Projects for Tik 61.183 --- Spring 2002
2 to pass
3 to excellent grade
The projects should be returned with reports for each project
separately, using an article format.
- Files pec1.dat, pec33.dat and pec52.dat have three-channel recordings of the PCG, ECG, and carotid pulse signals (sampled at 1 KHz). pec1 and pec52 are from adult male subjects and pec33 is from a female baby with systolic murmur (for details see, e.g., pages 234 and 312, 4th projects, of Rangayyan's book).
You may find the script plotpec.m of some help.
- Apply the Pan-Tompkins method for QRS detection to the ECG channel and the Lehner and Rangayyan method to detect the dicrotic notch in the carotid pulse channel;
- Extrapolate the timing information from the ECG and carotid pulse channels to segment the PCG signal into two parts: the systolic part, from the onset of an S1 to the onset of the following S2; and the diastolic part, from the onset of an S2 to the onset of the following S1;
- Compute the PSD of each segment. Extend the procedure to average the systolic and diastolic PSDs over several cardiac cycles. Compare the PSDs obtained for the three cases;
- Again for the systolic and the diastolic segments, apply the AR modeling procedure and derive a model PSD. Compare these PSDs with the previous ones?
- Calculate the average envelogram over several cardiac cycles using the ECG as the trigger. (specify, in particular, how you handled the variations in the duration of the signals from one beat to another)
- The file safety.dat contains the speech signal for the word "safety" uttered by a male speaker, sampled at 8 KHz. The signal has a significant amount of background noise.
You may find the script safety.m of some help.
- Develop procedures to derive short-time RMS, turns count, and ZCR in moving windows of durations in the range of 10 - 100 ms;
- Study the variations in the parameters in relation to the voiced, unvoiced, and silence (background noise) portions of the signal;
- Compute the PSD for each segment that you obtain and study its characteristics.
- How does the durations of the windows affect the trends in the parameters computed and on the segmentations?
- The MEG data in MEG.dat has a number of artifacts in it. These include ocular (both blinking and horizontal saccades), cardiac and myographic contaminations. You still have a weak (but real) digital watch. Underneath it all, there is some alpha activity (artificially added :-) ). Your task is to:
- Identify all the artifacts and alpha activity named above;
- Make use of any method studied in the book (in case of despair you can get one or two artifacts from your preferred bag of tricks);
- Bonus question: there may be some additional artifact that hasn't been mentioned... finding any will be acknowledged.
Monday, 18-Mar-2002 11:13:48 EET