CARDIAC ARRYTHMIAS MONITORING

01/03/2025 1
Cardiac arrythmias monitoring
 Pervasive health care (PHC) is a new health care model that enables patient mobility,
continuous health monitoring, and timely detection of anomalies.
 A pervasive cardiac care (PCC) prototype that supports real-time indoor and outdoor
continuous cardiac arrhythmias monitoring.
 Heart disease ranks as the top threat to human health. Elderly people and middle-
aged person in long-term sub-health state are among the high risk population of
cardiac sudden death (CSD).
 Clinical experiences prove that the most efficient way to prevent CSD is to detect the
cardiac abnormalities in time by making long term regular cardiac monitoring.
 The PCC system provides a real-time continuous cardiac monitoring service for
different subjects with no constraints of time or space.

01/03/2025 2
PCC System Architecture
 The PCC system integrates the advanced technologies of wireless communication,
embedded system, automatic ECG diagnosis (AED), and telemedicine, which
consists of four main functional components:
 Wireless ECG sensor (WES), Local access server, Remote access server, and
Remote surveillance terminal.
 Figure shows the architecture and operation fashions of the PCC system.

01/03/2025 3
01/03/2025 4
PCC System Architecture-Wireless ECG sensor
 Wireless ECG sensor (WES) conforms to the basic characteristics of wireless sensor
device: tiny resource, tiny power-consumption, and associated tiny cost.
 Furthermore, the basic embedded software such as a distributed real-time fault
tolerant microkernel, dedicated hardware and firmware and a TCP/IP protocol stack
are implemented and ported into WES.
 The WES prototype is a real-time wireless embedded portable sensor based on TI
MSP430 microcontroller. The key features of WES are
 Gain: 1000
 CMMR (min): 120 dB
 Bandwidth: 0.05 Hz to 125 Hz
 Programmable sample frequency more than 500 Hz
 Analogue to digital converter (ADC): 12 bit
 Leakage current: 10 μA

01/03/2025 5
PCC System Architecture-Wireless ECG sensor
 WES enables the capture of 4-leads ECG signals sampled at 500 Hz (sample
frequency) in real-time. Note that the sample frequency is reprogrammable (100 Hz
to 2000 Hz).
 The sample data is sent to the local server over a wireless medium such as Wi-Fi or
Bluetooth. In offline mode, ECG signals can be stored in the flash memory card.
 The signal store duration depends on the flash card capacity, the sample frequency,
and the number of ECG leads.

01/03/2025 6
PCC System Architecture-Local access server
 Local access server may be implemented on a standard PC, a PDA, a mobile phone,
or a dedicated network access device.
 It provides two kinds of network connection services: connection with WES via a
local wireless medium (Wi-Fi or Bluetooth) and connection with remote system via
infrastructure network, such as cable modem, ADSL, 2G/3G, etc.
 In view of the difference of network mediums, local servers, and transmission
speeds caused by network traffics, the PCC system provides an adaptable
communication mechanism to guarantee a reliable data transmission service
between local server and remote server.
 Furthermore, in order to provide a real-time data transmission, it is important to
minimize the amount of transmission data to reduce network traffics over low-
bandwidth connection links. A lossless ECG signals compression algorithm is thus
implemented.

01/03/2025 7
PCC System Architecture-Local access server
 If the local access server is a high-performance machine, the ECG diagnostic
module can be migrated from WES into the local server and the cost of WES will
be reduced.
 In addition, patients’ video information is necessary as an aided method for online
diagnosis. Hence, a ”webcam” can be installed in the local server to provide
patients’ images.

01/03/2025 8
PCC System Architecture-Remote access server
 The remote access server provides the capability of network connection between
local access server and remote surveillance server. In view of network types, it can
support two types of access servers: PPP server and WAP server.
 The PPP server establishes connections between patients and cardiologists via
PSTN, while the WAP server establishes the connections via wide wireless network.
 In addition, if the patients and the cardiologists are located in a same area (e.g., in
hospital or at clinic) and share local network infrastructures (e.g., high-speed LAN),
or they utilize the global network connection services provided by commercial ISPs
(Internet service providers), this component is no more necessary and can be
removed from the PCC system architecture.

