**6. Biomeasurement system and instrumentation**

#### **6.1. Biomeasurement system constitution**

The biomeasurement system is shown in figure 36 to measure some biological signals such as quantity, property, or condition which are bioelectrical signal generated by muscles or the brain, or a chemical or mechanical signal that is converted into an electrical signal. Biomea‐ surement system is composed of sensor, analog processing circuit, A/D conversion, digital signal processing, output display, and data storage. A/D conversion is used in bioinstrumen‐ tation to acquire the enough system gain.

**Figure 36.** Basic biomeasurement system using sensors to measure a biological signal with data acquisition, storage and display capabilities, data transmission, along with control and feedback

With the invention of telephone and with appearance of internet, signal can be required with a device in one location, perhaps in patient's home, and transmitted into another device for transmission or storage. For example, if a biological signal from bioinstrumentation system in rural area could be transmitted into a diagnosis center in hospital, doctor would quickly judge some diseases to make patients gain an accurate treatment or diagnosis in time.

Two other components play important roles in bioinstrumentation system. The first is the calibration signal. A signal with known frequency and amplitude is applied to the bioinstru‐ mentation system at sensor's input. The calibration device permits the system components to be adjusted so that it's known that the output and input have a certain linear relationship. Without such information, it's impossible to convert the output of an instrument system into a meaningful representation of the measurand.

Another important component, a control or feedback element, is not a part of all instrument systems. These parts include pacemakers and ventilators that could stimulate the heart and lungs. Some feedback devices collect physiological data and stimulate a response— a beat or breath—when needed or are part of biofeedback systems in which patients are made aware of a physiological instrument, such as blood pressure, and uses conscious control to change the physiological response.

### **6.2. Biomeasurement circuit**

filtering of biological signal may be necessary to meet the requirement of hardware specifica‐ tions of signal conversion procedure. Continuous signal needs to be limited to a certain band of frequencies before signal can be digitized with an analog-to-digital converter, prior to

The biomeasurement system is shown in figure 36 to measure some biological signals such as quantity, property, or condition which are bioelectrical signal generated by muscles or the brain, or a chemical or mechanical signal that is converted into an electrical signal. Biomea‐ surement system is composed of sensor, analog processing circuit, A/D conversion, digital signal processing, output display, and data storage. A/D conversion is used in bioinstrumen‐

**Figure 36.** Basic biomeasurement system using sensors to measure a biological signal with data acquisition, storage

With the invention of telephone and with appearance of internet, signal can be required with a device in one location, perhaps in patient's home, and transmitted into another device for transmission or storage. For example, if a biological signal from bioinstrumentation system in rural area could be transmitted into a diagnosis center in hospital, doctor would quickly judge

some diseases to make patients gain an accurate treatment or diagnosis in time.

and display capabilities, data transmission, along with control and feedback

storing in a digital computer.

220 Advances in Bioengineering

**6.1. Biomeasurement system constitution**

tation to acquire the enough system gain.

**6. Biomeasurement system and instrumentation**

Figure 36 shows the basic elements which constitute basic bioinstrumentation system. Circuits play a very important role in bioinstrumentation system. If a bioinstrumentation needs to be developed or improved to be fit for new condition, function circuits in different blocks from figure 36 have to be respectively designed to form bioinstrumentation system with relative indexes. Among all kinds of circuits, amplifiers and A/D converters are very important component for detecting the biological signal. Hence, amplifying circuits will be only intro‐ duced here in detail, and other function circuits could be read or utilized in relative books.

**•** Bioelectric amplifier

In order to record the bioelectric potential from the body, biological amplification is always required. The simplest form is shown in figure 37 which uses a single-input amplifier. Here amplifier only amplifies one input signal which is applied in the input and the reference "earth" or "ground".

