Week 1 - Intro, Taking Measurements, Signal & Noise Flashcards
1
Q
What are the essential considerations for a measuring system?
A
- ensuring the system is specific to what you are trying to measure
- it is not too complex/sensitive/accurate
2
Q
What are some purposes of the measurement?
A
- improving understanding/ establishing a patient condition
- monitoring patients
- using measurements within a control system to maintain a certain level/temperature for e.g.
3
Q
How to collect a good signal of interest?
A
- improve position of the transducer
- improve specificity of the transducer
- improve signal conditioning to optimise signal of interest
4
Q
What is an example of how the positioning of a transducer can be improved?
A
An ultrasound to monitor fetal heartbeat (palpate for fetal position to estimate location of fetal heart before placing the doppler in that place)
2016, transducer positioning aid, multiple ultrasounds in an array, positioned main one where the others gave off the most signals
5
Q
What does improving the specificity mean/involve?
A
Increasing the number/proportion of identified negatives to reduce the instances that a positive could be identified incorrectly
6
Q
What are the 3 main sources of error?
A
- positioning of the transducer (internal positioning can be limited by anaesthesia for example as lung pressures are different when patient is lying supine)
- presence of the transducer interfering with signal being measured
- characteristics of the transducer & signal processing (static and dynamic characteristics influence transducer behaviour)
7
Q
What does it mean when we talk about static transducer characteristics?
A
Quality of measurements done when maintaining the measured quantity at a constant value or it is moving very very slowly
8
Q
What does it mean when we talk about static transducer characteristics?
A
Quality of measurements done when the measured quantity is changing overtime
9
Q
What are some examples of transducer characteristics?
A
- sensitivity
- linearity
- frequency
- dependence
10
Q
What type of characteristics is linearity?
A
A static characteristic
11
Q
What is linearity?
A
Closeness of the input/output to a straight line
12
Q
How can we determine linearity?
A
- calculate the % non linearity by seeing how close values are to the linear region
- or by calculating the least squares fit between theoretical & true curves of the transducer
- allows us to answer the question: how well does the equation relating input and output fit?
13
Q
Equation for % non linearity
A
(max. input deviation/full-scale input) x 100
generally given in transducer data sheet
14
Q
What is sensitivity defined as?
A
Ratio of output quantity to an input quantity
15
Q
How do we determine sensitivity?
A
- See what a small change in one control does to the other value among different transducers
- A greater change in the value as a result of a small change indicates greater sensitivity
- e.g. a 4V increase in voltage as a result of a 1 degree temperature increase is more sensitive than a 0.6V increase
16
Q
Are linearity and sensitivity a trade off?
A
- can be
- linearity however is optimal as if not linear, sensitivity would depend on magnitude of interest
17
Q
What is resolution?
A
- associated with sensitivity
- the smallest change in input the transducer can detect
18
Q
What are some dynamic transducer characteristics?
A
- how stable is the output over time?
- does it drift off value?
- needs to be calibrated every year or few months depending on how much use it gets to ensure it remains accurate
- critical in medical applications
- less critical in other applications -e.g. a home oven
- need to consider the environmental conditions when these measurements are being taken as can affect output and therefore drift
19
Q
How can changes in conditions influence accuracy/value of the output?
A
- displacement transducers are often made of metal and measure resistance through dimensional changes
- AND metal resistance is affected by temperature
20
Q
What is hysteresis?
A
- does the past loading history affect the output reading
- error size in output per input if moving in opposite direction
21
Q
What is mechanical hysteresis?
A
- backlash in gears period when changing direct so have a period of slack before they engage again
22
Q
What is elastic hysteresis?
A
Material loading
force against extension
23
Q
What is electrical hysteresis?
A
- Schmitt trigger
24
Q
What is system overload?
A
- associated with hysteresis
- load beyond capacity = permanent damage
- may result in wrong read out/no read out
- need to look out for/check when analysing signals
25
Why do we need to process the output signal from a transducer?
- to amplify to a readable value
- to remove it from noise
- to digitise it for display
(first two considered together)
26
Why is amplification needed?
- most signals are in milli volt so need to amplify this small signal to match recorded device amplification
- also help improve signal to noise ratio
- also need to match range of analogue to digital convertors to increase resolution and sensitivity
27
What are the potential issues with amplification?
