The Photoplethysmograph as an instrument for physiological measurement

1
The Photoplethysmograph as an instrument for
physiological measurement

• Tomás Ward
• Department of Electronic Engineering,
• NUI Maynooth

2
Overview

• What is the PPG?
• Common Uses of the PPG
• How we use the PPG
• How we intend to use the PPG
• Conclusion

3
What is the Photoplethysmograph?

• The PPG is an optical means of conducting a
  plethysmography.
• So what is a plethysmography and how do we do it
  optically?

4
What is Plethysmography?

• PG is a term for a set of noninvasive techniques
  for measuring volume changes in parts of the body
  (even the whole body)
• Commonly measured volume changes are
• those caused by breathing (lung and chest
  expansion)
• those caused by blood being forced into vessels
  (such as arteries,veins and capillaries)
• those caused in the heart as it pumps

5
How are PGs commonly acquired ?

• The 2 main techniques are
• Volume Displacement Plethysmography
• Electrical Impedance Plethysmography
• The above two methods are flexible
• It is also possible to acquire a PG using
• Ultrasonic or X -Ray imaging
• A photoplethsymograph refers to a technique
  whereby localised volume changes due to an
  optically absorbant/scattering substance (e.g
  .blood ) are measured.

6
The PPG

• Usually the tissue under investigation is bathed
  with light of a suitable wavelength (usally NIR)
  and the resultant scattered light is measured
  with a silicon photodiode
• Two modes
• Transmissive mode - fingers / toes / earlobe
• Reflective - forehead / cheek
• Received signal is assumed to be a measure of
  volume changes due to localised blood flow

7
Common uses of the PPGThe Finger PPG
This signal is very similar to the peripheral
blood pressure waveform
8
How does the PPG work?

• 15 of blood by weight is hemoglobin inside the
  Red blood cells (RBC or Erythrocytes)
• The total Hb (THb) can have one of the following
  forms
• reduced or non-oxygenated Hb (HbR)
• Oxyhemoglobin (HbO2)
• Carboxyhemoglobin (HbCO)
• Methemoglobin (metHb)
• How do these various forms interact with light?

9
Optical measures from Radiative Tranfer theory - I

• Transmittance of light through an absorbing
  medium is defined by
• where I is the transmitted intensity and I0 is
  the incident intensity.
• Absorbance is given by

10
Optical measures from Radiative tranfer theory -
II
It can be shown that the absorbance can be
further expressed as where ? is the molar
absorptivity (in cm-1 M-1), and l is the path
length (usually in cm), and c is the molar
concentration. This is known as Beers Law
11
Optical Absorbance of HbR and HbO2 and H2O in
1cm cuvette vs ?
In cuvette obeys Beers Law ie we can relate I/I0
to c, ? and l In real blood Hb in erthrocytes
(RBCs) resulting in much scattering and
reflection by the RBC membranes and other tissues
12
IR - LED
I0
Skin
Fat
Capilliary Bed
Arteriole
Venule
I
Transmissive PPG
Photodiode
13
IR - LED
Photodiode
I0
I
Skin
Fat
Capilliary Bed
Arteriole
Venule
Reflective PPG
14
So what really is the PPG a measure of?

• Hard to say! Literature unsatisfactory on the
  subject
• The name conventionally suggests that this device
  should measure volume by optical methods. Really
  it detects changes in blood perfusion in limbs
  and tissues.
• As arterial pulsations fill the capillary bed the
  changes in volume of the blood vessels modify the
  absorption, reflection and scattering of the
  light. Also the amount of HbO changes resulting
  in additional modulation. So the picture is more
  complicated than Beer Lambert Law
• As a raw signal it is best used to show the
  timing of events such as heart beats,
• With additional processing it can provide a
  fairly accurate measure of relative peripheral
  volume change and relative blood pressure change
• Principle involved most useful for oximetry

15
The use of a PPG for determining oxygen
saturation levels - Oximetry and Pulse Oximetry

• Oximetry is the determination of the oxygen
  content of tissue blood
• The measure used is oxygen saturation SpO2 which
  is HbO/THb (as a percentage)

16
Principle of Pulse Oximetry

• By using light at 2 different wavelengths one at
  the isobestic wavelength we can determine the
  ratio of HbO2 to HbR and hence local oxygen
  levels (ideally!!)