01/03/2025 9
PCC System Architecture-Remote access server
 Several function modules can be loaded in the remote access server, including
database service and Web service. The medical history records of patients are
important for the diagnosis of heart diseases.
 In the remote access server, the medical records with multimedia formats are stored
in the patient database system, which include the ECG signals sequences,
cardiologists’ diagnostic reports, patients’ video, voice and individual profile
information, etc. The Web service provides the capacity of Internet access to the
patients’ database system.

01/03/2025 10
PCC System Architecture-Remote surveillance server
 The remote surveillance server has an interactive visualization Graphical User
Interface(GUI) which enables cardiologists to diagnose cardiac arrhythmias events in
real-time and to respond to alarm messages by monitoring the ECG signals sequence
and patients’ images.
 This server supports multi-patients surveillances and one patient on-line diagnosis (by
cardiologists) at one time.
 The 4-lead ECG signals and related diagnostic results (by the AED algorithm) are
displayed on the screen and stored into data files with the format of WaveForm
DataBase (WFDB).
 The diagnostic reports can be produced automatically and be printed with the ECG
signals sequence and the related statistic results.

01/03/2025 11
PCC System Architecture-Remote surveillance server
 When a patient’s number of a PCC application is few, e.g., an application of small
clinic, the cardiologist of the remote surveillance server can directly connect with the
patient in the local access server via the Internet. In this way, the database service and
Web service can be implemented directly in the remote surveillance server.

01/03/2025 12
PCC Operation Modes
 This PCC system enables four operation modes which work together to make the
system always adaptable to different application scenarios and requirements.
Cardiologists can reset the operation mode by taking into account patients’ physical
status and network medium access bandwidth. The key features of the four operation
modes are illuminated as follows:
Real-time continuous ECG signal:
 For the sake of remote real-time displaying and diagnosing, the data including
continuous ECG signals acquisition and its diagnosis report will be sent in real-time to
the remote system. This operation mode has the highest alarm level that enables online
diagnosis in real-time.
 Note that this mode does not fit to monitor a large number of patients due to the
limitations of network bandwidth, system resources and medical resources, but it has
remarkable performance in monitoring the CSD high risk patient.

01/03/2025 13
PCC Operation Modes
ECG signal sequence
 In order to satisfy remote real-time multi-patient detection and monitoring, WES is
configured to send automatically a sequence of ECG signals (pre- and post-
abnormality) to the remote system when a cardiac arrhythmia event defined by
cardiologists is detected.
 This operation mode is suitable for long-term multi-patient (lower risk of sudden death
than the previous class of patients) cardiac arrhythmias events surveillance.
Textual emergency message
 In this mode, only a short textual emergency message will be sent to cardiologists
when a cardiac arrhythmia event is detected. According to the gravity of the symptom,
the cardiologists can decide to intervene immediately or later. This mode may be
operated on any access medium (wire or wireless).

01/03/2025 14
PCC Operation Modes
Diagnosis report email
 It is the lowest level operation mode. The local server will send periodically a report
attached to an email to the remote server. The cardiologists predefined the reporting
period. This mode is suitable to monitor a large number of patients.

01/03/2025 15
Key Technologies of PCC System-Lossless ECG Signal
Compression
The key technologies developed for the PCC system, including
 The lossless signal compression technique,
 Adaptive real-time communication mechanism, and
 The adaptive ECG diagnosis (AED) algorithm.

Lossless ECG Signal Compression:

 Minimizing the amount of the data transmission is the most effective technique to
reduce network traffics, and it is thus necessary to compress the ECG signals before
data transferring.
 The 500 Hz sampling frequency in WES offers high quality ECG signals for the ECG
diagnosis, but also increases network traffics for the data transmission.

01/03/2025 16
Lossless ECG Signal Compression
 For example, a 4-lead ECG signals with 500 Hz sampling frequency and 12 bit sample
data (ADC resolution), the size of its 5 s frame is 20 KB (4 leads * 500 Hz * 5 s * 2
byte).
 In order to support real-time data transmission through classical modem, the network
bandwidth must be superior to 32 Kbps (20,000 * 8 bit/5 s) when PCC works in the
level-I operation mode.
 This speed is obviously unfitted for the low bandwidth networks. The ECG signals
with the high sampling frequency (500 Hz in WES) are absolutely necessary to
guarantee the accuracy of the AED algorithm.
 Note that ECG signals with the low sampling rate (often 128 Hz) are acceptable for
the purpose of the ECG observation in the remote visualization system.