In this amplifier, the resistor *R*1 is required to allow the "bias current" to flow into the noninverting (+) input of the operational amplifier and the resistor *R*<sup>2</sup> is required to balance the resistor *R*1 so that the bias currents do not produce a voltage difference between the two inputs of the amplifier. If there is no capacitor in the positive input, amplifier in figure 37 will become a voltage follower, that's to say, the bioelectric input signal could be transmitted completely to the output. Namely, the following equation is given:

$$
u\_{im\*} = 
u\_{out}$$

Unfortunately, in circuit shown in figure 37, the resistor *R*1 defines the maximum input impedance of the amplifier. The input impedance is an important consideration in bioelectric amplifiers because it can cause attenuation of a signal which is derived from electrodes with high impedances. For example, if the two electrode impedances were 10kΩand the input impedance of the amplifier was 1MΩ, then 1% of the signal would be lost by attenuation of two electrodes. The impedance presented by electrodes is termed the source impedance which has to be very much less than the input impedance of amplifier. Source impedance is very important when we consider differential amplifier shortly.

**Figure 37.** A simplest single-input amplifier

**Figure 38.** A differential amplifier

There is also capacitor introduced in figure 37 in series with the input signal. Capacitor *C* blocks any DC (direct current) signal by acting as a high-pass filter with the resistor *R*1. This function is usually referred as AC (alternative current) coupling. AC coupling will also cause some attenuation of the signal which may be important. We could determine the attenuation of signal by the following equations:

$$\mu\_{out} = \mu\_- = \mu\_{\ast \prime} \quad \frac{\boldsymbol{u}\_{in}}{R\_1 + \frac{1}{j\omega \boldsymbol{\alpha} \boldsymbol{\Sigma}}} = \frac{\boldsymbol{u}\_+}{R\_1 \cdot \prime}, \quad \text{then } \frac{\boldsymbol{u}\_{out}}{\boldsymbol{u}\_{in}} = \frac{R\_1}{R\_1 + \frac{1}{j\omega \boldsymbol{\alpha} \boldsymbol{\Sigma}}} = \frac{R\_1(R\_1 + j/\omega \boldsymbol{\Sigma})}{R\_1^2 + 1/\omega^2 \boldsymbol{\Sigma}^2}.$$

$$\left| \frac{\boldsymbol{U}\_{out}}{\boldsymbol{U}\_{in}} \right| = \frac{1}{\sqrt{1 + 1/\left(\omega^2 R\_1^2 \boldsymbol{\Sigma}^2\right)}} = \frac{1}{\sqrt{1 + \omega\_0^2 / \omega^2}}, \quad \text{where } \omega\_0 = 1/R\_1 \boldsymbol{\Sigma}.$$

If capacitor C is 1μF and resistor *R*1 is 1MΩ, then the attenuation of a 1Hz signal will be 1.25%. This is perhaps a significant attenuation for ECG which has considerable energy at 1Hz.

Unfortunately, even with capacitor C added, this type of amplifier is not suitable for recording small bioelectric signal because of interference from external electric fields. An electric electrode has to be connected to the amplifier via a wire and this wire is exposed to interfering signals. However, the interference will only appear on the input wire to the amplifier and not on the "ground" wire which is held at zero potential. An elegant solution to this problem is to utilize differential amplifier as shown in figure 38. The input of the type of amplifier has three connections marked '+', '-' and 'ground'.

**•** Differential amplifier

two electrodes. The impedance presented by electrodes is termed the source impedance which has to be very much less than the input impedance of amplifier. Source impedance is very

There is also capacitor introduced in figure 37 in series with the input signal. Capacitor *C* blocks any DC (direct current) signal by acting as a high-pass filter with the resistor *R*1. This function is usually referred as AC (alternative current) coupling. AC coupling will also cause some attenuation of the signal which may be important. We could determine the attenuation of

important when we consider differential amplifier shortly.

