BIOEN 316 Biomedical Signals and Sensors Spring 2020

Homework Assignment 1

Homework 1 consists of two parts. Part A has only one problem and is due next Monday.
Part B has two problems and is due the following Friday.

Homework 1, Part A
This is a handwritten assignment that should be scanned and uploaded to Canvas by
Monday, April 6. Exact due time to be determined.

The Problem is to derive the formulas necessary to plot the heart’s electrical axis as a
function of time, so you can implement these formulas in MATLAB for Part B of the
homework. You do not need to get the formulas exactly right to get full credit for this
problem, but you do need to make a good effort and at least get similar equations.
You should end up with a pair of equations that allow you to calculate the heart
electrical vector VH, given lead I and lead II voltages. You can derive the vector in Cartesian
form (x and y components), or in polar form (as an angle and a magnitude or length).
Either way will require some similar triangles and trigonometric functions.
You will earn points for having (1) a figure, based on Einthoven’s triangle, that shows
the graphical relationship between the components of the heart axis, (VHX, VHY) or (|VH|, θ),
and the lead I and lead II voltages, and (2) a derivation of the two components of VH. You do
not have to arrive at exactly the correct figure or formula yet; you will have time to fix
mistakes when you do the MATLAB part in problem 1B. Try not to spend more than an
hour on this derivation, and always get help if you’re really stuck for more than 20 minutes.

1
BIOEN 316 Biomedical Signals and Sensors Spring 2020

Homework 1, Part B
Due Friday, April 10, as a pdf or Word document uploaded to Canvas

Problem B1 (long)

This problem uses the PTB Diagnostic ECG Database, which is available from the
PhysioNet signal repository. A description of the PTB database can be found here:
http://physionet.org/physiobank/database/ptbdb/,
and a list of the healthy patients may be found here:
http://physionet.org/physiobank/database/ptbdb/CONTROLS.
The other patients all have some cardiac pathology (abnormal heart condition).

a. Go to the Physiobank ATM here: https://archive.physionet.org/cgi-bin/atm/ATM.

Select the PTB Diagnostic ECG Database, and one of the healthy patient records.
Download 10 seconds of the signal, starting from anywhere within the signal. Note:
in the Toolbox, select “Export signals as CSV” and then under the blue bar, click the
tiny samples.csv. This should give you the 12 standard leads in a spreadsheet.

b. Process the lead I, II, and III columns to have a mean near zero by subtracting the
mean over every half second using a MATLAB function provided for you. You will
learn to write a similar function in BIOEN 317.

c. Use the lead I and lead II signals, your derivation from HW 1A, and MATLAB to make
a polar plot of the heart axis as it changes over time. If you like, you may choose a
signal segment with fewer heartbeats (less time) to make your plot look cleaner.

d. Use leads I & III or II & III to create a new polar plot. Compare the two polar plots
you have created. A large difference is probably a math error, but they might match
well or be a little off. Suggest two reasons why the plots might be slightly different.

Notes
• There is a polar command in MATLAB that can make the appropriate plot.
• You are welcome to compare the plot from a healthy heart to one from an unhealthy
heart. To do so, go to http://physionet.org/physiobank/database/ptbdb/ and select
a patient whose number is not on the controls list. Each folder contains a .hea file
that describes the patient’s conditions.

Problem B2 (short)

In the PTB database home page (http://physionet.org/physiobank/database/ptbdb/),

under Data Description, find the values for Input voltage and Resolution. Show that the
resolution information can be used to calculate the given input voltage range. You may
ignore the “compensated offset voltage.”