Optical Heart Rate Monitor: Design and Implementation - Prof. Gary R. Swenson, Study Guides, Projects, Research of Electrical and Electronics Engineering

Download Optical Heart Rate Monitor: Design and Implementation - Prof. Gary R. Swenson and more Study Guides, Projects, Research Electrical and Electronics Engineering in PDF only on Docsity! OPTICAL HEART RATE MONITOR by Doris Chan Tomasz Bala Brian Sobocinski ECE 345 TA: Shao Hsia May 1, 2001 Project # 36 ABSTRACT This project combines hardware and software to determine the pulse of a patient and display it on a user friendly LabVIEW display. The optical heart rate monitor uses a Honeywell sensor that produces an ac signal for every change in skin reflectance. The amplifier circuit then amplifies the signal and filters so that only select frequencies that correspond to a human heart rate remain. A digital version of the filtered signal is then sent to a PC via a NI-DAQ. The signal is then analyzed and processed by LabVIEW software. The interface allows cardiologist to analyze the patient’s heart rate and blood pressure. Furthermore, LabVIEW software allows a physician to record the patient’s heart rate at regular intervals for more in depth analysis. ii 2. DESIGN PROCEDURE 2.1 Sensor Theory The Honeywell HLC1395-001 is an optical reflective sensor that provides the input for the heart rate monitor. It functions based on the change in contrast that occurs in the patients skin from the expansion and contraction of capillaries as blood circulates through the body. Each cycle of change in skin contrast corresponds to a heartbeat. The HLC1395-001 is a miniature infrared sensor that operates by detecting reflective objects. It is comprised of both an infrared emitter and a phototransistor in a single package as shown below in Figure 2.1. The package includes a GaAs infrared emitting diode (IRED) and an NPN transistor. A schematic of the sensor is also shown below in Figure 2.2. Figure 2.1 Sensor Diagram Figure 2.2 Optical Sensor The sensor design is highly sensitive, and capable of detecting a 1 % change in skin contrast. Unfortunately, the sensitivity of the design leads to compromised reliability. Without proper care, the sensor fails to give accurate readings. It was found that the sensor must be cleaned with isopropyl alcohol at the beginning of each work session. In addition to the initial cleaning, it was sometimes advantageous to clean the sensor once every hour or two. 2 2.2 Hardware Design Procedure The hardware design for this project consisted of building filter and an amplifier circuits. The Honeywell sensor produces an ac signal of around 500 V for every 1% change in skin reflectance and the rest of the circuit amplifies and filters signals. A high-gain adaptive filter rejects ambient optical and electrical noise and presents a cleaned-up signal to a comparator so that it can extract a digital pulse-rate signal. The circuit has a number of component parameters that must be met in order for it to operate according to design specifications. An important element of this design is that the first stage amplifier (preamplifier) must have a high input impedance and low input bias current. The high input impedance is necessary in the circuit to minimize loading effects. Loading occurs when the gain of the second stage amplifier affects the gain of the preamplifier. Extremely low input impedance can cause loading, which can result in a distortion of the output signal. Another factor that can cause the distortion of the output signal is that the first stage amplifier must have a gain that falls within a desired range. The output signals from the sensor are usually on the order of microvolts and they must be amplified to a degree such that the noise does not interfere with the desired signal. Amplifying the signal by a factor that is below the desired range will not allow us to adequately filter the signal. If the amplifying factor is too high, the noise will also be too great. The normal human hearts usually beat at rates within the 4:1 range of 50 to 200 beats per minute (bpm) around 0.83 Hz to 3.3 Hz. Consequently, the second stage amplifier and filter circuit should contain a bandpass filter suited to pass these frequencies, but to cut off all others. This frequency response can be achieved by adjusting resistor and capacitor values according to equation (1) on the following page. 3 fc = (2RC)-1 (1) The design of the amplifier circuit was modeled after the preamplifier presented in Electronic Design by Stephen W. Woodward [5] (see figure 2.3) Figure 2.3 Amplifier Circuit The actual design of the amplifier circuit in the heart rate sensor circuit differs slightly than the one presented by Woodward. Since the design uses reflective sensor the values of the resistors and capacitors had to be adjusted in order to reduce electromagnetic noise and to 4 Figure 2.4 PCI-MIO-16E-1 Pin layout 2.3.2 LabVIEW Software LabVIEW is a graphical programming language that is used for communication with the data acquisition hardware. LabVIEW software contains comprehensive libraries for data collection and analysis. The front panel is GUI that combines user input with graphical programming language to display the desired output. The graphical language uses a simple set of graphical symbols rather than textual language for easier debugging. Therefore, the ECG signal is analyzed in LabVIEW and not another language. LabVIEW is capable of measuring real time data from the DAQ board. 