coronary vasospasm, and thrombus formation. The ST-segment analysis performed by the system, which provides uninterrupted real-time information about the occurrence, frequency, and severity of ischemic episodes over the course of the dynamic occlusive process, and is one of the critical features of the Wireless Mobility EKG system. In addition to unstable angina and acute myocardial infarction, a number of other acute situations occur in which the ST-segment analysis is important. [0072] The Wireless Mobility EKG system uses normal EKG signal profiles and baselines data as a benchmark for analyzing the user's EKG data in real-time. For example, nine parameters can be calculated from the EKG data and sixteen context parameters can be calculated from a 0.75-second window of EKG data surrounding the QRS signal complex. Of course, more or less parameters can be used for each type of parameter. The Wireless Mobility EKG system uses a dynamic window- based EKG signal classifier that is a trainable unit via its input/target pairs. The input side of the EKG signal classifier performs normalization for the signal pattern. The core of the EKG signal classifier preferably performs feed-forward and error back-propagation calculations. [0073] The EKG signal classifier's feed- forward calculations are used both in training mode and in operation. The EKG signal classifier's error back-propagation calculations are preferably used in training mode only. The EKG signal classifier has a transfer function that is a sigmoid described, for example, as 1/(1 + exp (-input)), which is fed the summation of all previous layer outputs multiplied by their interconnecting weights. When the EKG signal classifier is in training mode, the feed-forward output state calculation is combined with error back-propagation and weight adjustment calculations that reflect the multi-layer neural networks learning/training. [0074] The EKG signal classifier will generate an error that is preferably a function of the difference between an output value and a target value for each signal pattern. The objective is to minimize this error by propagating this error back through the neural network and appropriately adjust the weights. This can be done on-line in a single signal pattern-training mode or epoch (batch) mode, for example. The Wireless Mobility EKG system preferably displays the online or offline analyses as graphical plots.

[0075] Scatter plots of QT versus RR, QT versus time and RR versus time, provide the user with a method for viewing changes in cardiac function. EKG trace recordings can contain regions of signal noise generated by electrical interference such as 50/60 cycle AC power lines induced noise. After EKG heartbeats are detected, the Wireless Mobility EKG system's window for the EKG heartbeat analyzer displays heartbeats according to activity and signal noise. Large activity values indicate high frequency noise, whereas large signal noise values correspond to either T- wave contamination or isoelectric line (baseline) wander. The Wireless Mobility EKG system utilizes several window-based moving average convolution filters for minimizing the effects-induced high frequency noise. [0076] The Wireless Mobility EKG system's window for the EKG heartbeat analyzer allows the user to detect and identify changes in the shape of the QRS complex, ST segment shape and the RR interval. For example, heartbeats such as premature ventricular contractions (PVC) have a smaller form factor than normal beats. Apart from reducing the effects of high-frequency noise, 50/60 cycle power line interference and movement artifacts, averaging is useful when comparing effects on the EKG waveform before and after mobility/activity intervention. [0077] hi the EKG heartbeat averaging mode, the user can preview each average individually and manually adjust PQRST indicators, if desired. The Wireless Mobility EKG system checks for signal quality, EKG electrode placement, cable integrity, and QRS signal profile on EKG electrode attachment. The interactive display provides corrective action when necessary. The Wireless Mobility EKG system also confirms that the algorithm is running each time the user changes EKG electrodes. PQRST signal profile analysis allows the Wireless Mobility EKG system to differentiate electronic noise and muscular artifact from real cardiac events. [0078] The Wireless Mobility EKG system's event monitor can be designed to autonomously capture heart rate, rhythm and PQRST signal profiles that include both symptomatic and asymptomatic cardiac event detection and capture. This feature makes it an ideal monitor for a wide range of populations such as geriatric, diabetic and pediatric class of users. The Wireless Mobility EKG system allows for capturing an elusive life-threatening cardiac event. The Wireless Mobility EKG system monitors and analyzes the user's EKG for the entire procedure. [0079] The Wireless Mobility EKG system facilities accurate electrode attachment at the baseline and provides the user with verification whenever they change EKG electrodes. The collector analyzer part of the system stores EKG data and mobility data on a broadband-connected PC or similar device attached to it. When the device, which is attached to the user, comes within wireless range of the collector analyzer server, the EKG data and mobility data wirelessly transmitted using AES 128-bit encryption, for example, is processed and archived in real-time on the collector analyzer server. When the user is out of wireless range of the collector analyzer server, the device attached to the user commences recording or storing the EKG data and the motion data in its own memory or memory device without any wireless transmissions until it detects that it is within wireless range again with its securely authenticated collector analyzer server partner. [0080] Figure 5 is an exemplary time domain plot showing the distance traversed by the user wearing the device, which is sending three dimensional acceleration data (Ax, Ay, Az) at a preferred minimum rate of five times a second (an adjustable parameter) to the collector analyzer server, which calculates the distance traversed preferably using normalized position vectors. The collector analyzer server system preferably performs three dimensional double integrations a minimum rate of five times a second (an adjustable parameter) where the Path (x,y,z,t) = ∑JJ A_xdt + ∑IJAydt + jjA_zdt, and each integration result is summed and accumulated over the entire observation and monitoring period to provide location data as it relates to the device and its user. In the exemplary time domain plot of Figure 5, two of the dimensions are plotted since the device user only moved in a two dimensional plane (x and y and z = 0, indicating no height change up or down, for example going up/down stairs, etc.). [0081] The collector analyzer server can generate alarms and alerts based on predetermined rules and the type of application used through a securely attached internet-enabled PC or similar device. These alarms and alerts are incidents which are dispatched to individuals identified as responders (neighbors, friends/family, emergency service providers such as local community police, fire or ambulance) and medical managed service providers (e.g., doctors, nurses, doctors' offices, insurance offices).

