Electrocardiogram (ECG or EKG) is the most common way to identify various ailments, especially when the ailment is related to the heart. To perform an ECG, the medical personnel places the leads on the patient’s skin. The leads measure the heart’s electrical activity of one heartbeat cycle and record it as a continuous line tracing on paper to produce a graph. The ECG signal may indicate:
- Cause of unexplained chest pain
- Cause of symptoms of heart disease
- How well medicines are working for the patient
- How well implants such as pacemakers are working
- Other heart-related disease
Today, there is a wide array of cardiac equipment that displays and interprets ECG signal patterns, and medical equipment designers need a flexible, reliable, and accurate way to seamlessly generate ECG signal patterns to verify their designs (Figure 1).
ECG Signal Generation
In this scenario, a maker of cardiac monitoring equipment needs to test its latest design’s ability to capture and interpret ECG signals. The company’s design engineers want to simulate gradual changing sequences of normal and abnormal ECG signals to test and tune the design’s input signal and condition hardware and firmware interpretation algorithms to ensure the design does not produce false positives or life-threatening false negatives. A few key factors need to be taken into consideration when generating ECG signal, such as:
- Low-amplitude signal ±1mV or ±0.5mV
- Large offset due to half potential developed at electrode
- System noise (50/60Hz) from the power lines that forms the common mode signal
- Complex ECG signal pattern
The firmware design of the ECG equipment can resolve the offset developed by electrode and common mode signal from power lines. The designer needs to use firmware to remove the power line noise, implement a filter to remove the electrode offset, and correct noise.
This article focuses on two considerations of ECG signal generation: the clean, low-amplitude signal generation; and the complexity of the ECG pattern generation. The best solution for this is to use an Arbitrary Waveform Generator (AWG).
Real-world cardiac signals typically are very low in amplitude — often only a couple of millivolts or even less. This poses a problem for simulation using AWGs because typically their lowest amplitude setting is between 10 mV and 1 mV. When they are used at their lowest amplitude, the AWG’s signal-to-noise ratio can become a problem. One way to overcome this drawback is to use a voltage divider at the output of the AWG. Since ECG signals are at such low frequencies, the divider only needs resistors, as reactive effects can be ignored. When constructing the voltage divider, it is important to remember that the amplitude accuracy of the divider’s output signal depends on the precision of the resistors used in the divider. For example, a voltage divider that uses a 10-kohm resistor and a 10- ohm resistor will reduce amplitude of 1 V down to 1 mV, as shown in Figure 2.
Complex ECG patterns
There are various methods to create and store an ECG on an AWG, including:
- User can use a digitizer or an oscilloscope to capture the actual ECG signal from a patient and upload the points to the AWG. Most modern AWGs are able to support this.
- User can use mathematical software such as MATLAB. 3. The AWG has the built-in typical ECG waveform.
In a real-life scenario, the ECG waveform is not standard or typical; rather, it is a dynamic signal that varies from one to another, and seamless transition from one complex pattern to another is an important feature that a designer would desire. In addition, arb sequencing ability is very important to the designer. This allows an actual ECG waveform cycle to run continuously and further stimulate an actual ECG signal for the designer to test the ECG equipment (Figure 3).
In the medical industry, a precision oscilloscope is very important test equipment that the ECG designer needs for their design and verification phases. The oscilloscope needs to be able to capture, display, and analyze the ECG patterns.
Measuring Very-Low-Amplitude Signals
One of the key test challenges for a test engineer is to capture the ECG signal, especially in a noisy environment and most importantly, the low amplitude of the signal. Typically, the amplitude value is in the 0.5 mV to 1 mV range for heart rate between 60 beats per minute and 100 beats per minute. Hence a designer normally uses an amplifier to strengthen the signal, set an offset to the scope channel, and even use waveform averaging to capture an accurate ECG signal. Waveform averaging is not suitable for noise reduction due to the fact that the ECG signal is not repetitive, and the signal needed is too weak and no signal can be observed (Figure 4).
The best solution is to select an oscilloscope with a fast waveform update rate in order to capture the intermittent signal anomalies. The faster the waveform update rate, the shorter the “dead time” or the time the oscilloscope misses the signal completely. Hence during acquisition mode, extra samples are averaged in order to reduce random noise and produce a smoother trace, and effectively increase vertical resolution, as shown in Figure 5.
A mask test is one of the best ways to verify a healthy ECG signal against the measures signal. The oscilloscope’s display defines the region where the waveform must remain in order to comply with the golden signal. This helps a designer to visually identify any abnormality in the measured signal instantly.
Solution for ECG Signal Generation
State-of-the-art AWGs have arb sequencing capability and deep arb memory to provide an excellent solution for ECG simulation. The sequencing provides the ability to seamlessly transition through various ECG signal conditions. The deep arb memory complements the sequencing capability to allow storage of a large library of ECG waveforms. This allows an engineer to add subtle changes from waveform to waveform for high-resolution testing of designs.
An oscilloscope is the best tool in analyzing very-low-amplitude signals (such as ECG signals) with superb accuracy and suitability for medical application. Designers can analyze the full spectrum of the signal in high-resolution acquisition. The signal analysis is also economical in that it does not require additional accessories to achieve high vertical resolution for the ECG signal amplitude. In addition to this, masking the golden signal allows designers to identify anomalies in the ECG pattern with ease.
This article was written by Kah-Meng Chew and Doris Lau of Keysight Technologies, Santa Rosa, CA. For more information, Click Here.