- large electrical noise so it is difficult to record
- noise can result in errors in recording in situations where we need error free measurements/ good accuracy, sensitivity, resolution etc.
28
What can help minimise problems with amplification and noise?
- signal conditioners (especially if located close to signal source or transducer to then increase signal to noise ratio before noise is introduced by environment)
- must amplify whole signal evenly (this means also amplifying noise and have to accommodate for capacitors which have different reactance at different frequencies)
- use gain = allows amplification to be adjustable and can calibrate it
29
How do we define gain?
Mean ratio of amplitude/power at the input port to the amplitude/power at the output port
30
How to minimise noise in an ECG?
- use a reference electrode = put on muscle/ mains
- can incorporate baseline drift filters
- uses a differential amplifier
31
How does a differential amplifier work?
- pulls out difference between signals and removes common signals
- only amplifies difference between 2 signals
- can remove general noise/ other biopotentials/mains
32
Common Mode Signals
- need to be rejected so we can cancel noise out by cancelling these identical components between two inputs (common)
33
Common Mode Rejection Ratio
- measurement of amplifiers ability to reject common mode signals between 2 inputs
- ratio between differential mode gain and common mode gain
34
What do filters do?
Allow us to pass/amplify certain frequencies whilst also attenuating other frequencies themselves
- split into passive and active
35
Passive filter
- only passive components involved (resistors, capacitors, inductors)
- most response to 100Hz - 300MHz ranges
- limitation on lower end as at lower frequencies inductance or capacitance would have to be quite large
- limitation on upper end frequency is due to effect of parasitic capacitance/inductance
- with careful design you can increase upper limit
- can remove some unwanted frequencies
36
Active filter
- uses active components such as operational amplifiers with resistors + capacitors (passive elements)
- NOT inductors
- deals with very low frequencies (towards 0Hz)
- can also provide voltage gain unlike passive
- design higher order filters without needing inductors
- essentially amplify at the same time as removing unwanted frequencies
37
How do we know which filter is best to use?
- no perfect cut off
- need to account for range of partially attenuated frequencies
- trial and error
- need to be clear about signal frequencies to then optimise bandwidth (can maintain signal and maximise noise removal)
38
What does displacement refer to?
```
position
motion
strain
velocity &acceleration
force & pressure
```
39
Example of measuring displacements directly
- changes in blood vessel diameter
- muscle contraction
- lung movement in breathing
40
Example of measuring displacements indirectly
- measuring displacement of a diaphragm to infer displacement of heart beneath (or blood pressure)
41
Strain Gauges
- often metal/semiconductor
- used in osteotomy
- when wires stretch, CSA is reduced, increases total resistance
- lattice structure is altered so there is resistance increase
- both elements affect strain sensitivity
- elongation causes change in resistance
- increase in length increases R
- L is length
- R is resistance
- A is CSA
- rho is resistivity
- R = rho (symbol) (L/A)
42
Gauge Factor
- factor relating R and L together
- shows us sensitivity to strain itself
- ratio of fractional change in electrical resistance to fractional change in length/strain
- constant for certain materials
- larger change in resistance relative to length = higher gauge factor = higher sensitivity
- very high gauge factors in silicone = high high sensitivity
- piezoresistive effect?
- see formula on slide
43
How are resistive gauges designed?
- wires or deposited metal films
- miniaturised as allows localised measurement of strain and reduces effect of sheer strain
- gauge also detects strain on cross sectional axis as well as sensing axis, corners are thickened to minimise cross-axis sensitivity, typically around 2%
44
Advantages of strain gauges
- small in size
- very high sensitivity (high gain) = maximum strain a wire gauge can permit is very small
- for a gauge factor of 2, 5000microstrain = 1% change in resistance
45
Disadvantages of strain gauges
- need large forces to deform them
- care needed as do not influence parameter to be measured
- influenced by temperature (R depends on resistivity)
46
Displacement is the basis range of what other measurements?
Velocity & Acceleration (via differentiation)
| Force & Pressure (via F=kD)
47
Dummy strain gauge
- allows us to compensate for temperature fluctuation and use it to get rid of noise
- temperature affects both gauges but not the bridge balance
48
How are strain gauges used?
- 8 gauges allows for high sensitivity and good temperature regulation
- movement is converted into electrical signal using strain gauge
- silicon strain gauge can be used in catheter tip (can be placed into blood and give accurate readings of BP)
- disposable sensor system
- good temperature compensation
- semi conductor strain gauges
49
What are resistive displacement transducers also known as?