[0082] Inactivity concerns can be monitored based on the collector analyzer server's predetermined template-based software rules. If there is excessive inactivity detected within a selected time period, notification will be sent to the medical managed service provider and the appropriate alarms and alerts will be generated via predefined personalized call-lists. The collector analyzer server can activate commands (rule sets) for desired function as a result of excessive inactivity events. [0083] For data reliability, the system uses the wireless IEEE 802.15.4 ZigBee mesh network technology standard for protection against failure. By placing the wireless IEEE 802.15.4 ZigBee receivers and transmitters in groups, the mesh network that results provides redundant paths to ensure alternate data path routes exist and there is no signal point of failure should a node fail. Wireless IEEE 802.15.4 ZigBee routers having extra specialized software running in the node are used to greatly extend the range of the network by acting as relays for nodes that are to far apart to communicate directly. The wireless IEEE 802.15.4 ZigBee technology standard is preferred for communication between the device and the collector analyzer server. Of course, other technology standards can be used. [0084] The wireless data communications preferably implement a 128-bit

Advanced Encryption Standard (AES) algorithm for encryption and incorporate the strong security contained within IEEE 802.15.4. The security services implemented preferably include methods for key establishment and transport, device management and frame protection. The system leverages the security concept of a "Trust Center." The "Trust Center" allows the system's node devices to access the network, distribute keys and enable end-to-end security between the device and the collector analyzer server. Device Hardware

[0085] The device uses an IEEE 802.15.4 compliant 2.4 GHz Industrial, Scientific, and Medical (ISM) band Radio Frequency (RF) transceiver. It contains a complete 802.15.4 Physical layer (PHY) modem designed for the IEEE 802.15.4 wireless standard which supports peer-to-peer, star, and mesh networking. It is combined with a MPU to create the wireless RF data link and network. The IEEE 802.15.4 transceiver supports 250 kbps OQPSK data in 5.0 MHz channels and full spread-spectrum encode and decode. [0086] All control, reading of status, writing of data, and reading of data is done through the RF transceiver interface port. The device MPU accesses the device RF transceiver through interface "transactions" in which multiple bursts of byte-long data are transmitted on the interface bus. Each transaction is preferably three or more bursts long depending on the transaction type. Of course, shorter can be used. Transactions are read accesses or write accesses to register addresses. The associated data for single register access can be 16 bits in length. Of course, greater or lesser bit lengths can be used.

[0087] Receive mode is a state where the device RF transceiver is waiting for an incoming data frame. The packet receive mode allows the device RF transceiver to preferably receive a whole packet without intervention from the device MPU. The entire packet payload is preferably stored in RX Packet RAM, and the micro controller fetches the data after determining the bit length and validity of the RX packet.

