Lecture 2-3 Classification of Biosignals - kpi.ua

66
Lecture 2-3 Classification of Biosignals Digital Signal Processing and Analysis in Biomedical Systems

Transcript of Lecture 2-3 Classification of Biosignals - kpi.ua

Page 1: Lecture 2-3 Classification of Biosignals - kpi.ua

Lecture 2-3Classification of

Biosignals

Digital Signal Processing and Analysis in Biomedical Systems

Page 2: Lecture 2-3 Classification of Biosignals - kpi.ua

Contents

- What and why biosignals?

- Classification of biosignals

- Examples

2

Page 3: Lecture 2-3 Classification of Biosignals - kpi.ua

What are biosignals?

All types of biomedical systems either generate the signals to influence the

human body, or analyze biosignals to extract useful information about

functioning of human body.

Signal – is the parameter that is observable from the object.

Biosignal is a description of physiological phenomenon of any nature.

Bio+Signal = “living object” + “function that carries information about the

behavior or state”. Biosignals are the key objects in Biosystems.

3

Page 4: Lecture 2-3 Classification of Biosignals - kpi.ua

Variety of tasks - 1

- Signal generation (therapy, calibration, control)

- Coding and compression (security)

- Transmission (telemedicine)

- Registration and measurement (diagnostics)

- Processing (filtration)

- Transformation (sampling, representations in other spaces)

- Defining the parameters (feature extraction)

- Analysis (extraction of information)

4

Page 5: Lecture 2-3 Classification of Biosignals - kpi.ua

Variety of tasks - 2

- Connectivity and influence (between various processes)

- Prediction (forecasting the future values, states and events)

- Classification (Machine Learning)

- Control (biofeedback systems)

- Visualization (signals, analysis results)

- Storing (databases, cloud storages etc.)

5

Page 6: Lecture 2-3 Classification of Biosignals - kpi.ua

Signal analysis

At the level of signals, the biomedical

system for diagnosis is based on

Machine Learning.

6

Page 7: Lecture 2-3 Classification of Biosignals - kpi.ua

Why biosignals?

Biosignal carries all information about the living object. We analyze signals

which are coming from the body (ECG, EEG etc.) or are connected to the body (X-

ray images, ultrasonic images).

Biosignal can be used to understand the underlying physiological mechanisms

of a specific biological event or system.

7

Page 8: Lecture 2-3 Classification of Biosignals - kpi.ua

Biosignals through centuries

8Hippocrates palpating a young patient (500 BC) ICU in modern hospital (2016 AD)

Page 9: Lecture 2-3 Classification of Biosignals - kpi.ua

Example - Polysomnogram

9

Page 10: Lecture 2-3 Classification of Biosignals - kpi.ua

PSG signal

10

Page 11: Lecture 2-3 Classification of Biosignals - kpi.ua

Example - 2

11

http://wexnermedical.osu.edu/mediaroom/pressreleaselisting/new-device-allows-brain-to-bypass-spinal-cord-move-paralyzed-limbs

Page 12: Lecture 2-3 Classification of Biosignals - kpi.ua

Classification of biosignals - 1

According to the physical nature of biosignals- Electric- Magnetic- Chemical- Mechanical (acoustic)- Optical- Thermal

12

Page 13: Lecture 2-3 Classification of Biosignals - kpi.ua

Classification of biosignals - 2

According to the system of origin of biosignals- Endocrine system- Nervous system (Central and Peripheral)- Cardiovascular system- Vision system- Auditory system- Musculoskeletal system- Respiratory system- Gastrointestinal System- Blood system

13

Page 14: Lecture 2-3 Classification of Biosignals - kpi.ua

Classification according to the physical nature of

signal

14

Page 15: Lecture 2-3 Classification of Biosignals - kpi.ua

Electric signals

Electric field is generated in cells (nerve and muscle) and organs because of

intra- and extracellular ionic currents. They are the results of electrochemical

processes in the single ionic channels.

15

Page 16: Lecture 2-3 Classification of Biosignals - kpi.ua

Action potential generation

16

Page 17: Lecture 2-3 Classification of Biosignals - kpi.ua

Types of electrical signals

Neural cellsENG – electroneurogram

EEG – electroencephalogram

ERG – electroretinogram

Muscle cellsECG – electrocardiogram

EMG – electromyogram

Other cellsEOG – electrooculorgam

GSR – galvanic skin response17

Page 18: Lecture 2-3 Classification of Biosignals - kpi.ua

ECG generation

18

Page 19: Lecture 2-3 Classification of Biosignals - kpi.ua

Bioelectric signals

Amplitude and spectral ranges of

bioelectric signals

19

Page 20: Lecture 2-3 Classification of Biosignals - kpi.ua

Magnetic biosignals

Weak magnetic fields are generated by different organs and

cells.

