Lab Report 2: Electrocardiography (BENG355L, Fall 2021)

Blue text indicates questions that require graphs pasted in.

1. Background (15pts total)

1.1. The P-wave, corresponding to atrial depolarization, has a positive magnitude. So

why does the T-wave, corresponding to ventricular repolarization, also have a positive
magnitude? (5pts)

- Let’s assume a Lead I orientation where the positive end is on the left arm and the
negative end is on the right arm. Under this orientation, the P-wave has a positive
magnitude because it reflects the conduction across the atria that moves toward
the positive electrode. Using similar logic, we can determine that the T-wave has a
positive magnitude because the free left ventricular wall repolarizes from the
epicardium to the endocardium. The repolarization here produces a negative
intracellular current that is also moving away from the positive electrode, and the
combination of these two negative potentials forms a positive deflection.

1.2. How does the electrical activity of the heart reflect its mechanical properties?
Describe the relation of different electrical events to what occurs mechanically
throughout a cardiac cycle. (5pts)

- The electrical activity of the heart is closely correlated with the different steps of
the cardiac cycle. During the P wave, atrial depolarization occurs and acts as the
precursor to ventricular depolarization. Thus, the PR interval corresponds to the
time lag between atrial depolarization and ventricular depolarization, during
which the ventricles fill in preparation for ventricular systole and the atria carries
out atrial systole. Moving forward, the QRS interval is effectively where
ventricular depolarization occurs, allowing for isovolumetric contraction of the
heart. At the end of the QRS interval, the aortic valve opens, and the heart ejects
blood. The ejection process aligns with the T wave, which is when ventricular
repolarization occurs, resetting the system for the next iteration of the cardiac
cycle.

1.3. Give two examples of how the ECG is used as a diagnostic tool. (5pts)

- ECGs are extremely versatile as diagnostic tools. For example, ECGs can be used
to identify abnormalities in heart rate or beat rhythm such as tachycardia
 (abnormally fast heart rate) and ventricular fibrillation (uncoordinated ventricular
contractions). In addition, ECGs can be utilized to evaluate potassium levels in
people and thus diagnose conditions such as Hyperkalemia and Hypokalemia.

2. Time Analysis (35pts total)

2.1. For the data from Lead I:

2.1.1. Paste in the graph of the ECG trace. Measure the duration of each portion
of the wave (P-wave duration, QRS complex duration, T-wave duration, PR
interval, QT interval, ST interval, PR segment, ST segment). Using data cursor
and tabulating results is likely the easiest way. (9pts)

2.1.2. Measure the beat frequency. (4pts)

Beat frequency = 1.6988 s-1

2.1.3. Create a file containing one beat (P-QRS-T complex). Import the data into
a spreadsheet. Consider the PQ segment as baseline and measure the maximum
amplitude and peak to peak values for each portion of the wave, factoring in the
amplification factor so that the values are for the actual body signals. Report
results in a table and summarize observations. (5pts)
The P, QRS, and T waves are all present as expected in this Lead I ECG.

2.2. Paste in the graph of the ECG trace from Lead II. Repeat 2.1.3 for Lead II. (6pts)
The P, QRS, and T waves are all present as expected in this Lead II ECG. Compared to Lead I,
however, the Q and S waves are of smaller amplitude.

2.3. Paste in the graph of the ECG trace from Lead III. Repeat 2.1.3 for Lead III. (6pts)
The QRS waves are all present as expected in this Lead III ECG. There is no distinguishable P
wave in this ECG due to there being much more noise than is present in Lead I and II. In
addition, the T wave is inverted.

2.4. Discuss the values obtained. Compare to nominal values. (Refer to the chart of nominal
values and the ECG wave diagram in the preparation portion of the lab manual.) (5pts)

- All of the max amplitude values for each wave and lead type were within the normal
range of nominal values. The one major difference is that we had an inverted T wave for
Lead III, so the value for our max amplitude was negative. Information in literature
suggests that inverted T waves do occasionally occur with the Lead III configuration, so
our results are not unreasonable.