[0088] The device RF transceiver waits for a preamble followed by a Start of

Frame Delimiter. From there, the Frame Length Indicator is used to determine length of the frame and calculate the Cycle Redundancy Check (CRC) sequence. After a frame is received, the device application determines the validity of the packet. Due to noise, it is possible for an invalid packet to be reported with either of the following conditions: A valid CRC and a frame length (0, 1, or 2) and/or invalid CRC/invalid frame length. [0089] The device application software determines if the packet CRC is valid, and that the packet frame length is valid with a value of 3 or greater. Of course, this value threshold can be greater than or less than 3. In response to an interrupt request from the device RF transceiver, the device MPU preferably determines the validity of the frame by reading and checking valid frame length and CRC data. The receive Packet RAM register is accessed when the device RF transceiver is read for data transfer.

[0090] The device RF transceiver preferably transmits entire packets without intervention from the device MPU. The entire packet payload is preferably preloaded in TX Packet RAM, the device RF transceiver transmits the frame, and then the transmit complete status is set for the device MPU. When the packet is successfully transmitted, the transmit interrupt routine that runs on the device MPU reports the completion of packet transmission. In response to the interrupt request from the device RF transceiver, the device MPU reads the status to clear the interrupt, and check for successful transmission.

[0091] Control of the device RF transceiver and data transfers are preferably accomplished by means of a Serial Peripheral Interface (SPI). Although the normal SPI protocol is based on 8-bit transfers, the device RF transceiver imposes a higher level transaction protocol that is based on multiple 8-bit transfers per transaction. A singular SPI read or write transaction preferably comprises an 8-bit header transfer followed by two 8-bit data transfers. The header denotes access type and register address. The following bytes are read or write data. The SPI also supports recursive 'data burst' transactions in which additional data transfers can occur. The recursive mode is intended for Packet RAM access and fast configuration of the device RF transceiver.

Wireless Mobility EKG device Software

[0092] The software architecture for the Wireless Mobility EKG device's MPU uses an interrupt-driven architecture. The interrupt routines preferably include the reading of the ADC (Analog Digital Converter), timers for creating the sampling frequency and handling interrupts from the IEEE 802.15.4 RF Transceiver. Non- interrupt routines run on the device's MPU are system initializations and the wireless communications to the collection analyzer server are shown in Figure 7. [0093] There a number of interrupt handlers that process data asynchronously from the non-interrupt main loop routine previously described. The first interrupt is the Timer interrupt routine, which is used as a time base, and generates the sampling rate frequency used by the ADC. The second is the ADC interrupt routine, which occurs when the ADC conversion of the three acceleration vectors (A_x, A_y, A_z) and the EKG data is complete. It formats the ADC readings for read by the non- interrupt main processing loop. The third is the device's RF transceiver status and data transfers interrupt handler. [0094] This routine is used to process the device's RF transceiver events, transmit acceleration (A_x, A_y, A_z) vectors, EKG and link energy data via the device RF transceiver to the collector analyzer server, and receive control/acknowledgement data via the device RF transceiver from the collector analyzer server. Of course, other suitable interrupt handlers can be used. [0095] Figure 8 illustrates an exemplary block diagram of the described interrupt handlers.

[0096] Figure 9 is an exemplary sequence diagram of successful transmission of acceleration (A_x, A_y, A_z) vectors and EKG data from the device to the collector analyzer server. [0097] The collector analyzer server software is preferably a multithreaded Java- based server that handles one or more device communications channels for data gathering/control and secure Internet communications with a medical managed service provider. The Java language was chosen to provide the broadest base of support for the collector analyzer server hardware platform. Figure 10 illustrates exemplary internal subsystems of the collector analyzer server.

[0098] The Wireless Mobility EKG collector analyzer server preferably collects device three dimensional acceleration (A_x, A_y, A_z) vectors, EKG with the signal strength (Link energy) data associated with the wireless communications channel between the device and the collector analyzer server. The device three dimensional acceleration vector data, which is preferably sampled a minimum of five times a second (this is an adjustable parameter) for each dimension, reflects the motion dynamics experienced by the user of the device in real-time.