Neural cellsMNG – magnetoneurogram

MEG – magnetoencephalogram

Muscle cellsMCG – magnetocardiogram

MMG – magnetomyogram

20

Page 21: Lecture 2-3 Classification of Biosignals - kpi.ua

21

Page 22: Lecture 2-3 Classification of Biosignals - kpi.ua

Magnetic signal strength comparison

22

Page 23: Lecture 2-3 Classification of Biosignals - kpi.ua

Magnetoencephalography (MEG)

MEG is based on measuring the magnetic field outside the head using an array

of very sensitive magnetic field detectors (magnetometers). The signals

recorded by EEG and MEG directly reflect current flows generated by neurons

within the brain. The temporal frequency content of these signals ranges from

less than 1 Hz (one cycle per second) to over 100 Hz (100 cycles per second).

Because MEG and EEG measure neuronal activity in “real time” the connections

activated either at rest or during task can be measured, giving us a picture of the

dynamic interactions among brain networks. MEG has much greater temporal

resolution than fMRI so MEG-based analysis provides high temporal resolution

data for analyzing the neuromagnetic correlates of fMRI connectivity, its time-

frequency content, and temporal interactions.23

Page 24: Lecture 2-3 Classification of Biosignals - kpi.ua

MEG measurement

Magnes 3600 with 248 magnetometers within a shielded room with 64 EEG Voltage Channels and 23 MEG reference

channels (5 gradiometer, and 18 magnetometer).

Five current coils attached to the subject, in combination with structural imaging data and head surface tracings, are

used to localize the brain in geometric relation to the magnetometers. Auditory (earphone), visual (screen external to

shielded room) and somatosensory (air-driven tactile) stimuli can be delivered. 24

Page 25: Lecture 2-3 Classification of Biosignals - kpi.ua

MEG signal

25

Page 26: Lecture 2-3 Classification of Biosignals - kpi.ua

fMRI+MEG during rest

Source-space band-limited power (BLP) correlation maps obtained for the Dorsal Attentive Network. Left: fMRI

seed based conjunction maps. Right: MEG t-statistic images across subjects computed using epochs of

maximum correlation. The topography of the non-stationary MEG Resting State Network (RSN) is similar to the

RSN obtained by fMRI.26

Page 27: Lecture 2-3 Classification of Biosignals - kpi.ua

Magnetocardiography (MCG)

MCG is the measurement of magnetic fields emitted by the human heart from

small currents by electrically active cells of the heart muscle.

27

Page 28: Lecture 2-3 Classification of Biosignals - kpi.ua

Data derived from MCG

28

Map of magnetic field distribution

Averaged MCG 36-channels

Map of current density distribution

Page 29: Lecture 2-3 Classification of Biosignals - kpi.ua

Myocardium current density maps

29

Page 30: Lecture 2-3 Classification of Biosignals - kpi.ua

Chemical biosignals

Signals providing information about concentration of various chemical agents in

the body

- Level of glucose (diabetes)

- Blood oxygen level (asthma, obstructive pulmonary disease, heart and kidney

failure)

- Gases in blood and breathing airflow (anesthetic gases, carbon dioxide etc.)

- pH

30

Page 31: Lecture 2-3 Classification of Biosignals - kpi.ua

Blood saturation of gases

SaO2 – arterial blood oxygen saturation, describes the percentage of

hemoglobin molecules carrying oxygen.

SvO2 – venous oxygen saturation, describes how much oxygen the body

consumes

SpO2 – peripheral capillary oxygen saturation – the same as SaO2 but in the

capillary system

SpCO2 – concentration of carboxyhemoglobin in blood

31

Page 32: Lecture 2-3 Classification of Biosignals - kpi.ua

Photoplethysmography

32

Page 33: Lecture 2-3 Classification of Biosignals - kpi.ua

33

Page 34: Lecture 2-3 Classification of Biosignals - kpi.ua

34

Page 35: Lecture 2-3 Classification of Biosignals - kpi.ua

Mechanical biosignals

Biomechanical signals reflect mechanical functions of body parts

Examples:

- Blood Pressure

- Accelerometer signals describing human movements, gait, balance and pose

(Parkinson disease, mobile applications, fitness)

- Chest movements during respiration

- Air flow characteristics during MLV

35

Page 36: Lecture 2-3 Classification of Biosignals - kpi.ua

3D tracking using accelerometer

36

Page 37: Lecture 2-3 Classification of Biosignals - kpi.ua

Acoustic biosignals

Subset of mechanical signals that describe the acoustic sound produced by the

body (vibrations and motions). Bioacoustic signals give access to diverse body

sounds:

- Cardiac sounds (phonocardiography)

- Snoring (Obstructive Sleep Apnea detction)

- Swallowing

- Respiratory sounds

- Crackles of joints and muscles

Often measured at the skin using acoustic transducers such as microphones

and accelerometers. 37

Page 38: Lecture 2-3 Classification of Biosignals - kpi.ua

Phonocardiogram (PCG)

PCG reflects sounds of heartbeats, produced by heart sounds corresponding to

two consecutive heart valve closures. Indicates closure strength and the valve’s

stiffness.

38

Page 39: Lecture 2-3 Classification of Biosignals - kpi.ua

Respiratory sounds

Reflect normal breathing sounds superimposed with crackles, cough sounds,

rhonchus, snoring, squawk, stridor and wheeze sounds, which are associated

with pulmonary disorders.