7 The first program is design to use a digital input channel from the DAQ board to measure the TTL signal. However, due to difficulties from the bit conversion and the display for the graph of the TTL signal, the second program is written. This second program takes data from the analog Channel and measures the frequency of TTL signal. Then, it calculates the heart rate, and displays a simulation of the analog heart beat waveform on the front panel. 8 3. DESIGN DETAILS 3.1 Hardware Design Details The amplifier circuit was designed so that it would pass frequencies from 0.9 Hz – 3.5 Hz. To construct a bandpass filter meeting these parameters, the actual resistor and capacitor values were measured to achieve the desired cutoff frequencies. The design was implemented with a combination of a low-pass and a high-pass filter. These values are listed below in Table 3.1 and Table 3.2. Table 3.1 Low-pass filter data R1 C3 Frequency 880 k 0.0504 F 3.588 Hz Table 3.2 High-pass filter data R3 C2 Frequency 750 k 0.230 F 0.923 Hz 9 edge component of the signal is determined from the Power Spectrum. VI. This is equal to the frequency that is used in the calculation of the heart rate. 3.3.3 Determining Pulse The frequency measurement subprogram does not yield an accurate measurement of the frequency all the time because it is too varied over a few seconds. For this reason, the program must calculate the average frequency. This part of program uses a "for loop" to make ten arrays which is written with a double "for loop". The inner for loop is the number of the columns, and the outer for loop is number of the rows for the array of frequency data. Summing all the array data yields an average frequency that is equal to the sum of the frequency divided by the number of counts in the loop. After getting the average frequency, the bpm calculation is performed as in equation (5): bpm = 60*1/f (5) The front panel display shows the average frequency and bpm. 3.3.4 Heart Rate Waveform This part of the subVI uses a sinc pulse from the waveform generation palette. The delay is set for zero, and the amplitude default is set as 1. Delta t is the sampling interval and set to 0.1 sec. for the width of the main sinc lobe. The number of samples is set to average frequency of the bpm because it generates one pulse. The frequency for the heart rate is the major peak to peak of the signal. The program uses a for loop to generate the required pulses and shows the heart rate graph on the front panel. 12 4. DESIGN VERIFICATION 4.1 Hardware Design Verification After completing the circuit design, the waveform was tested at various points in the circuit with an oscilloscope and voltmeter. The digital output measured in a trial run is shown below in Figure 4.1. Figure 4.1 Digital Output It was discovered that the filters did not originally perform as desired. It was found that the original values allowed too much high frequency interference. After adjusting to the final component values given in Tables 3.1 and 3.2, the circuit operated as expected. A variable resistor was used to aid in the determination of the cutoff values. 13 4.2 Software Design Verification After each major section of the software was completed, a patient’s finger was applied to the sensor. The first component of the software to be completed was the real time graph of the input waveform. As previously mentioned, the first part of the program takes the scan rate at the minimum of 1000 scan/sec. If the scan rate is less than this, the output results display erroneous data and the frequency measurement of the TTL signal is not accurate for second part of the program. The third and fourth parts are calculation and generation of sine wave. For this part of the testing process, the program uses random number generation instead of real data from DAQ board. Random numbers were used as frequencies to build the array. Then the average frequency and bpm could be calculated. After that, a heart rate display is generated on the graph. 14 features to record peak heart-rate would make this project useful as a fitness device that would be good for sale in a sporting goods store. The optical heart-rate sensor constructed this spring successfully incorporated many of the original design features, and opened many possibilities for future development that could possibly be explored by subsequent ECE 345 design teams. 17 REFERENCES [1] B. Leece and S. Nikiforos, "Heart rate monitor and data acquisition system," presented at the Department of Electrical and Computer Engineering at the University of Illinois, Urbana- Champaign, Aug. 2, 1999. [2] National Instruments, LabVIEW User Manual, National Instruments Corporation, 1996. [3] D. J. Stadtmiller, Electronics Project Management and Design. New Jersey: Prentice Hall, 2001, pp. 14-90. [4] J. G. Webster, Medical Instrumentation: Application and Design. 3rd Ed. Philadelphia: W.B. Saunders Company, 1998. [5] W. S. Woodward, “Build Your Own Optical Heart-Rate Sensor,” Electronic Design, pp. 104-106 December 15, 1997. 18