[0099] Once receiving the device three dimensional acceleration data, the collector analyzer server preferably performs normalization functions on the acceleration data to remove zero gravity (g) offsets. Next, the collector analyzer server can apply several signal averaging and finite impulse response (FIR) filtering algorithms to the acceleration data for smoothing and signal noise reduction. This processed acceleration data represents a time-series of dynamic events which are reordered and analyzed for standard linear motion (such as walking, running, repetitive movement), fall detection, shaking, and tremor events or the like. [0100] The collector analyzer server preferably has numerous differential acceleration templates ([d(A_x)/dt]² + [d(A_y)/dt]² + [d(A_z)/dt]²) in memory that profile the changes in acceleration data that exist when standard linear motion (such as walking, running, repetitive movement), falls, shaking, and/or tremor events occur. The differential acceleration time derivatives are preferably generated by the collector analyzer server's algorithm which uses as input, the real-time acceleration data sent by the device. These templates are used to correlate the real-time acceleration data from the device with known events such as standard linear motion (such as walking, running, repetitive movement), falls, shaking, and/or tremor events contained in the differential acceleration templates. When the collector analyzer server detects a fall (or any other significant event), it immediately notifies all persons and services on a preprogrammed call list for this user wearing this device. The templates are preferably personalized or customized to a particular user, who wears the device, via profiling data and the neural network. [0101] The collector analyzer server preferably archives data locally and at the medical managed service provider, when necessary. When analyzing specific situations such as disease progression, data needs to be archived for data mining purposes, and in this case, may require the additional storage of a medical managed service provider. It is preferably that the amount of data be quite large. The collector analyzer server can correlate events such as standard linear motion (such as walking, running, repetitive movement), falls, shaking, and/or tremors with preprogrammed schedules of medication, exercise, and other bodily events. [0102] The collector analyzer server is preferably designed with layered software architecture that supports multithreading for concurrent processing of device requirements, real-time data analysis, event processing, and medical managed service provider communication. The collector analyzer server preferably runs on a Java Virtual Machine (JVM) architecture so as to support a broad range of computing platforms. Of course, other architectures may be used.

[0103] The collector analyzer server software preferably uses a default finite impulse response (FIR) filter that is implemented using a eleventh-order moving average convolution filter whereby the filter coefficients are found via: B(i) = 1/(P + 1) for i = 0, 1, 2 P

Where P = 10 for creating the eleventh-order filter. The impulse response for the resulting filter is:

h(n) = δ(n)/l 1 + δ(n - 1)/11 + δ(n - 2)/l 1 + δ(n - 3)/l 1 + δ(n - 4)/l 1

+ δ(n - 5)/l 1 + δ(n - 6)/l 1 + δ(n - I)Il 1 + δ(n - 8)/l 1 + δ(n - 9)/l 1

+ δ(n - 10)/l l + δ(n - l l)/l l

[0104] Figure 11 illustrates an exemplary block diagram of this eleventh-order filter.

[0105] The collector analyzer server software also preferably uses a dynamic sized

(ordered) finite impulse response (FIR) filters based on profiling requirements that are implemented using n^th-order moving average convolution filters whereby the filter coefficients are found via:

B(i) = 1/(P + 1) for i = 0, 1, 2, .... P

Where P = n - 1 for creating the n^th-order filter. The impulse response for the resulting filter is:

h(t) = δ(t)/n + δ(t - l)/n + δ(t - 2)/n + . . . . +δ(t - n )/n

[0106] Figure 12 illustrates an exemplary block diagram of this n^th-order filter. [0107] The moving average convolution filter size is a function of the application that would run above the collector analyzer server software layer. The application could be for athletic sports mobility/EKG profiling, or pre-testing youngsters before they perform stressful field and track gym activities, or monitoring elder's mobility/EKG during normal daily activities for analyzing disease progression, or a monitor anyone's mobility/EKG for correlating against medication schedules/dosing. These applications have their own specialized requirements based on mobility dynamics/EKG data to monitored and profiled. [0108] The collector analyzer server preferably uses a EKG signal classifier that has a transfer function that is preferably a sigmoid described as 1/ (1 + exp (-input)). The collector analyzer server preferably uses a back-propagation gradient descent algorithm. The collector analyzer server uses as a secondary technique, for example, the Widrow-Hoff learning rule, where the network weights are moved along the negative of the gradient of the performance function. The collector analyzer server preferably uses a conjugate gradient and Newton methods for standardized optimization. The collector analyzer server's EKG signal classifier allows for "tuning" to the user's EKG signal profile/morphology, in essence, creating a "personalized" or "customized" EKG signal classifier.