01/03/2025 17
Lossless ECG Signal Compression
 On the other hand, the morphological features of ECG signals show that the signals
have huge data redundancies. A good data compression algorithm can thus reduce
network traffics.
 To minimize energy consumption and ease VLSI integration of WES, we adopt a very
simple no-loss compression algorithm: only the difference value between two
consecutive samples is sent to the remote server.
 According to the data sizes of the difference values, the sample compression format is
defined in Figure.

01/03/2025 18
Lossless ECG Signal Compression
 Where the ‘‘ + /−” field indicates the positive/negative property of the difference
value: 1 — positive and 0 — negative;
 The “type” field indicates the type of data length, and the “data” field stores the
difference data.
 The statistical results show that this compression algorithm has 50%–60%
compression ratios.

01/03/2025 19
Adaptive Communication Mechanism
 In order to support a real-time data transmission, the UDP protocol is adopted in PCC to
transmit ECG signals. Because the UDP protocol does not offer a guaranteed datagram
delivery service, a reliable data transmission mechanism must be implemented in the
application layer.
 In fact, the network bandwidth of the PCC communication system is normally
fluctuated and affected by network traffics and some interference factors, an application
layer communication mechanism dedicated to PCC is thus implemented, which enables
reliable real-time data transmission in various healthcare environments.

01/03/2025 20
PCC data frame
 The transfer data unit between remote and local peers in PCC is named a PCC frame,
which is used to establish/terminate connection, to deliver data (ECG signals or
images), and to configure operation modes or other system parameters.
 Each frame consists of two parts: a frame header followed by data. Figure shows the
PCC frame format.
 The type field is used to identify the type of PCC frame. Three frame types are
implemented in PCC:
 a value of 0 × 01 indicates a system control frame;
 a value of 0 × 02 indicates an ECG signal frame; and
 a value of 0x03 indicates an image frame.
 Each type of PCC frames has a unique system priority identified by the type value (1 to
3, from high to low).

01/03/2025 21
 a

01/03/2025 22
PCC data frame
 Each ECG frame has a unique identifier specified by the sequence number. Every time
when the local server sends an ECG frame, the sequence number increments
automatically by one for next frame.
 Basing upon this value, PCC calculates the surveillance time of heart monitoring. The
value of signal number field indicates the channel number of ECG signals and the
default channel number is four in WES.
 The value of QRS number fields represents the number of QRS complexes detected by
the AED algorithm in the ECG frame.
 The default sampling frequency of WES is 500 Hz, which is identified in the sampling
frequency field. This value is alterable by the sub-sampling frequency operation
because the signal compression algorithm changes the size of original ECG frame.
 Hence this frequency value is useful to define the length of the uncompressed original
frame in the remote server.

01/03/2025 23
PCC data frame
 The following area of the ECG frame stores the diagnostic results of the AED
algorithm. It is a QRS structure queue, where the QRS member number is indicated in
the QRS number field.
 Each QRS member named QRS Result consists of three elements: QRS position, QRS
length, and QRS state.
 The QRS position indicates the onset position of a QRS complex in the uncompressed
ECG frame. The heart rate of each beat can be calculated basing on the positions of two
consecutive QRS.
 The QRS length represents the time interval of a QRS.
 The QRS state shows the type of heart rhythm and its related heart statue classified by
the AED algorithm.
 The ECG signals field stores the compressed ECG signals. In the PCC system, each
ECG frame contains 5 s of ECG signals.

01/03/2025 24
PCC data frame
 Figure b shows the format of the image frame. The sequence number field is unused for
an image frame. The image field contains an image in the jpeg format.
 Figure c shows the format of the system control information frame. The value of control
code field indicates the type of control code. The related control information is stored in
the following control information field.

01/03/2025 25
PCC communication mechanisms
 Three kinds of UDP connections are established in PCC. They are responsible for the
System control (udp CMD),
the ECG signal transmission (udp SIG), and
the image transmission (udp IMG).

PCC system control

PCC control frames are responsible for the system remote configuration, the patient
online/offline notifications, and the ECG frames retransmission management. The
control frame is transferred via the UDP CMD connection. PCC defines nine types of
control frames:
• REQ_Connect /ACK_Connect; REQ_Terminate /ACK_Terminate
• REQ_Configure /ACK_Configure; REQ_Restra
• ACK_5Frames; ACK_Image.

