• General Information
• Syllabus
• Laboratory Projects
• Lectures
• Bibliography

SYLLABUS

Laboratory Projects

Introduction to signal  processing  and  visualization  using Matlab
(1 week - Obeid)
 
1. The electrocardiogram (ECG): Design a ventricular arrhythmia detector

The lab includes recording ECG from a student in class, designing and testing filters for noise reduction, and finally using those filters as one component of a system to detect irregular heart rhythms. The detector will be tested on normal and abnormal ECG signals.
(2 weeks - Greenberg)
 

2. Array processing: Multimicrophone arrays and binaural hearing

Simple array configurations will be designed to create directionally-sensitive systems. Filter approximations to these designs will be generated, and the resulting systems will be evaluated and simulated. A simple scheme to create an array system with binaural output will be investigated.
(2 weeks - Greenberg)
 

3. Analysis of single-unit data: Identify a model of auditory-nerve fibers

System identification for a functional model of an auditory-nerve fiber. Identification is based on single-unit analysis techniques such as post-stimulus-time histograms and crosscorrelation. Students will compare prediction of identified model with measured auditory-nerve fiber responses to speech utterances.
(2 weeks - Delgutte)
 

4. Linear prediction of speech: Design an LPC vocoder

A complete linear-production  anaylsis-synthesis  system (vocoder) will be developed and evaluated using speech utterances.   The  vocoder comprises  a linear-prediction analyzer, a pitch detector based
  on the prediction error signal, and a synthesizer.
(2 weeks  - Delgutte)
 

5. Post-processing brain MRI:  Isolate and visualize anatomical structure

Clinical MRI scans will be processed to reduce noise, label tissues and extract brain contours.  Structures extracted from the  imagery will be visualized in 3D.

Lectures

A. Fundamentals of Digital Signal Processing (J. Greenberg)
1. Biological signals and noise. Data acquisition (sampling, interpolation. quantization). Oversampling and sigma-delta modulation. (B. Delgutte)
2. Digital filters. Difference equations. FIR and IIR filters. Basic properties of discrete-time systems: linearity, shift-invariance, stability, causality. The convolution sum.
3. Fourier representation of discrete-time signals. The discrete-time Fourier transform and its properties.  FIR filter design using windows.
4. The electrocardiogram. Relation of ECG components to cardiac events. Neural control of heart rate. Clinical applications.
5. Continuous-time Fourier transforms. Sampling theorems. Aliasing. Fourier series. Spectral analysis. Uncertainty principle.
6. The discrete Fourier transform. FIR filter design by frequency sampling. Cyclic convolution. Overlap-save method of convolution.
7. The Z transform. Regions of convergence. Stability of digital filters. Array signal processing. Multi-microphone arrays and their application to hearing aids.

B. Random signals and speech processing (B. Delgutte)
1. Random signals. Time and ensemble averages. Autocorrelation and crosscorrelation functions.
2. Effects of digital filters on random signals. Matched filters. Estimation of parameters of random signals.
3. Fourier representation of random signals. Power spectra and cross spectra. System identification and Wiener filters.
4. Signal processing by the mammalian ear. Models of auditory nerve activity.
5. Speech. Acoustic theory of speech production. The source filter model. Spectral and temporal characteristics of speech sounds.
6. Spectrographic analysis of speech. The short-time Fourier transform. Time and frequency resolution.
7. Linear prediction of speech and its relation to the source filter model.

C. Image processing, visualization and medical imaging (W. Wells)
1. Medical image modalities: ultrasound, X-ray, CT, MRI, PET, SPECT. Image perception by the human vusual system.
2. 2D and 3D image processing techniques I: Image representation, Fourier methods, 2D visualization, filtering.
3. Physics and signal processing for MRI.
4. 2D and 3D image processing techniques II: interpolation, noise reduction methods, homomorphic filtering.
5. Image segmentation I: thresholding, statistical classifiers.
6. Image segmentation II: edge detection, morphological operators.
7. 3D visualization: surface rendering, volume rendering.
8. Image registration: correlation and other intensity-based methods, surface-based methods, rigid and non-rigid methods.
9. Selected medical applications: surgical planning and guidance, disease quantification.

Quizzes

Quiz 1 will cover the material in Part A of the lectures and Part B, Lectures 1-3.
Quiz 2 will cover the material in Part B, Lectures 4-7and Part C.
 

---

BIBLIOGRAPHY

• Oppenheim and Schafer (1989). Discrete-time Signal Processing, Prentice-Hall.
• Schwartz and Shaw (1975). Signal Processing, McGraw Hill.
• Papoulis (1984). Probability, Random Variables, and Stochastic Processes, McGraw Hill.

• Siebert (1986). Circuits, Signals, and Systems, McGraw Hill.
• Oppenheim, Wilsky and Nawab (1997). Signals and Systems. Prentice Hall.
• Karu (1994). Signals and Systems Made Ridiculously Simple.

• Rabiner and Schafer (1978). Digital Processing of Speech Signals, Prentice-Hall.

• Lim (1990) Two-Dimensional Signal and Image Processing, Prentice Hall.
• Gonzalez and Woods (1993) Digital Image Processing, Addison-Wesley.
• Webb (1988). The Physics of Medical Imaging, Adam Hilger.