[0109] The Wireless Mobility EKG system can also use multi-band filters, which are program selectable such as Butterworth, Chebyshev, elliptic, Yule- Walker, window-based, least squares, and Parks-McClellan (real and complex) type filters. The filter structures supported are preferably the direct forms I and II, lattice, lattice- ladder, and second-order sections. Additional work will be done to explore the use of multi-rate EKG signal processing, which can include decimation, up- and down- sampling, re-sampling, and spline interpolation. The Parks-McClellan algorithm, which uses a Remez exchange algorithm and Chebyshev approximation theory allows for designing digital filters with optimal fits between the desired and actual frequency responses. These digital filters designs (called min-max digital filters) become optimal because they minimize the maximum error between the desired frequency response and the actual frequency response. [0110] The EKG digital signal processing core preferably uses zero-phase digital filtering by processing the input data in the forward direction. The EKG digital signal processing core implements zero-phase digital filters that process the input data in both the forward and reverse directions. After filtering in the forward direction, the EKG digital signal processing core's zero-phase digital filter reverses the filtered sequence and runs it back through the filter. The resulting sequence has zero-phase distortion and double the filter order. The EKG digital signal processing core algorithm preferably minimizes the start-up and ending transients by matching initial conditions, and works for both real and complex inputs. [0111] The EKG digital signal processing core reduces ringing and ripples occurring in the signal response near the band edge (i.e., "Gibbs effect") by utilizing a nonrectangular window, which can reduce the ringing and ripples signal magnitude. A multiplication by a window in the time domain causes a convolution or smoothing in the frequency domain. A primary adjustable Hamming window digital filter can be implemented to reduce the ringing and ripples signal magnitude. The EKG digital signal processing core preferably uses secondary Kaiser window digital filters that utilize various filter orders, cutoff frequencies, and Kaiser window beta parameters. The EKG digital signal processing core determines a vector class of frequency band edges and a corresponding vector class of magnitudes, as well as maximum allowable ripple, and uses this to determine the appropriate input parameters for dynamically creating the windowed Kaiser digital filters. [0112] The EKG digital signal processing core leverages constrained least squares (CLS) finite impulse response (FIR) digital filter algorithms for creating additional FIR digital filters without explicitly defining the transition bands for the magnitude response. The ability to omit the specification of transition bands is used in a situation where it is not clear where a rigidly defined transition band will appear if noise and signal information appear together in the same frequency band. [0113] For the EKG digital signal processing core's CLS FIRs implementation, omitting the specification of transition bands is possible if they appear to control the results of Gibbs (band edge) phenomena that appear in the FIR filter's response. Instead of defining pass-bands, stop-bands, and transition regions, the EKG digital signal processing core's CLS algorithm accepts a cutoff frequency for the high-pass, low-pass, band-pass, or band-stop cases for the desired response. The EKG digital signal processing core's CLS algorithm can also accept a cutoff frequency for pass- band and stop-band edge for multi-band cases to get a desired response. This is so that the EKG digital signal processing core's CLS algorithm preferably defines transition regions implicitly rather than explicitly. [0114] One of the features of the EKG digital signal processing core's CLS algorithm is that it enables defining upper and lower thresholds that contain the maximum allowable ripple in the magnitude response. Given this constraint, the EKG digital signal processing core's algorithm applies the least square error minimization technique over the frequency range of the FIR filter's response, instead of over specific bands. In the EKG signal classifier weight adjustments can be done on-line in a single signal pattern-training mode or epoch (batch) mode. The EKG signal classifier supports dynamic initialization weight adjustments to support the Wireless Mobility EKG system's "personalization" feature. [0115] The Wireless Mobility EKG system can support a television (TV) interface (when desired) for use with a standard TV set as a monitor display. This feature addresses the households where PCs are not available for the Wireless Mobility EKG system application. There are situations where the home PC hardware is used for other applications and is not available or does not exist, which is the case in many elders homes. TV sets are usually available especially older ones that could be used with the Wireless Mobility EKG system. The TV interface used by the Wireless Mobility EKG system preferably supports TV sets as old as 40 years. [0116] The optional Wireless Mobility EKG system TV interface generates all the aspects of the TV signal such as vertical sync/blanking, horizontal sync/blanking and the TV Line. A digital to analog converter (DAC) is used to mix the video and sync pulses. Using a resistor matrix, a 3.3-VDC voltage source and in conjunction with the TV's 75-ohm load produces sync and the white voltage levels of 0.3-volts to 1.0 volts. [0117] The foregoing description is of a preferred embodiment of the invention and has been presented for the purposes of illustration and description of the best mode of the invention currently known to the inventors. This description is not intended to be exhaustive or to limit the invention to the precise form, connections or choice of components disclosed. Obvious modifications or variations are possible and foreseeable in light of the above teachings. This embodiment of the invention was chosen and described to provide the best illustration of the principles of the invention and its practical application to thereby enable one of ordinary skill in the art to utilize the invention in various embodiments and with various modifications as are suited to the particular use contemplated by the inventors. All such modifications and variations are intended to be within the scope of the invention as determined by the appended claims when they are interpreted in accordance with the breadth to which they are fairly, legally and equitably entitled.

