Transcript Outline Nature of Biomedical Signals Biomedical Signal Analysis

Biomedical Signal Processing
Lecture 1
梁 勝 富
成功大學 資訊工程系
[email protected]
Office: 資訊系館 12F 65C06, Tel: Ext. 62549
Lab:神經運算與腦機介面實驗室 (3F 65301)
http://ncbci.csie.ncku.edu.tw/
Outline
Nature of Biomedical Signals
 Biomedical Signal Analysis

Ref: Chapter 1 Introduction to Biomedical Signals in Biomedical Signal
Analysis, Rangaraj M. Rangayyan, Wiley, ISBN: 0-471-20811-62.
2
Nature of Biomedical Signals

Living organism are made up of many
component systems




Nervous system
Cardiovascular system
Musculoskeletal system
Each system carries on many physiological
processes


The cardiac system performs the rhythmic
pumping blood throughout the body and the
pulmonary system.
They are accompanied by or manifest themselves
as signals that reflect their nature and activities.
3
Nature of Biomedical Signals

What is a signal?
 A signal is a single-valued
representation of information as
a function of an independent
variable (e.g., time).
4
Example-Speech Signals

“1”



Channels: mono
Sampling rate: 8kHz
Bits (resolution): 16 bits


Samples: 4409
Amplitudes: -18,000~20,000
5
Example-Speech Signals

“1”

Fundamental frequency: 159.4 Hz
6
Example-EEG Signals
7
Example-Blood Pressure

Systolic: 收縮壓/Diastolic: 舒張壓
8
Nature of Biomedical Signals

The types of biomedical signals:

Biochemical:



Electrical



Hormones
Neurotransmitters
Potential
Current
Physical


Pressure
Temperature
9
Action Potential (動作電位)

Action potential




the basic component of all bioelectrical signals.
provides information on the nature of physiological
activity at the single-cell level.
An excited cell displaying an action potential is
called depolarized (去極化).
caused by the flow of sodium (Na+), potassium
(K+), chloride (Cl-), and other ions across the cell
membrane.
10
The Neuron

cell body (soma)

Axon (軸突)

Dendrite (樹突)
11
The neuronal membrane


5 nm thickness
The membrane acts as a capacitor.
12
Action Potential (動作電位)

Resting (polarized)





Sodium (Na+) concentration inside the cell is far
less than the outside.
The outside of the cell is more positive than the
inside of the cell.
To balance the charge, additional potassium (K+)
ions enter the cell, causing higher K+
concentration inside the cell than outside.
Charge balance cannot be reached due to
differences in membrane permeability for the
various ions.
A state of equilibrium is established with a
potential difference, with the inside of the cell
being negative with respect to the outside.
13
Action Potential (動作電位)

Depolarization




A cell is excited by ionic currents or an
external stimulus.
Membrane allows sodium (Na+) ions rush
into the cell.
The potassium (K+) ions leaves the cell but
cannot move as fast as sodium (Na+) ions.
The inside of the cell becomes positive with
respect to the outside.
14
Action Potential (動作電位)

Repolarization


The predominant membrane permeability
is for potassium (K+) ions (to leave the
cell).
It makes the inside more negative back to
the resting potential.
15
Action potential
16
All-or-None law


The action potential is always the same for
a given cell, regardless of the method of
excitation or the intensity of the stimulus
beyond a threshold.
The duration of action potentials lasts only
about 3 ms because the channels are only
open for a fixed period of time.
(inactivation of Na+ channels)
17
Absolute refractory period

After an action potential, there is a period
during which a cell cannot respond to any
new stimulus.
18
Examples
19
Electroneurogram (ENG)




ENG is an electrical signal observed as a
stimulus and the associated nerve action
potential propagates over the length of a nerve.
Measuring the nerve conduction velocity (NCV)
of an action potential in a nerve by comparing
the latencies recorded at different locations.
ENG have amplitudes of the order of 10 uV.
Stimulus pulse:

amplitude: 100V, duration:100-300 s
20
Electroneurogram (ENG)



The stimuli was applied to the ulnar nerve (尺骨神經).
The grid boxes: 3ms in width, 2uV in height
NCV=Distance(Belbow,Wrist)/time
21
=20.96(cm)/3.23(ms)=64.9 m/s
Electroneurogram (ENG)