**Figure 37.** A simplest single-input amplifier

222 Advances in Bioengineering

**Figure 38.** A differential amplifier

signal by the following equations:

<sup>|</sup> *Uout Uin*

*uout* <sup>=</sup>*u*<sup>−</sup> <sup>=</sup>*u*+, *uin*

*R*<sup>1</sup> +

<sup>|</sup> <sup>=</sup> <sup>1</sup> 1+1 /(*ω* <sup>2</sup>

1 *jωC* = *u*+ *R*1

> *R*1 2 *C* <sup>2</sup> )

, then

*uout uin*

<sup>=</sup> <sup>1</sup> 1 + *ω*<sup>0</sup>

<sup>=</sup> *<sup>R</sup>*<sup>1</sup> *R*<sup>1</sup> +

1 *jωC* =

<sup>2</sup> / *<sup>ω</sup>* <sup>2</sup> , where *<sup>ω</sup>*<sup>0</sup> =1 / *<sup>R</sup>*1*C*.

*R*1

*R*1(*R*<sup>1</sup> + *j* / *ωC*)

<sup>2</sup> + 1 / *ω* <sup>2</sup>

*C* <sup>2</sup>

In figure 38, the signal which we wish to record is connected between the '+' and '-' points. Now both inputs are exposed to any external interfering electric fields so that the difference between the '+' and '-' input will be zero. This will not be quite true because the electric fields experienced by the two input wires may not be exactly the same, but if the wires are run close together then the difference will be small. Differential amplifier is not perfect in that even with the same signal applied to both inputs, with respect to ground; a small output signal can appear. This imperfection is specified by the common mode rejection ratio or CMMR. An ideal differential amplifier has zero output when identical signals are applied to the two '+' and '-' inputs. CMMR could be defined as the following equation:

$$\text{CMMR} = 20 \log \left[ \frac{\text{signal} \cdot \frac{\text{gain}}{\text{common} - \text{model}}}{\text{common} - \text{model}} \right]$$

Where, signal and common-mode gains are given by

$$\begin{aligned} \text{signal} \quad \text{gain} &= \frac{\mathcal{U}\_{out}}{\mathcal{U}\_{in}} = \frac{\mathcal{U}\_{out}}{\mathcal{U}\_{+} - \mathcal{U}\_{-}} \\ \text{common} &= \text{mode} \quad \text{gain} = \frac{\mathcal{U}\_{out}}{\mathcal{U}\_{cm}} = \frac{\mathcal{U}\_{out}}{(\mathcal{U}\_{a} + \mathcal{U}\_{b})/2} \end{aligned}$$

In practice, common-mode voltage *Ucm* can be as large as 100mV or even more. In order to reject this signal and record a signal Vin as small as 100μV, a high CMMR is required. If we wish the interfering signal to be reduced to only 1% of output voltage then

$$required - signal \quad gain = \frac{\mathcal{U}\_{out}}{\mathcal{U}\_{in}} = \frac{\mathcal{U}\_{out}}{\mathcal{U}\_{+} - \mathcal{U}\_{-}} = \frac{\mathcal{U}\_{out}}{100\,\mu V}$$

$$required - \text{CM} \quad gain = \frac{\mathcal{U}\_{out}}{\mathcal{U}\_{out}} = \frac{\mathcal{U}\_{out}}{1\% \,\mathcal{U}\_{in}} = \frac{\mathcal{U}\_{out}/100}{100\,mV}$$

Hence, the required CMMR could be given as:

$$CMMR = 20\log\left[\frac{\mathcal{U}\_{out}/0.1mV}{\mathcal{U}\_{out}/100^2mV}\right] = 100dB$$

In fact, it's not always easy to achieve a CMMR of 100dB. As we have known, electrode source impedances have a very significant effect on CMMR and hence electrode impedance affects noise rejection.

Of course, in detecting biosignals, the AC coupling shown in figure 37 and figure 38 degrades the performance of the amplifiers. If the input impedance and bias current of amplifiers is sufficiently high, then they can be connected directly to the input electrodes, without produc‐ ing electrode polarization. Furthermore, DC offset will occur from the electrode contact potentials, but if the amplifier gain is low (<10) DC offset will be not a significant problem. The offset can be removed by AC coupling at later stage.

However, there are some safety arguments against the use of DC coupling. If a fault arises in the operational amplifier, then it's possible for the power supply to be directly connected to the patient and so give rise to a hazard. DC currents will cause electrolysis and result in tissue necrosis. AC coupling could avoid this problem and is often used. Nonetheless DC coupling is also often used in biomedical field.