39

Page 40: Lecture 2-3 Classification of Biosignals - kpi.ua

Electronic stethoscope

40

Page 41: Lecture 2-3 Classification of Biosignals - kpi.ua

Optical biosignals

Optical methods are among the oldest and best-established techniques for

sensing biochemical analytes.

41

Page 42: Lecture 2-3 Classification of Biosignals - kpi.ua

Immunology (antibody-antigen interaction)

Surface plasmon resonance is the collective oscillation of electrons

stimulated by incident light.42

Page 43: Lecture 2-3 Classification of Biosignals - kpi.ua

Thermal biosignals

Body temperature in the point and temperature maps, may describe heat loss

and heat absorption in the body, or temperature distribution over the body

surface.

43

Page 44: Lecture 2-3 Classification of Biosignals - kpi.ua

Thermography (temperature maps)

- Cancer

- Varicose veins

- Osteochondrosis, osteoporosis

44

Page 45: Lecture 2-3 Classification of Biosignals - kpi.ua

Classification according to the system of origin

of signal

45

Page 46: Lecture 2-3 Classification of Biosignals - kpi.ua

Endocrine System

Is the collection of glands of an organism that secrete hormones directly into the

circulatory system to be carried toward a distant target organ.

Signals:

- Chemical

- Optical

46

Page 47: Lecture 2-3 Classification of Biosignals - kpi.ua

47

Page 48: Lecture 2-3 Classification of Biosignals - kpi.ua

Signals from Nervous System

Neurons and spinal cord

- Electroneurogram (Spike trains)

- Magnetoneurogram

Brain

- EEG, MEG

- Event-Related Potentials (acoustic, visual)

- Neurovisualization (MRI/fMRI, CT, PET, SPECT)

48

Page 49: Lecture 2-3 Classification of Biosignals - kpi.ua

Cardiovascular System

Heart & blood vessels

- ECG

- MCG (Current Density Maps)

- Blood pressure

- Heart Rate Variability

Visualization

- Ultrasonic Imaging

- MRI, Ultrasonic, X-ray

49

Page 50: Lecture 2-3 Classification of Biosignals - kpi.ua

Vision system

- EEG (visual cortex)

- VEP (Visual Evoked Potentials)

- EOG (Electrooculogram)

- ERG (Electroretinogram)

50

Page 51: Lecture 2-3 Classification of Biosignals - kpi.ua

Electroretinography

Electroretinography measures the electrical responses of various cell types in

the retina, including the photoreceptors (rods and cones), inner retinal cells

(bipolar and amacrine cells), and the ganglion cells.

51

Page 52: Lecture 2-3 Classification of Biosignals - kpi.ua

Auditory system

- EEG (Auditory Evoked Potentials)

- Audiometry

52

Page 53: Lecture 2-3 Classification of Biosignals - kpi.ua

Example of audiogram

53

Page 54: Lecture 2-3 Classification of Biosignals - kpi.ua

Musculoskeletal system

- EMG (Electromyogram)

- Visualization (MRI, X-Ray)

- Reography (myorelaxation)

- Accelerometry (gait)

- Stabilorgaphy (Parkinson’s)

54form, support, stability, and movement

Page 55: Lecture 2-3 Classification of Biosignals - kpi.ua

Stabilography

Analysis of balancing act and postural control

55

Page 56: Lecture 2-3 Classification of Biosignals - kpi.ua

Respiratory system

- Chemical signals (gas concentration)

- Mechanical (airflow, pressure, volume)

- Spirometry (flow-volume)

- Plethysmography (volume)

56

Page 57: Lecture 2-3 Classification of Biosignals - kpi.ua

Spyrography

57

Page 58: Lecture 2-3 Classification of Biosignals - kpi.ua

Gastrointestinal System

- MRI

- X-ray

- Ultrasound Imaging

- Chemical signals

- Electrogastrogram

58

Page 59: Lecture 2-3 Classification of Biosignals - kpi.ua

Elecrtogastrogram

59

Page 60: Lecture 2-3 Classification of Biosignals - kpi.ua

Blood System

- Chemical signals (concentrations)

60

Page 61: Lecture 2-3 Classification of Biosignals - kpi.ua

Other Classificationof biosignals

61

Page 62: Lecture 2-3 Classification of Biosignals - kpi.ua

Dimesionality

• 1D (ECG)

• 2D (temperature map)

• 3D (MRI image)

• 4D (fMRI image)

62Task-related activation

Page 63: Lecture 2-3 Classification of Biosignals - kpi.ua

Number of channels

• One channel (pulse wave)

• Three channels (accelerometer data)

• Multichannel (EEG)

63

Page 64: Lecture 2-3 Classification of Biosignals - kpi.ua

256 channels of EEG

64

Page 65: Lecture 2-3 Classification of Biosignals - kpi.ua

Analysis approach (signal models)

65

Page 66: Lecture 2-3 Classification of Biosignals - kpi.ua

Summary

Biosignals are the only source of information describing the functioning of the

human body in healthy and disease conditions.

Biosignals are of various nature and origin.

Many biosignals can contain information about same organ or system.

Researchers and practitioners should do

measurement,

processing,

analysis,

interpretation

66