Typical values of NCV




Nerve fibers: 45-70 m/s
Heart muscle: 0.2-0.4 m/s
Neural disease may cause a decrease in
conduction velocity.
An example: the nerve conduction report
22
Electromyogram (EMG) 肌電波


Skeletal muscles are make up of collection of
motor units (MUs).
Motor unit (MU):




Single-motor-unit action potential (SMUAP):


a motorneuron,
its axon (軸突),
and muscle fibers innervated by that axon
electrical signal of a motor unit that is the summation
of the action potentials of all of its constituent cells
Normal SMUAPs:




biphasic or triphasic
3-15 msec in duration
100-300 V in amplitude
appear with frequency in the range of 6-30 Hz
23
Electromyogram (EMG)
Biphasic
Triphasic
SMUP trains recorded simultaneously from three
channels of needle electrodes
24
Electromyogram (EMG)

EMG of the muscle: spatio-temporal summation
of the MUAPs of all of the active motor units.
25
Electromyogram (EMG)



EMG of a normal subject.
 SMUAPs are mostly
biphasic.
Patient with neuropathy.
 SMUAPs are polyphasic
and large in amplitude.
Patient with myopathy.
 SMUAPs are polyphasic.
26
Chin EMG
Wake
REM
27
Electrocardiogram (ECG)




ECG is the electrical manifestation of the
contractile activity of the heart and can be
recorded with surface electrodes on the
limbs or chest.
Two atria (心房)collect blood and two
ventricles (心室) pumps out blood.
The rhythm of the heart in term of beats
per minute (bpm) may be easily estimated
by counting the readily identifiable waves.
The normal (resting) heart rate is 70 bpm.
28
Electrocardiogram (ECG)



The right atrium (RA) collects impure blood
from the vena cava (靜脈).
During atrial contraction, the impure blood
is passed to the right ventricle (RV).
During ventricular systole, the impure blood
is pumped out to the lungs for purification.
29
Electrocardiogram (ECG)



The left atrium (LA) receive purified blood
from the lungs.
During atrial contraction, the purified blood
is passed to the left ventricle (LV).
During ventricular systole, the oxygenated
blood is pumped out to the body through
the aortic valve (動脈).
30
ECG --- not recording of action potential of heart;
but results from the production and conduction of
cardiac action potentials
31
From Mohrman, DE & LJ Heller: Cardiovascular physiology, 2005
32
Electrocardiogram (ECG)





P wave: atria contraction (心房收縮), 0.1-0.2mV, 60-80ms
PQ segment: propagation delay at atrio-ventricular node (房室結),6080ms
QRS wave: ventricle contraction (心室收縮), 1mV, 80ms
ST segment: slow action potential of ventricular muscle cells, 100120ms
T wave: ventricle repolarization (relaxation), 0.1-0.3mV, 120-160ms
33
34
35
Myocardial ischemia (心肌缺血)
Right
coronary
artery
Aorta
Left
coronary
artery
Blockage
Dead muscle
tissue
From Campbell, et al. : Biology: concepts & connections 5th edi., 2006
Fig. 13.31
36
Electrocardiogram (ECG)

Some important features of the standard clinical
ECG:




peak value is normally about 1mV
amplifier gain used is about 1,000
recommended sampling rate is 500 Hz.
filtered bandwidth is about 0.05-100 Hz (for clinical)
or 0.05-500 (for high resolution).
37
Electroencephalogram (EEG)



EEG measures the cortical potential (皮質
電位) of Brain.
Cortical potentials are generated due to
excitatory and inhibitory post-synaptic
potentials (突觸後電位)developed by
pyramidal neurons (錐體神經元).
Scalp EEG is the average of multifarious
electrical activities of many small zones of
the cortical surface beneath the electrode.
38
Electroencephalogram (EEG)
39
侵入與非侵入式量測

Noninvasive-Scalp EEG

(非侵入式顱外腦波)

Electrocorticography (ECoG)-intracranial EEG (iEEG)

(顱內腦皮質電位)


