Lab Report on the Development of a ‘Transmission Finger Infrared PPG (Photo PlethysmoGraphy) Probe’
This report investigates the photoplethysmography technique and details the circuitry used to obtain the photoplethysmographic signals (PPG). PPG techniques are used to monitor the heart rate and the level of oxygen in the blood stream of a human body (oxygen saturation). Light emitting diodes (LED) emit light at different wavelengths, through a fingertip, ear lobe or toe, where the transmitted light is detected by an optical sensor. It is based on the principle where oxygenated haemoglobin has a higher absorption coefficient for infrared light than the deoxygenated haemoglobin while the deoxygenated haemoglobin has a higher absorption coefficient for red light than the oxygenated haemoglobin. The report highlights the significance of different components in a circuit that enable us to obtain a PPG signal, which include infrared emitter driver circuit to provide a constant current, an inverting amplifier and low-pass and band-pass filters to segregate the AC and DC components of the PPG signals. It also provides in detail the methods used to construct the complete circuit and how the PPG signals obtained are displayed on the two channel digital oscilloscope. The PPG signals are observed, frequency response of band pass filter is plotted and the results are discussed. Furthermore, the significance of photoplethysmography and its applications are also outlined in this report.
Table of Contents
The principle of photoplethysmography is based on the differences in the absorbance of light caused by any changes in the arterial configurations which determine the various stages within the cardiac cycle. The heart undergoes rhythmic contractions and relaxations known as systole and diastole, during the cardiac cycle, which in turn causes changes in the blood pressure in the blood vessels. During relaxing and re- filling of heart, no blood is pumped out. The arteries are specialized to serve as pressure reservoirs in order to ensure a continuous flow of blood . The elastic fibres present in the walls of arteries enable them to stretch and expand, so that it is easy to accommodate the extra blood volume during the systole. This behaviour of arteries could be matched that of balloons, where the diameter increases or decreases with the volume of blood .
Figure 1: Arteries acting as pressure reservoirs by varying cross-sectional area 
The arterial diameter increases during systole, as there is more blood in arteries during this phase. More light is absorbed by the blood as the optical path of the light increases, causing a decrease in the amount of transmitted light, thus the PPG waveform reaches its minimum peak . During diastole, the opposite happens causing the PPG waveform to reach its maximum peak. Due to this pulsatile arterial blood flow in the vessels, the PPG waveform consists of a distinguishable AC component, whereas the DC component of the waveform represents the absorption on light due to the non-pulsatile arterial blood and the tissues such as muscles, veins and bone .
Figure 2: Light absorption through tissue as a function of pulsatile blood flow 
The changes triggered during the cardiac cycle, in the arterial blood volume are therefore described as pulsations in the arterial blood flow. Various signal processing algorithm can therefore be applied to calculate the Pulse rate which corresponds to the Heart Rate and Pulse rate Variability.
The sensor probe of the Pulse oximeter consists of a photo detector and a light emitting diode. The photo detector detects the light transmitted from the LED, in form of current which is proportional to the amount of transmitted light and is converted into voltage by the trans-impedance amplifier incorporated within the pulse oximeter. Further conditioning of the detected signal is made possible by the filters and amplifiers used to extract the PPG waveform.
Methods for Light Detection
The sensors used can be classified into two types primarily based on the relative position of the LED with respect to the detector, known as transmittance mode and reflectance mode sensors .
In this mode, the detection of signal is dependent on the light that is transmitted through the tissue. A soft tissue and minimal presence of bone tissue is thus required to achieve the maximum transmittance of light, as the bone tissue reflects the light. Transmittance mode sensors are therefore limited to several peripheral locations of the body such as fingers, toes, nasal septum and earlobe, so that the tissue length is small, as the LED and the detector are designed to be placed on opposite sides of the tissue 
In this mode, the detection of signal is dependent on the light that is reflected from the tissue, therefore it can be used in various parts of body such as forehead, where there is a presence of bone tissue to facilitate the reflection of light. The sensors are thus designed so that the placement of LEDs and the photo detectors are adjacent to each other .
Figure 3: Transmittance (a) and reflectance (b) PPG probes 
Transmittance versus Reflectance
Finger sensors are the most popular probes used commercially in hospitals and clinical researches for patient monitoring. Transmittance probes are advantageous as the light is transported through the soft tissue, but they are limited to peripheral locations, where they are susceptible to inaccuracies caused by the environmental aspects, such as vasoconstriction [9,10].