#### **6.3. Bioinstrumentation design**

The purpose of using bioinstrumentation is to monitor the output of a sensor or sensors and to extract some useful information from signals that are produced by sensors.

Acquiring discrete-time signal and storing this signal in computer memory from a continuoustime signal is accomplished with analog-to-digital (A/D) converter. After analog signals have been processed which are based on analog filters such as low-pass or high-pass filters, A/D converter uniformly samples the continuous-time waveform and transforms it into a sequence of numbers, one every *tk* seconds. The A/D converter also transforms the continuous-time waveform into a digital signal, which is converted into computer words and stored in computer memory. To adequately capture the continuous-time signal, the sample frequency has to be carefully selected to ensure any signal information is not lost. The minimum sampling frequency is twice the highest frequency content of the signal based on the sampling theorem from communication theory. In reality, we often adopt the sampling frequency from five to ten times the highest frequency content of the signal so as to achieve better accuracy by reducing aliasing error.

**•** Biological signal categories in human body

The electrical, chemical and mechanical activity that occurs during this biological event often produces signals that could be detected and analyzed. Biological signals are the record of a biological event such as a beating heart or a contracting muscle. Hence, biological signals contain useful information which could reflect human's activities and physiology, that's to say, biological signal could be used for biomedical diagnosis. Biological signals are classified into bioelectric signals, biomagnetic signals, biochemical signals, biomechanical signals, bioacoustic signals and biooptical signals.

Nerve and muscle cells generate bioelectric signals that are the result of electrochemical changes within and between cells. When plenty of cells are stimulated, an electric field is then generated that propagates through biological tissues. These changes in extracellular potential may be measured on the surface of tissue or organism by using surface electrodes. The electrocardiogram (ECG) is an example of this phenomenon. Different organs in body, including the heart, brain, lungs, and liver, also generate weak magnetic fields that could be detected with magnetic sensors. The strength of magnetic field is much weaker than the corresponding physiological bioelectric signal. Magnetic sensors could be used to detect biomagnetic signals. Magnetocardiography (MCG) is a specific example of such phenomenon.

Biochemical signals contain information about changes in the concentration of various chemical agents in the body. The concentration of various ions such as calcium and potassium in cell can be measured and recorded. Oxygen sensor is used to detect oxygen concentration in body. Mechanical functions of biological systems, including motion, displacement, tension, force, pressure and flow, also produce measurable biological signals. Blood pressure sensor is a measurement of the force that blood exerts against the walls of blood vessels. Change in blood pressure can be recorded as a waveform by blood pressure sensor. Bioacoustics' signals are a special subset of biochemical signals which involve vibrations. Many biological events could produce acoustic noise. For example, the flow of blood through the valves in the heart can be used to determine whether motion is operating properly. Besides these, the respiratory system, joints and muscles could also produce bioacoustic signals that propagate through the biological medium and can be often measured at the skin surface by acoustic sensors. Bioop‐ tical signals are generated by the optical or light induced attributes of biological systems. Biooptical signals can occur or be introduced to measure a biological parameter with an external light medium such as the measurement of health of a fetus by red and infrared light.

**•** Noise

*CMMR* =20log

offset can be removed by AC coupling at later stage.

is also often used in biomedical field.

**6.3. Bioinstrumentation design**

reducing aliasing error.

**•** Biological signal categories in human body

noise rejection.