Local Field Potential
(顱內局部腦電位)
40
各種腦電訊號
Scalp EEG (eyes open)
Scalp EEG (eye closed)
Intracranial EEG (interictal, opposite of epileptic zone)
Intracranial EEG (interictal, epileptic zone)
Intracranial EEG (seizure)
(Andrzejak et al., 2001)
41
認知功能實驗室
NuAmp
NeuroScan
Electrode cap
Data Store
Server
42
睡眠實驗室
Polysomnography
43
Electroencephalogram (EEG)

10-20 system for electrode placement


Recommended by International Federation of
Societies for Electroencephalography and Clinical
Neurophysiology
10-20 indicates that the electrodes are placed at
10, 20, 20, 20, 20, 10 % of the total nasion-inion
distance.
44
45
Electroencephalogram (EEG)

The rhythm of EEG
: 0.5≤f<4 Hz, deep sleep stage
: 4≤f<8 Hz, beginning stage of sleep
: 8≤f≤13 Hz, principal resting rhythm;
eye closed
: f>13 Hz, background activity in
tense and anxious status
46
Event-Related Potentials (ERP)



Event-related potential (ERP) represents
the EEG in response to visual (light),
auditory (sound), electrical, or other
external stimuli.
ERPs are weak signals buried in ongoing
activity of associated systems
Signal-to-noise ratio (SNR) improvement
is usually improved by synchronized
averaging and filtering.
47
Event-Related Potentials (ERP)
 Visual ERPs
uV
+
P300
P100
N150
48
Event-Related Potentials (ERP)
 Auditory ERPs
N1
N2
P2
49
Comparison of Biomedical Signals
Voltage[V]
10m
Respiration (呼吸率)
Electrocardiogram Electromyogram
( 心電圖, ECG ) ( 肌電圖, EMG )
1m
100μ
10μ
ENG(electronystagmography, 神經電位傳導圖)
Brain Waves
( 腦波圖, EEG )
1μ
Evoked Potential (Brain Wave, 誘發性腦波圖)
0.01
0.1
1
10
100
1K
10K
Frequency [Hz]
50
Biomedical Signal Analysis

Objectives





Information gathering-measurement of phenomena to
interpret a system
Diagnosis-detection of malfunction, pathology, or abnormality.
Monitoring-obtaining continuous or periodic information about
a system.
Therapy and control-modification of the behavior of a system
based on the outcome of the activities to ensure a specific
result.
Evaluation-objective analysis to determine the ability to meet
functional requirements, obtain proof of performance, perform
quality control, or quantify the effect of treatment.
51
Biomedical Signal Analysis

A complete biomedical signal Analysis
System contains:



Signal data Acquisition
 Sensor and transducers
 Amplifiers and filters
 Analog to digital conversion
Signal Processing
 Filtering to remove artifacts
 Detection of events and components
Signal Analysis
 Analysis of events and waves
 Feature extraction
 Pattern recognition and classification
 Diagnostic decision
52
Biomedical Signal Analysis

Computer-aided diagnosis and therapy
53
Biomedical Signal Analysis

Wireless Polysomnography and automatic sleep
scoring
54
Biomedical Signal Analysis

Aspects to be considered in the design of
biomedical instruments





Isolation of the subject or patient-avoid
electrocution
Range of operation-the minimum to maximum
values of the signal or parameter being measured.
Sensitivity-the smallest signal variation measurable.
Linearity-desired over at least a portion of the range
of operation.
Hysteresis (遲滯)- should be corrected.
55
Biomedical Signal Analysis

Aspects to be considered in the design of biomedical
instruments




Frequency response-most system exhibit a lowpass
behavior and the compensation may be required.
Stability-an unstable system could preclude repeatability
and consistency of measurements.
Signal-to-noise ratio (SNR)- power-line interference,
thermal noise, etc., could compromise the quality of the
acquired signal. A good understanding of the signaldegrading phenomena is necessary to design appropriate
filtering and correction procedures.
Accuracy-includes the effects of errors.
56
Biomedical Signal Analysis

Difficulties encountered in biomedical signal acquisition
and analysis







Accessibility of the variables to measurement.
Variability of the signal source.
Inter-relationships in interactions among physiological
systems.
Effect of the instrumentation or procedure on the system.
Physiological artifacts and interference.
Energy limitation.
Patient safety.
57
Study

Chapter 1 of Biosignal and Biomedical
Image Processing, John L. Semmlow,
Marcel Dekker, ISBN: 0-8247-4803-4 3.
58