A higher amplitude can be obtained from regions such as forehead and chest, where reflectance sensors are used, as the PPG signal is facilitated by the bones present in the part being monitored . The advantage of reflectance probes is that they can be used either invasively or non-invasively in many parts of the body that are not accessible by the transmittance probes . Reflectance probes are highly applicable for heart rate monitoring during compromised peripheral blood flow .
The accuracy of PPG signals is adversely affected by any motion artefacts present during measurement. Transmittance sensors are more prone to provide fluctuating readings due to motion artefacts, such as the finger or toe probes are very easily susceptible to artefacts due to any minute movement of limbs during recordings. In a study conducted by Johnston et al It was demonstrated that the reflectance sensors provide a greater stability from forehead sensors during motion, than any other sensor, therefore reflectance sensors offer a reduction in any inaccuracies that might occur due to motion artefacts . However, the reduction of motion artefacts still remain as a challenge for designing the long-term health monitors which require a high specificity . Several approaches have been made in order to attenuate the inaccuracies occurring due to motion artefacts. Some of them include the software adaptive filtering and removal or correlation of motion artefacts in simple analogue filtering . Power consumption is also considered to be an important criterion for the selection of LED‟s. The intensity of light transmitted or reflected must be high enough to be detected by the detector in order to obtain an appropriated PPG waveform. The drive current of LED and the intensity of light are directly proportional to each other. Most of the energy of LED is dissipated as heat as they are 2-10% efficient . While designing battery operated or portable pulse oximeters, the power consumption must be considered, as most of the power is consumed by the LED, for example 70% of power is consumed by the LED‟s and RF transmitter in the microcontroller pulse oximeters . In a study conducted by Savage et al, it was demonstrated that the reflectance sensor with a large area for photo detection show almost 18 times higher battery efficiency than the transmittance sensors, this is due to the fact that the current requirements for the reflectance sensors are much lesser, about 1.9-3mA, than the transmittance sensors for which it is about 19.6-46mA .
A detailed block diagram of the infrared PPG circuit was used to construct the circuit from the point of application to the point of the results being displayed.
Fig. 4: Block Diagram of the Infrared PPG System (Source: Lab Handout – MELM003)
In this section the methods to put the circuit together are discussed and it also includes the description of the role of each part of the circuit. Moreover, the calculations for the constant current driver and the filters that were used in the circuitry are shown along with an explanation of the purpose of filters in the circuit.
To conveniently build and test the circuit, a mini sized WISH Breadboard was used as that ensured a tidy and small working area and provided room for alterations if any required to make sure the circuit worked. A TENMA Handheld Digital Multimeter was used to take the Voltage and Resistance readings at various points of process of building and testing of the circuit. Furthermore, a FARNELL Triple Output Power Supply TOPS 3D was used to supply the required power to the circuit that was being built. During the later stage, a Tektronix 2-channel Digital Oscilloscope was also used to display the AC and DC PPG signals from the output of the filters as shown in the figure 4.
IRED Driver Circuit
To begin with, the Infrared Emitting Diode (IRED) was connected to the breadboard and to provide a constant current source an IRED driver circuit was built. This driver circuit consisted of a low power operational amplifier, TLO84 (Texas Instruments, Dallas, Texas, USA) and a series NPN transistor, BC184L. It also included two fixed resistors, R1 and R3, and one variable resistor, R2. The values for resistors R1, R2 and R3 were then calculated using the Voltage divider rule:
The Vin is 9V and the Vout, V1
7V, substituting the values in the equation, the following equation was obtained:
After choosing a value of resistor, R1 i.e. 1.2kΩ and solving the above equation the value for the valuable resistor, R2 was obtained i.e. 4.2kΩ.
Using the Ohm’s Law i.e.
, the value for R3 was calculated by dividing the Voltage across R3 i.e. 7V by the current coming out from the emitter pin of the transistor, iethat was to be maintained at 0.04A (40mA). The resistance of the R3 calculated was therefore 175Ω, however, a 180Ω resistor was used instead due to the unavailability of 175 ohms resistor.
Photodetector and I-V Amplifier
After building, and verifying that the current was maintained at 40mA, a photodetector I-V amplifier was added to convert the output current signal from the photodetector to a Voltage in order for it to be passed on to the PPG processing unit. Before the circuit was supplied with the power, the correct position and connection of the photodetector pins was checked as recommended.
, if RF = 120kΩ then R1 = 12kΩ
And, R1 = R2 so R2 = 12kΩ
Figure 6: Circuitry from the photodetector I-V amplifier output to the display
The output signal from the photodetector I-V amplifier consists of two different sized amplitudes combined; a large amplitude DC PPG component and a small amplitude AC PPG component. Hence why, the combination of filters as shown in figure 6 above was used in order to separate the signals into 2 independent channels, AC and DC respectively.