17
Probe - Transmissive
18
Probe - Reflective
19
Measures of the received signals are processed to
yield SpO2 values
20
Practical Pulse Oximetry

• Due to scattering effects the actual output of a
  pulse oximeter is not the linear function1 of
  average SpO2 that theory predicts
• Actual SpO2 is found via a lookup table
• For absolute measurement calibration with blood
  sample required
• Does show relative changes - still clinically
  useful

1
Beer-Lambert Law
Vout normalised
Empirical Calibration
0
0 100 - SpO2
21
Section Summary

• PPG produces a measure of blood perfusion changes
  in a local area of tissue
• Pulse Oximetry signal is produced using a PPG
  calculated at two or more wavelengths and
  provides a measure of SpO2 and hence relative
  local oxygen consumption by tissue as a function
  of time

22
Our current use of the PPG

• Measurement of Pulse Transit Time (PTT)
• MEng work Michael Maguire
• Collaboration Diarmuid OShea, Leo Kevin,
  Charles Markham
• Assessment of vascular function
• New research
• Collaborators Michael Maguire, Douglas Leith,
  Patricia Fitzgerald

23
Measurement of Pulse Transit Time using the PPG

• What is PTT?
• Pulse transit time is the time an arterial
  pressure wave takes to travel between two points
  along the same artery.
• Why measure PTT?
• Because it allows a noninvasive measurement of
  arterial blood pressure
• may also allow measurement of certain other
  cardiovascular parameters noninvasively

24
Direct Invasive Measurement of PTT

• Elastic Theory2 linear relationship Pulse Wave
  Velocity and Diastolic BP
• PTT decreases with increasing BP

PTT
PWV
25
Conventional Noninvasive Measurement of Pulse
Transit Time using the ECG and finger PPG
26
Typical Results (Geddes et al., 1981)
High scatter, averaging required for even
moderate accuracy Can we improve on this?
27
Our Method of Measurement of Pulse Transit Time

• Direct measurement of PTT
• Brachial Reflective PPG replaces ECG
• Experiment Actual continuous BP taken with
  Portapress system along with ECG
• Allows PTT as measured using both methods to be
  correlated with BP
• Collaboration St Vincents Hospital

28
Pulse Transit Time

• Improved result over ECG method
• Discrepancy between measures could be an
  indicator of isovolumetric contraction
  variablilty (C. Markham)
• Integration of additional data (HR) may yield
  improved relationship (Barschdorf et al.)

29
Potential of Additional parameterization of
Peripheral Vascular system

• Currently looking at conducting step-response
  measurements for assessing state of peripheral
  vascular system
• Occlude artery under investigation
• Rapidly allow blood back into arterial system
• Monitor PPG
• May yield information on compliance / arterial
  narrowing
• Collaboration Beaumont Hospital

30
Future uses of the Photoplethysmograph in our
research - a Brain Computer Interface (BCI)

• Collaborators Charles Markham, Gary McDarby
• A BCI in the context we discuss here is a
  wearable device that will allow a human user to
  control their environment via thought processes
  alone.
• Next slides
• Why a BCI?
• And How.

31
Why bother?

• There exist people with such profound
  disabilities that they have NO means of
  communicating with the outside world.
• People with amyotrophic lateral sclerosis and
  brainstem stroke for example
• The immediate goal is to provide these users, who
  may be completely paralyzed or "locked in," with
  basic communication capabilities so that they can
  express their wishes to caregivers, operate
  simple word processing programs, or even control
  a neuroprosthesis.

32
How should we proceed?

• Many severely disabled people can communicate
  through the use of switches

No
Yes
Scanning Communication S/W
33
Can we make a Mind Switch?

• Yes
• IF we can come up with a physiological
  measurement modality that will allow different
  neurological or thought processes to be
  distinguished
• Then if a user can voluntarily reproduce a
  thought process we can measure noninvasively then
  we have our Mind switch

34
Current BCIs ( EEG-based )

• Are all based on electrical potentials
  (electroencephalograph/EEG) recorded from the
  scalp (25bits/min, long training)
• Visual Evoked Potentials, P300, Slow Cortical
  potentials, Sensorimotor Cortex Rhythms
• Problems
• Long training times - non-intuitive
• Messy electrodes
• Signal averaging required

35
EEG problems

• The EEG is a representation of groups of waves
  produced by the electrical activity of the cortex
  averaged at a given point.
• EEG is a crude modality, akin to trying to
  discern what is going on at a football game
  through listening to the reactions of the crowd!!
• We require an imaging modality allows us to see
  the brain function related to its anatomy.
• One such modality is Functional magnetic
  resonance imaging (fMRI)

36
fMRI - basic principle

• Application of a large external magnetic field
  causes magnetically active atomic nuclei to
  become oriented parallel to the applied field.
• This resting orientation may be disturbed with an
  external RF (radio frequency) pulse.
• After the RF pulse, the nuclei fall back in line
  with the external magnetic field and, in so doing
  reemit the radio-frequency energy as a signal
  that can be detected by a receiver coil. The
  frequency of this signal reflects number of
  elements in the nucleus, the strength of the
  external magnetic field, and the effect of
  surrounding material.
• fMRI is tuned to the magnetic properties of
  hemoglobin and can distinguish HbO2 from HbR and
  so can image neural activity which results in an
  increase in local oxygen levels

37
fMRI - principle more detail - I

• When neurons fire, they consume oxygen and this
  causes the local oxygen levels to briefly
  decrease and then actually increase above the
  resting level as nearby capillaries dilate to
  allow more oxygenated blood into the active area.
  fMRI works by imaging blood oxygenation, a
  technique called BOLD (Blood Oxygen Level
  Dependence). The BOLD paradigm relies on brain
  mechanisms which overcompensate for oxygen usage
  (activation causes an influx of oxygenated blood
  in excess of that used and therefore the local
  oxyhemoglobin concentration increases.