01/03/2025 26
PCC communication mechanisms
Signal retransmission mechanism
 PCC guarantees a reliable ECG signal delivery service by providing a signal
retransmission mechanism.
 Two ECG frame queues are defined, respectively, for the local access server (Hold
Queue) and the remote surveillance server peer (Wait Queue).
Data competition mechanism
 Considering the network quality and communication cost, PCC employs a data
competition mechanism to guarantee those important data (i.e., ECG signals) being
transferred in a higher priority, so that in the remote surveillance peer, cardiologists
can detect and diagnose the ECG signals in real time.

01/03/2025 27
AED (Automatic ECG Diagnosis) Algorithm
 In recent years, the study of automatic ECG diagnosis (AED) techniques addressed
mainly clinical applications, but rarely PHC applications, due to two reasons:
1) to be applied on various scenes, i.e., different objectives and environments, the
system signals are ambulatory, easily being disturbed by interferences;
2) to meet the needs of portable, the system node has strict resource constraints,
traditional AED techniques are resource-consumption in views of PCC services.

 A novel AED algorithm dedicated to the PCC applications has been developed. By
offering the PCC service, this system can reduce the risk of cardiac sudden death by
detecting cardiac arrhythmia events in time.

01/03/2025 28
AED-Signal preprocessing and conditioning
 Due to the nonstationary and easy-to-be-disturbed features, the ambulatory ECG
signals must be de-noised before decision making. Most of artifacts, such as baseline
drifts, electrical noises, and muscle tremor interferences, can be eliminated or
restrained by adopting suitable filters.
ECG time series
 Three ECG signal series, i.e., R(t), AD(t), and RC(t), are adopted in the AED
algorithm. The R(t) series is raw ECG signals acquired from electrodes. It is generally
contaminated by different kinds of noises.
 The AD(t) series is the adaptive differential signals with the processing of the
differential filter and the adaptive filter. The inferences of the baseline drift and the
motion artifacts can be eliminated in the AD(t) series; hence, this series is used to
detect and localize the QRS complexes.

01/03/2025 29
 AED-Signal preprocessing and conditioning
 The RC(t) series is the de-noised ECG signals with the operations of the band pass
filter and the linear amplifier. Since electrical noises and muscle tremors have been
removed from the RC(t) series, the RC(t) series is used to extract the characteristics of
the QRS complexes.
Adaptive filter
 Classical filters for the ECG series, e.g., notch filter, low-pass filter, and high-pass
filter, can effectively remove or reduce most of interferences.
 But for motion artifacts, because of their irregular occurrences and irregular
morphological attributes, these filters cannot eliminate these disturbances. These
artifacts can cause much trouble in QRS detection when encountering QRS-like
artifacts.
 This algorithm adopts an adaptive filter (AT) to reduce motion artifacts. The resultant
signal series, named A(t), is generated by performing AT operation in the raw series
R(t).
01/03/2025 30
 AED-Signal preprocessing and conditioning
 The AT expression is

where α is the balance coefficient of AT, which is a key factor to influence the
performance of the AT filter.

01/03/2025 31
 AED-QRS complex detection
 New QRS detector has been developed to eliminate noises and artifacts by exploiting
a self-adaptive threshold (SAT) method and designing state transition recognition
(STR) procedure.
 The SAT method is used to estimate the peaks of ECG sub-segments and the means of
contextual thresholds, which allow estimating the optimum thresholds in a segment
space.
 The STR procedure traces the waveform changes of signal series and identifies QRS
complexes based on the optimum thresholds and the rules of state transition.

01/03/2025 32
AED-QRS complex detection

01/03/2025 33
AED-Feature extraction: Geometric analysis method
 QRS complex has triangular-alike or triangular-component morphological
characteristics (see Fig.).
 This chapter thus adopts the geometric analysis method (GAM) to extract the
features of QRS complexes.
 GAM has simple operations and low resource consumption, being able to predict and
estimate the key points of QRS complexes under noisy situations, such as R wave
peak, end point of Q wave (Qt), and onset point of S wave (Si ).

01/03/2025 34
AED-Feature extraction: Geometric analysis method
 Therein, R wave peak can be obtained from Tpeak1 or Tpeak2, and it has mono-peak or
poly-peaks. The measurement and detection phases of Qt and Si points are
illuminated as follows.

01/03/2025 35
Conclusion
 The system provides high quality ECG signals and accurate QRS detection PCC is an
efficient system for diagnosing cardiac arrhythmias. It provides a new clinical
approach adapted to monitor accurately and efficiently large scale high-risk patients.

01/03/2025 36