For the AC PPG component, a band pass filter was used to only allow a certain band of frequencies. This band pass filter has a high pass passive filter to prevent the DC PPG component to pass followed by a Butterworth 2-pole low pass active filter to attenuate higher frequencies. This band pass filter then gives out a really low amplitude AC signal, so this issue was dealt with using an inverting amplifier with a closed loop gain of 10 as can be seen in the figure 6 so the end signal that is seen on the oscilloscope is amplified. The choice of the resistors for this amplifier can be seen from the calculations in figure 6 above.
For the DC PPG component only one unity gain 2-pole, active low pass Butterworth filter was used to filter through the low frequencies. Both separate, individual AC and DC signals were then displayed on the dual channel digital oscilloscope.
The band pass filter has two stop bands and two cut off frequencies. It rejects the frequency less than 0.48Hz and the frequency higher than 19.40Hz, while it passes the band of frequencies between 0.48Hz and 19.40Hz. Thus, 0 < ω < 0.48Hz and ω > 19.40Hz are the stop bands while 0.48Hz < ω < 19.40Hz is the pass band.
The PPG circuit was connected to the dual channel digital oscilloscope after all the filters and components were in place and the observations did suggest that the DC signals appeared as expected i.e. a constant flat line signal. However, for the AC signal, the frequencies were changed on the oscilloscope to get a frequency-amplitude relationship. For each frequency value, the corresponding Voltage amplitude was recorded and were put in a table as shown below:
Table 1: Frequency-Amplitude relationship for AC PPG component
|Frequency (Hz)||Amplitude (V)|
These results in the table were then plotted on a graph with an aim to work out the actual cut off frequencies and compare it with the theoretical ones calculated in the Methods section earlier. The theoretical cut off frequencies for the band pass filter were 0.482Hz (lower cut off frequency) and 19.40Hz (higher cut off frequency). The graph is shown below:
Half power point
Lower cut off frequency
Higher cut off frequency
Figure 7: Band Pass filter Frequency response
To work out the cut off frequency, the peak voltage was taken from the graph and then a half power point was worked out i.e.
of the peak voltage. At this point, the output power reduces to half of its peak value. Using the peak value i.e. 4.2V, the half power point came out to be =
At the practical frequency response, it showed that the cut off frequencies came out to be very similar to the theoretical value i.e.
0.482Hz (lower cut off frequency) and
19.40Hz (higher cut off frequency).
How the circuit works
Relevance of the findings
Important aspects/difficulties associated with the building PPG equipment
Choice of filters
Ways of improvement in the circuit
PPG technique is used in hospitals for measuring oxygen saturation, heart rate, detecting the other haemoglobin species such as carboxyhaemoglobin and methaemoglobin using multi-wavelength analysis [17, 18]
Monitoring patients’ intravascular volume status
In mechanically ventilated patients, the dynamic indicators of fluid responsiveness (including the respiratory variations in the amplitude of photoplethysmogram) rely on cardio- pulmonary interactions . During lung inflation, a decrease in right ventricular venous return is induced by the increase in intra-thoracic pressure, leading to a decrease in right ventricular stroke volume that spreads to the left ventricle. These all factors are responsible for respiratory variations in the left ventricular stroke volume, which is detected as systemic arterial pulse pressure . The respiratory variation in the photoplethysmogram’s amplitude (ΔPOP) shows sensitivity to venous return and could be used for the accurate prediction of fluid responsiveness.
Peripheral vasodilation occurs due to the local anaesthetic-induced sympathectomy during regional anaesthesia and could be quantified using the plethysmogram. Perfusion Index (PI) has been evaluated as a good indicator of several injections given during general anaesthesia as the injections usually induce a significant decrease in PI .
Other morphological analysis
Information regarding venous pulsation could be derived using morphological analysis of the plethysmogram . False oxygen saturation readings are achieved due to the interference from the venous blood in the forehead, when using a forehead pulse oximeter . An elastic tension headband is therefore used to compress the veins in order to eliminate venous blood.
Carboxyhemoglobin and methemoglobin measurements Light transmission at two wavelengths is used to determine the arterial oxygen saturation as the ratio of oxyhaemoglobin to the total haemoglobin . Haemoglobins such as methaemoglobin, carboxyhaeoglobin and dyshaemoglobins also absorb light in the blood. Eight or more wavelengths of light could be used to ensure the concentrations of these haemoglobins in the blood noninvasively using multiple wavelength pulse oximeters .