38
fMRI - principle more detail - II

• Oxygen is carried to the brain in the hemoglobin
  molecules of red blood cells. Luckily for fMRI,
  the magnetic properties of hemoglobin differ when
  it is saturated with oxygen compared to when it
  has given up oxygen. Technically, deoxygenated
  haemoglobin is "paramagnetic" and thefore has a
  short T2 relaxation time. As the ratio of
  oxygenated to deoxygenated haemoglobin increases,
  so to does the signal recorded by the MRI.
  Deoxyhemoglobin increases the rate of
  depolarization of hydrogen nuclei creating the
  NMR signal thus decreases the intensity of the T2
  image. The bottom line is that the intensity of
  images increases with the increase of brain
  activation. The problem is that this increase is
  small (usually less than 2) and easily obscured
  by noise and different artifacts.

39

• Anatomy fMRI

Moving fingers on right hand - the anatomy
Moving fingers on right hand - the fMRI image
40
Typical patterns as you read this text!
41
So why not just use fMRI?

• fMRI is highly sensitive to movement of the head
  - the head must be clamped in place.
• Subjects responses must not involve speaking and
  at most only small movements etc.
• Useful imaging still requires task repetition and
  image averaging.
• The MRI machine is very noisy and somewhat
  claustrophobia-provoking.
• The technique can induce heating of the brain.
• Very expensive (several million dollars)
• Large magnetic fields required (up to 4 Tesla)
• Not portable

42
Alternative to fMRI Monitoring Cerebral Surface
activity with Near-Infrared imaging

• Use of oximetry approach (double PPG) at suitable
  NIR wavelengths
• Use an array of
• POX sensors
• Build up map of
• cortical neural activity
• Called Diffuse Optical Tomography

43
Typical DOT system
44
Right finger movement experiment as seen by DOT
system
45
How DOT compares with other brain imaging
modalities
46
Can we do this ?

• Wait and see
• Project is funded and will commence start of
  April 2002

47
Conclusion

• The PPG is a deceptively useful physiological
  instrument
• Last 12 months has spawned a number of
  interesting experiments
• Expect more useful applications in the future

48
References 1 p397 in Noninvasive Instrumentation
and Measurement in Medical Diagnosis, Robert B
Northrop, CRC Press 2002, NUI Maynooth Library 2
Geddes, Hughes and Babbs, 1969 (reference
incomplete from Geddes Psychophysiology paper) 3
p241 in Noninvasive Instrumentation and
Measurement in Medical Diagnosis, Robert B
Northrop, CRC Press 2002, NUI Maynooth Library
49
Partial Pressure

• John Dalton (1766-1844) - (gave us Dalton's
  atomic theory)
• The total pressure of a mixture of gases equals
  the sum of the pressures that each would exert if
  it were present alone
• The partial pressure of a gas
• The pressure exerted by a particular component of
  a mixture of gases

50
Volume Plethysmography

• Simplest Example Measurement of limb volume
  using pneumatic sphyganometer cuff
• Inflated to P0 ltlt BPdiastolic
• If the limb expands against bladder by ?V
• it will cause ?PP0(?V/V0) allowing ?V to be
  calculated3

51
Impedance Plethysmography

• Usually ac current source (30-75kHz, high freq
  has less physiological effect such as
  electroshock lt1mA pk), impedance changes with
  blood flow
• Major applications
• occulsive impedance plethysmography used to
  detect clots in deep leg veins
• Measurement of depth of respiration and rate in
  ICU (air intake varies impedance)

52
Pulse Oximetry Theory - I
At 650nm (NIR) with concentration C and path
length L held constant the absorbitivity or
extinction coefficient varies with Saturation
percent of Hb as
Sp02Hb/THb
Also we can say
53
Pulse Oximetry Theory - II
Beers Law
Br is the non-Hb absorption of tissues at 650nm
this light is converted to a proportional (Ka)
voltage before being fed through a log10(x)
nonlinearity (KL) to yield
VLrKLlog10(KaIor)KLlog10(KaI0)-KL(Br?rCL) VLrK
Llog10(KaI0)-KL(BrCL(?rmax-mSp02)) Now the light
from the 805 nm isobestic LED yields indepen
dent of SpO2
54
Pulse Oximetry Theory - III
VLiKLlog10(KaI0)-KL(Bi) We subtract VLi from VLr
to get Vo Vo approx (SpO2)(KLCLm)KL(Bi-Br) ie
output is a linear fn of SpO2 but only if Beers
Law were to hold in fact output is a
monotonically inc. fn of SpO2