Claims

What Is Claimed Is:

1. A method and apparatus for wireless autonomous temporal-spatial mobility and electrocardiogram (EKG) simultaneous data capture for the purposes of critical and non-critical cardiac event detection, alarming, heuristic archiving, and correlation analysis comprising: (A) a Wireless Mobility EKG device hardware for wearing on the user's waistband or pants pocket or attached like a sports MP3 player with a Velcro strap to the arm measure and capture acceleration vectors that depict motion/mobility and EKG data simultaneously; (B) the Wireless Mobility EKG device sending acceleration vector and EKG data using a mesh-type wireless network to a wireless Wireless Mobility EKG collector analyzer server for data collection and further processing; (C) the Wireless Mobility EKG collector analyzer server signal averaging and temporally smoothing the collected Wireless Mobility EKG device acceleration vector and EKG data using dynamically sized moving average convolution filters and storing the results; (D) the Wireless Mobility EKG collector analyzer server using these results (from C) to create differential acceleration time derivatives ([d(A_x)/dt]² + [d(A_y)/dt]² + [d(A_z)/dt]²) data, which then is used to calculate the energy expenditure by the user of the Wireless Mobility EKG device by using a unique algorithm for calculating d(Sχyz)/dt = (([d(A_x)/dt]² + [d(A_y)/dt]² + [d(A_z)/dt]²))^'Λ and the user's work function which is proportional to Work(x,y,z,t) ~ M_h(A_x∑JjA_xdt + A_y∑|jA_ydt + AjJA_zdt; (E) the Wireless Mobility EKG collector analyzer server profiling and correlate the Wireless Mobility EKG device acceleration vector/EKG data (from D) with motion artifact analysis and multi-layer neural network EKG event signal analysis/classification using statistical parametric template profiles in memory and then measuring mobility/EKG event correlation; (F) the Wireless Mobility EKG collector analyzer server using these results (from E) to determine the path traversed by the Wireless Mobility EKG device wearer by calculating Path (x,y,z,t) = ∑JJA_xdt + ∑JjA_ydt + JjA_zdt, which implies an overall temporal-spatial work function (user energy expenditure) Work(x,y,z,t) ~ M_h(A_x∑IJA_xdt + A_y∑JJA_ydt + A_zj|A_zdt; (G) the Wireless Mobility EKG collector analyzer server performing mobility/EKG stress test recording, benchmarking, and archiving for physical conditioning; (H) the Wireless Mobility EKG collector analyzer server correlating the Wireless Mobility EKG device wearer's mobility/EKG event data with medication schedules/dosing; (J) the Wireless Mobility EKG collector analyzer server correlating the Wireless Mobility EKG device wearer's mobility/EKG event data with disease progression; (M) the Wireless Mobility EKG collector analyzer server archiving the stored Wireless Mobility EKG device wearer's result mobility/EKG event data with medical managed service provider for purposes of data mining that allow for profiling/correlating the Wireless Mobility EKG device wearer's result mobility/EKG event data with medication schedules/dosing and disease progression; (N) the Wireless Mobility EKG collector analyzer server profiling and correlating the spatial-temporal mobility/EKG event data of a user wearing the Wireless

Mobility EKG device for determining severity of alarm level and then generating those alarms for detected critical cardiac events.