224 Advances in Bioengineering

*Uout* / 0.1*mV Uout* / <sup>100</sup><sup>2</sup>

In fact, it's not always easy to achieve a CMMR of 100dB. As we have known, electrode source impedances have a very significant effect on CMMR and hence electrode impedance affects

Of course, in detecting biosignals, the AC coupling shown in figure 37 and figure 38 degrades the performance of the amplifiers. If the input impedance and bias current of amplifiers is sufficiently high, then they can be connected directly to the input electrodes, without produc‐ ing electrode polarization. Furthermore, DC offset will occur from the electrode contact potentials, but if the amplifier gain is low (<10) DC offset will be not a significant problem. The

However, there are some safety arguments against the use of DC coupling. If a fault arises in the operational amplifier, then it's possible for the power supply to be directly connected to the patient and so give rise to a hazard. DC currents will cause electrolysis and result in tissue necrosis. AC coupling could avoid this problem and is often used. Nonetheless DC coupling

The purpose of using bioinstrumentation is to monitor the output of a sensor or sensors and

Acquiring discrete-time signal and storing this signal in computer memory from a continuoustime signal is accomplished with analog-to-digital (A/D) converter. After analog signals have been processed which are based on analog filters such as low-pass or high-pass filters, A/D converter uniformly samples the continuous-time waveform and transforms it into a sequence of numbers, one every *tk* seconds. The A/D converter also transforms the continuous-time waveform into a digital signal, which is converted into computer words and stored in computer memory. To adequately capture the continuous-time signal, the sample frequency has to be carefully selected to ensure any signal information is not lost. The minimum sampling frequency is twice the highest frequency content of the signal based on the sampling theorem from communication theory. In reality, we often adopt the sampling frequency from five to ten times the highest frequency content of the signal so as to achieve better accuracy by

The electrical, chemical and mechanical activity that occurs during this biological event often produces signals that could be detected and analyzed. Biological signals are the record of a biological event such as a beating heart or a contracting muscle. Hence, biological signals contain useful information which could reflect human's activities and physiology, that's to say, biological signal could be used for biomedical diagnosis. Biological signals are classified

to extract some useful information from signals that are produced by sensors.

*mV*

=100*dB*

Measurement signals are always corrupted by noise in the bioinstrumentation system. Interference noise occurs when unwanted signals are introduced into systems by external sources such as telephone magnetic wave, power line and transmitted radio. Interference noise needs to be effectively reduced by careful attention to the circuit wiring configuration to minimize coupling effect.

Interference noise is introduced by power lines, fluorescent lights, AM/FM radio broadcasts, computer clock oscillator, laboratory equipment and cellphone. Electromagnetic energy radiating from noise source is injected into the amplifier circuit or into the patient by capacitive or inductive coupling. Even action potentials from nerve conduction in the patient generate noise at the sensor/amplifier surface. Filters are also used to reduce the noise and to maximize the signal-to-noise(S/N) rate at the input of the A/D converter.

Low frequency noise could be eliminated by high-pass filter with the cutoff frequency set above the noise frequency. High frequency noise could be reduced by low-pass filter with the cutoff frequency set below the noise frequency and above the frequency of biological signal which is being monitored. Power line noise is a very difficult problem in biological monitoring because the 50-or-60-Hz frequency is usually at the range of biological signal which could be monitored. Band-stop filters are commonly used to reduce the power line noise. The notch frequency in the band-stop filters is set to the power line frequency of 50 or 60Hz with the cutoff frequency located a few Hertz to either side.

The second type of noise is called inherent noise. Inherent noise arises from random processes that are fundamental to the operation of circuit's elements and thus is reduced by a good circuit design practice. While inherent noise is reduced, it can be never eliminated. Low-pass filters are used to reduce high-frequency components. However, noise signals within the frequency range of biological signals being amplified cannot be eliminated by this filtering approach.

**•** Computer

Computer is a main device which is used to display the biological signals being monitored. However some low or high level languages such as machine language, FORTRAN, visual C+ +, MATLAB or LabView, have to be used to realize the operation on the acquisition data from biological body. When computers are used to acquire physiological data, programming instruction tell computer when acquisition data should begin, how often samples should be taken from how many sensors, how long acquisition data should continue, and where the digitized data should be stored. The rate at which a system acquires sample depends on the speed of computer clock's frequency and the number of computer instruction that could be completed in order to realize a sample. Of course, some computers are utilized to control the gain on the input amplifiers so that biological signals could be adjusted during data acquisition. In other systems, the gain of data acquisition has to be adjusted.