 Sherwood, Lauralee. Human Physiology: From Cells to Systems. 4th ed. Pacific Grove: Brooks/Cole, 2001.
 Tam, Hak W and Webster, John G. “Minimizing Electrode Motion Artifact by Skin Abrasion.” IEEE Transactions on Biomedical Engineering 24 (1977): 134-139.
 Fine I and Weinreb A 1993 Multiple scattering effects in transmission oximetry Med. Biol. Eng. Comput. 31 516–22
 Webster, John G. Medical Instrumentation: Application and Design. Wiley and Sons, Inc, 1998
 Deni, Hassan, Diane M. Muratore, and Robert A. Malkin. “Development of a Pulse Oximeter Analyzer for the Developing World.” Bioengineering Conference, 2005. Proceedings of the IEEE 31st Annual Northeast (2005): 227-228.
 Webster, John G. Design of Pulse Oximeters. CRC Press, 1997.
 Burch, G. E., Cohn, A. E., and Neumann, C. (1942). A study by quantitative methods of the spontaneous variations in volume of the finger tip, toe tip, and postero-superior portions of the pinna of resting normal white adults. Am.J. Physiol., 136,433.
 Leach, R M., and D F. Treacher. “ABC of Oxygen Transport—1. Basic Principles.” BMJ 317 (1998): 1302-1306. 10 Aug. 2007
 Guyton A C 1982 Textbook of Medical Physiology 7th edn (Philadelphia, PA: Saunders) pp 345, 692
 Intaglietta M 1990 Vasomotion and flowmotion: physiological mechanisms and clinical evidence Vasc. Med. Rev. 1 101–12
 Pujary, Chirag J. Investigation of Photodetector Optimization in Reducing Power Consumption by a Noninvasive Pulse Oximeter Sensor. Diss. Worcester Polytechnic Institute, 2004.
 Ewing D J, Neilson J M and Travis P 1984 New method for assessing cardiac parasympathetic activity using 24 hours electrocardiograms Br. Heart J. 52 396–402
 Kyriacou, P A., S Powell, R M. Langford, and D P. Jones. “Esophageal Pulse Oximetry Utilizing Reflectance Photoplethysmography.” IEEE Transactions on Biomedical Engineering 49 (2002): 1360-1368. IEEE Xplore. Worcester Polytechnic Institute. 9 Aug. 2007
 Johnston, W S., P C. Branche, C J. Pujary, and Y Mendelson. “Effects of Motion Artifacts on Helmet-Mounted Pulse Oximeter Sensors.” Bioengineering Conference, 2004. Proceedings of the IEEE 30th Annual Northeast (2004): 214-215.
 Such, O, and J Muehlsteff. “The Challenge of Motion Artifact Suppression in Wearable Monitoring Solutions.” 3rd IEEE/EMBS International Summer School on Medical Devices and Biosensors (2006): 49-52.
 Savage, M, C Pujary, and Y Mendelson. “Optimizing Power Consumption in the Design of a Wearable Wireless Telesensor: Comparison of Pulse Oximeter Modes.” Bioengineering Conference, 2003 IEEE 29th Annual, Proceedings Of (2003): 150- 151.
 Wesseling K H, Settels J J, van der Hoeven G M A, Nijboer J A, Butijn M W and Dorlas J C 1985 Effects of peripheral vasoconstriction on the measurement of blood pressure in a finger Cardiovasc. Res. 19 139–45
 Wilson B C and Adam G 1983 A Monte Carlo model for the absorption and flux distribution of light in tissue Med. Phys. 10 824–30
 Michard F: Changes in arterial pressure during mechanical ventilation. Anesthesiology 2005.
 Shamir M, Eidelman LA, Floman Y, Kaplan L, Pizov R: Pulse oximetry plethysmographic waveform during changes in blood volume. Br J Anaesth 1999.
 Mowafi HA, Ismail SA, Shafi MA, Al-Ghandi AA: The efficacy of perfusion index as an indicator for intravascular injection of epinephrine-containing epidural test dose in propofolanesthetized adults.
 Bartstra, M., and Zeelenberg, H. J., (1969). The use of different drugs in circulatory care during anaesthesia; in L’AnesthesieVigile el Subvigile; Traveaux du symposium international d’Ostende. 1, III, 107.
 Zeelenberg, H. J. (1970). Maintenance of an adequate circulation during anesthesia. Arch. Chir. Neerl., XXII-IV, 247.
 Barker SJ, Curry J, Redford D, Morgan S: Measurements of carboxyhemoglobin and methemoglobin by pulse oximetry:a human volunteer study. Anesthesiology 2007,105:892-7.