2. A method for wireless autonomous temporal-spatial mobility and electrocardiogram (EKG) simultaneous data capture for the purposes of critical and non-critical cardiac event detection, alarming, heuristic archiving, and correlation analysis, as recited in claim 1, wherein said human autonomous temporal-spatial mobility and electrocardiogram (EKG) simultaneous data capture detection and analysis uses wireless smaller than cell phone size battery-powered portable device (Wireless Mobility EKG device) to send simultaneously acceleration motion vectors and EKG data.

3. A method for wireless autonomous temporal-spatial mobility and electrocardiogram (EKG) simultaneous data capture for the purposes of critical and non-critical cardiac event detection, alarming, heuristic archiving, and correlation analysis, as recited in claim 1, wherein said a collector analyzer server (Wireless Mobility EKG collector analyzer server) is used to collect Wireless Mobility EKG device acceleration motion data such as walking, running, climbing (such as stairs), static, rollover, free-fall, impact, shaking, complex linear and angular motion simultaneously with EKG data for archive storage and correlation/heuristic analysis.

4. A method for wireless autonomous temporal-spatial mobility and electrocardiogram (EKG) simultaneous data capture for the purposes of critical and non-critical cardiac event detection, alarming, heuristic archiving, and correlation analysis, as recited in claim 1 , wherein said a collector analyzer server (Wireless Mobility EKG collector analyzer server) is used to signal average and use dynamically sized moving average convolution filters as a function of profile template on collected Wireless Mobility EKG device acceleration motion vectors and EKG data.

5. A method for wireless autonomous temporal-spatial mobility and electrocardiogram (EKG) simultaneous data capture for the purposes of critical and non-critical cardiac event detection, alarming, heuristic archiving, and correlation analysis, as recited in claim 1, wherein said a Wireless Mobility EKG collector analyzer server is used to calculate, profile and correlate the differential acceleration time derivatives ([d(A_x)/dt]² + [d(A_y)/dt]² + [d(A_z)/dt]²) which then is used to calculate the energy expenditure by the wearer of the Wireless Mobility EKG device by using a unique algorithm for calculating d(Sχyz)/dt = (([d(A_x)/dt]² + [d(A_y)/dt]² + [d(A_z)/dt]²))^'/2 on data as recited in claim 4.

6. A method for wireless autonomous temporal-spatial mobility and electrocardiogram (EKG) simultaneous data capture for the purposes of critical and non-critical cardiac event detection, alarming, heuristic archiving, and correlation analysis, as recited in claim 1, wherein said a Wireless Mobility EKG collector analyzer server is used to profile and correlate the data, as recited in claim 5, to determine critical EKG events as a direct and/or indirect result of various states of motion such as walking, running, climbing (such as stairs), stressful static, rollover, free-fall, impact, shaking, complex linear and angular motion generated from the user wearing the Wireless Mobility EKG device.

7. A method for wireless autonomous temporal-spatial mobility and electrocardiogram (EKG) simultaneous data capture for the purposes of critical and non-critical cardiac event detection, alarming, heuristic archiving, and correlation analysis, as recited in claim 1, wherein said a Wireless Mobility EKG collector analyzer server is used, as recited in claim 6, to generate alarms/alerts to a medical managed service provider.

8. A method for wireless autonomous temporal-spatial mobility and electrocardiogram (EKG) simultaneous data capture for the purposes of critical and non-critical cardiac event detection, alarming, heuristic archiving, and correlation analysis, as recited in claim 1, wherein said a Wireless Mobility EKG collector analyzer server is used, as recited in claim 5, to profile and correlate the spatial- temporal mobility/EKG data of the user wearing the Wireless Mobility EKG device for the purposes of disease progression.

9. A method for wireless autonomous temporal-spatial mobility and electrocardiogram (EKG) simultaneous data capture for the purposes of critical and non-critical cardiac event detection, alarming, heuristic archiving, and correlation analysis, as recited in claim 1 , wherein said a Wireless Mobility EKG collector analyzer server is used, as recited in claim 5, to profile and correlate the spatial- temporal mobility/EKG data of the user wearing the Wireless Mobility EKG device for the purposes of medication schedules/dosing.