Potentiometers
50
How do potentiometers work?
- applied displacement -> change in resistance within sensor
- need to be calibrated and put within circuit to give linear voltage output
- tend to be larger components
51
Advantages of potentiometers
- contactless
| - can achieve excellent resolutions
52
Disadvantages of potentiometers
- limited by size constraints (overall size)
| - resolution depends on wire windings (minimum -20 micrometers)
53
Inductive Displacement Transducers
- passive
- requires signal conditioning (needs a very complex demodulator)
- require AC excitation and a conditioning circuit
- the measured quantity causes change in inductance (L)
54
2 types of inductive displacement transducers
1) simple inductive type (movement of core within coil)
| 2) two coil mutual inductance (movement of one coil relative to another)
55
What is LVDT?
linear variable differential transformer = common inductive transducer
- used by catheter tip blood pressure transducers to measurement displacement of a diaphragm
Passive as it requires AC
56
Advantages of common inductive transducer
- 3 coils with core moving differently relative to each other = high sensitivity
- also no contact = no wear/friction
57
Advantages of inductive displacement transducers in general
- light weight
| - minimal resistance to movement
58
What are capacitive displacement transducers used for?
- displacement
- displacement changes size of gap (d)
- can change amount of overlap between capacitor plates (A)
- permitivity is combined with Eo and Er
59
What are some properties of capacitive displacement transducers?
- require AC excitation and a conditioning circuit
- measured quantity causes a change in the capacitance C
- capacitance changes generally very small
- variation in capacitance must be detected accurately
- formula see slide
- inverse relationship between sensitivity and plate separation (d)
- ideal capacitor is a small plate separation and large area of plates
60
Typical sensitivities and full scale deflections of inductive displacement transducers
Sensitivites=> Approx 0.5-2.0 mV/0.001 cm displacement
| Full scale deflections=> Approx 0.01-25cm--> displacements of less < 0.1um can be detected
61
What is Eo
Dielectric constant of free space
62
What is Er
Relative dielectric constant of the insulation (between 2 charged plates of the capacitor)
63
Types of biomedical signals
- bioelectric
- bioacoustic (noises of blood flow/air)
- biomechanical (motion/displacement, pressures, flows)
- biomagnetic
- bio-optical
- bioimpedance
64
What does any system require?
- measurand (measured quantity)
- transducer (sensing element, converts 1 energy form to another)
- signal conditioning (convert transducer output into electrical form to be displayed)
- display
65
Sensor
- physical parameter to electrical output
| type of transducer
66
Actuator
- electrical signal to physical output
| type of transducer
67
Direct measurement
- measure of temperature with a thermistor for example
68
Inferential measurement
- of cardiac output (using Fick's principle)
| - measure that is estimated based on assumptions
69
Why are living systems difficult to obtain good measurements from?
- body signals interact (lots of feedback loops)
- lots of highly inaccessible sites
- cannot predict expected outcomes as highly time dependent signals
- limitations as cannot be highly invasive (patient safety, applying energy for function, must physically connect system to patient for measurement)
- signals are generally very small so need to make them readable and understandable
- signals are usually low frequency range leading to motion artefacts
- very high change of signal artefacts (movement, mains, crosstalk)
70
List of transducer characteristics
- linearity
- sensitivity
- response speed/responsiveness
- accuracy
- repeatability
- size
- ruggedness
- dynamic range
- environmental resilience
- hysteresis
- drift
- impedance
71
What is accuracy?
- closeness of output to the true value
| - usually described by how inaccurate it is using % error
72
What is precision/repeatability?
- ability to reproduce set of readings within accuracy range
| - reproduce under same conditions, same quantity, same observer etc.
73
Differences between accuracy and precision
Look at image on slide! (good image)
74
What is range/span?
What are the minimum to maximum values that can be measured by system
75
What is drift?
Baseline output variation caused by sensitivity change due to interreference
- baseline shift from 0 when switch machines off overtime
76
What is threshold?
smallest change in measurand resulting in output
| - limits capability of a transducer
77
What is saturation?
- no further change in output for given input
78
What is conformance?
closeness of calibration curve to specified curve
| - specifically for non-linear transducers
79
Types of filters
- low pass
- high pass
- band pass
- notch
Look at image from slide