3. Find the electrical axis of the heart (20 pts total)

3.1. Calculate the area under the QRS complex for Leads I and II. (5pts)

Lead I area: 0.0305 volts*sec

Lead II area: 0.0446 volts*sec

 3.2. The 'mean QRS vector' lies in the frontal plane of the chest. Its component in the
Lead I direction is proportional to the area under the QRS complex for that lead. The
same is true for the other leads. (Refer to the preparation part of the lab manual for a
more detailed explanation of plotting the cardiac vector into the Einthoven triangle).
Draw the Einthoven triangle and show the orientation of the mean QRS vector using your
data from part 3.1. (5pts)

3.3. The mean QRS vector roughly corresponds to the anatomical longitudinal axis of the
heart. Is the orientation you found in part 3.2 reasonable? (5pts) Why don't you need to
use the Lead III data? (5pts)

- The orientation we found in part 3.2 is reasonable because it points down and to the right
(anatomical left) from the center of the plane of the chest. This is the orientation we
would expect because the QRS vector also corresponds to the mean direction of
ventricular wave depolarization, which we know also moves toward the anatomical left
 and the positive lead. We don’t need to use Lead III data because it is redundant and
would give us the same intersection point as above.

4. Signal Averaging (30pts total)

4.1. For the 10 beat data you collected in Section 1.8, divide it into ten beats, each
starting at the R wave. Obtain the MEAN composite signal as: (5pts)
𝑁
1
𝐶𝑚𝑒𝑎𝑛(𝑛) = 𝑁
∑ 𝑥𝑘(𝑛)
𝑘=1

for 0 ≤𝑛≤𝐿 − 1 where 𝑥𝑘, 𝑘 = 1, …, 𝑁 are noisy epochs of length 𝐿.

Show a graph of the mean composite signal for a single beat.

4.2. Obtain the MEDIAN composite signal as: (5pts)

{
𝐶𝑚𝑒𝑑𝑖𝑎𝑛(𝑛) = 𝑚𝑒𝑑𝑖𝑎𝑛 𝑥1(𝑛), 𝑥2(𝑛), …, 𝑥𝑁(𝑛) }
for 0 ≤𝑛≤𝐿 − 1 where 𝑥𝑘, 𝑘 = 1, …, 𝑁 are noisy epochs of length 𝐿.

Show a graph of the median composite signal for a single beat.

4.3 Compare the results. (5pts)

- For the most part, the median and mean composite signals are quite similar to
each other. For instance, the peaks and valleys of the R, S, and T waves are all at
about the same points voltage-wise between the two signals. What ultimately
separates the two signals, however, is that the median composite signal practically
does not show the P wave at all, whereas the first half of the P wave is visible in
 the mean composite signal. In addition, while it is difficult to see from the pasted
images above, if you zoom in, it becomes evident that the median composite
signal is slightly more noisy than the mean composite signal.

4.4 Paste in a graph of an ECG trace of noisy data, either from a poor signal you acquired
in 1.3 or by applying erroneous filtering or processing. For the noisy data, repeat 4.1 and
4.2. (showing new graphs of median and mean signals) (5pts) Compare the effect of the
artifact on the results. (5pts) Which method (mean or median) gives a better composite
signal when there is a single glitch in one of the beats? (5pts)
- Even with the addition of noise, the mean composite signal still produced a better signal
than the median composite signal. This is because while both signals exhibit a fair
amount of noisiness, the median composite signal still did not appear to have a P wave. In
addition, the minimum value of the S wave for the median composite signal was slightly
lower than what would be expected, and the peak of the R wave was slightly higher than
what would be expected.
- Interestingly, the median composite signal performed better for the glitch scenario. This
is because the glitch caused the mean composite signal to incorrectly elevate the T-P
segment of the ECG wave. On the other hand, the median composite signal was able to
maintain fairly accurately the overall shape of the waveform and did not contain any
abnormal components.