10. A method for wireless autonomous temporal-spatial mobility and electrocardiogram (EKG) simultaneous data capture for the purposes of critical and non-critical cardiac event detection, alarming, heuristic archiving, and correlation analysis, as recited in claim 1, wherein said a Wireless Mobility EKG collector analyzer server is used, as recited in claim 5, to profile and correlate the spatial- temporal mobility/EKG data of the user wearing the Wireless Mobility EKG device for performing mobility stress tests such as running, climbing (such as stairs), general athletic/sports exercise movements, and weight lifting and provide recording, benchmarking, and heuristic archiving.

11. A method for wireless autonomous temporal-spatial mobility and electrocardiogram (EKG) simultaneous data capture for the purposes of critical and non-critical cardiac event detection, alarming, heuristic archiving, and correlation analysis, as recited in claim 1, wherein said a Wireless Mobility EKG collector analyzer server is used, as recited in claim 4, to determine the path traversed by the Wireless Mobility EKG device user by calculating Path (x,y,z,t) = ∑|jA_xdt + ∑JJA_ydt + jjA_zdt and an overall work function (user energy expenditure) Work(x,y,z,t) ~ M_h(A_x∑I|A_xdt + A_yΣJlA_ydt + A_zIJA_zdt for purposes of mobility/EKG data analysis of the user wearing the Wireless Mobility EKG device.

12. A method for wireless autonomous temporal-spatial mobility and electrocardiogram (EKG) simultaneous data capture for the purposes of critical and non-critical cardiac event detection, alarming, heuristic archiving, and correlation analysis, as recited in claim 1 , wherein said a Wireless Mobility EKG collector analyzer server is used, as recited in claim 5, to profile and correlate the spatial- temporal mobility/EKG data of an user wearing the Wireless Mobility EKG device and which mobility/EKG data with the user's preset cardiac profile, dynamic initialization weight adjustments are made in single signal pattern-training mode or epoch (batch) mode to the Wireless Mobility EKG collector analyzer server's EKG signal classifier to support the Wireless Mobility EKG system's user "personalization" feature.

13. A method for wireless autonomous temporal-spatial mobility and electrocardiogram (EKG) simultaneous data capture for the purposes of critical and non-critical cardiac event detection, alarming, heuristic archiving, and correlation analysis, as recited in claim 1, wherein said the Wireless Mobility EKG system supports a television interface for using with a standard TV set as a monitor, which feature addresses the households where PCs are not available for the Wireless Mobility EKG system application and where the home PC hardware is used for other applications and is not available or does not exist, which is the still the case in many households, the Wireless Mobility EKG system can support TV sets manufactured as far back as 1965 and which are usually available for such use.

14. A method for the Wireless Mobility EKG system to help create a relatively consistent electrode site preparation process with the use of the immediate EKG electrode/site impedance measurement feedback to the user, the archiving of electrode, cable and preparation characterizations for future reference, the application of algorithms with automated EKG electrode/cable impedance testing hardware which allows for characterizing the Wireless Mobility EKG system and any other EKG monitoring and recording systems in use in the marketplace.

15. A method for wireless autonomous temporal-spatial mobility and electrocardiogram (EKG) simultaneous data capture for the purposes of critical and non-critical cardiac event detection, alarming, heuristic archiving, and correlation analysis, as recited in claim 1 , wherein said a Wireless Mobility EKG collector analyzer server is used, as recited in claim 5, to profile and correlate cardiac ST- segment monitoring from captured spatial-temporal mobility/EKG data of a user wearing the Wireless Mobility EKG device using the Wireless Mobility EKG collector analyzer server's EKG signal classifier which uses a multi-layer neural network that embodies a parts-based representation of the inputs, which allows every neuron in the multi-layer neural network to represents a part of the EKG signal identification/classification solution and incorporates an adaptive model that learns from the EKG signal data, generalizes things learned for the EKG data and extracts the EKG signal characteristics from the EKG signal data instead of archiving all the data.

16. An apparatus for wireless autonomous temporal-spatial mobility and electrocardiogram (EKG) simultaneous data capture for the purposes of critical and non-critical cardiac event detection, alarming, heuristic archiving, and correlation analysis, comprising: a wireless device for simultaneously measuring acceleration data and EKG data and transmitting the measured data; and a wireless collector analyzer for receiving and processing the measured data, wherein the collector analyzer will output the results of the processing to a storage device or user.