Physiological Signal Processing

Code:

EBE0052

Acronym:

PSFI

Keywords
Classification	Keyword
OFICIAL	Biomedical Engineering

Instance: 2010/2011 - 1S

Active?	Yes
Web Page:	http://moodle.fe.up.pt/1011/course/view.php?id=82
Responsible unit:	Telecommunications
Course/CS Responsible:	Master in Bioengineering

Cycles of Study/Courses

Acronym	No. of Students	Study Plan	Curricular Years	Credits UCN	Credits ECTS	Contact hours	Total Time
MEB	1	Syllabus	1	-	6	56	162
MIB	41	Syllabus	3	-	6	56	162
			4
			5
MIEEC	4	Syllabus (Transition) since 2010/2011	4	-	6	56	162
			5
		Syllabus	4	-	6	56	162
			5
PRODEB	1	Syllabus	1	-	6	56	162

Teaching language

Suitable for English-speaking students

Objectives

The objective of this course unit is to motivate students to the nature and diversity of physiological signals (for example: EMG, EEG, ECG), and to familiarize students with the theory foundations in the area of digital signal processing as well as their valorization as practical skills allowing students to understand and design important processes in physiological signal processing including acquisition, conditioning, filtering, analysis and representation of relevant information.
After successful completation this course unit, students will be able to use techniques and technologies of physiological signal processing, by strengthening not only their application to diagnosis objectives, therapy and rehabilitation, but also to foster specialization and innovation in these areas.

Program

1. Introduction to electrophysiology
Concept of electrophysiology
Membranes, bioelectrical currents, membrane polarization
Action potentials, propagation, extracellular signals
Physiological systems
Principles of electrocardiography (ECG), electromyography (EMG) and electroencephalography (EEG)

2. Signals and systems
Characterization and representation of discrete signals
Characterization and representation of discrete systems

3. Fourier analysis
The four faces of Fourier analysis
Review of the discrete-time Fourier transform
Discrete Fourier Transform (DFT) as a sampling of the discrete-time Fourier transform
Properties of the DFT
Fast calculation of the DFT using the Fast Fourier Transform (FFT)

4. Sampling and reconstruction of signals
Periodic sampling of continuous signals
Frequency analysis of periodic sampling
Reconstruction of continuous signals from samples
Discrete processing of continuous signals

5. Z transform
Definition, convergence region, properties, inverse Z transform

6. Discrete filters
Types of filters
Transfer function and frequency response
Group delay and inverse system
Minimum-phase and maximum-phase filters
Linear-phase filters: types I, II, III, and IV
The zero-pole representation of linear-phase filters
Design of IIR filters (impulse invariance, bilinear transformation)
Design of FIR filters (window method, optimum design using Parks-McClellan)
Structures for the realization of IIR and FIR filters

7. Fast FIR filtering using the FFT
Overlap-add method
Overlap-save method

8. The auto-correlation and cross-correlation
As special cases of the discrete convolution
The Wiener-Khintchine theorem: fast correlation using the FFT

9. Principles of spectral estimation
The DFT as an uniform bank of filters
Spectral estimation using DFT (periodogram, spectrogram)

10. Physiological signal processing
Examples of application

Mandatory literature

Oppenheim, Alan V.; Discrete-Time Signal Processing. ISBN: 0-13-216771-9

Complementary Bibliography

Bronzino, Joseph Daniel, 1937- 340; The biomedical engineering handbook
Enderle, Joseph Bronzino John; Introduction to Biomedical Engineering. ISBN: 0-12-238662-0
Bruce, Eugene N.; Biomedical signal processing and signal modeling. ISBN: 0-471-34540-7

Teaching methods and learning activities

This course unit will be based on theoretical-practical classes and practical classes. The former include: a) theory presentation and illustration of the topics of the course using multimedia support, b) conventional (paper-and-pencil) and Matlab based exercises allowing students to master the applied perspective of main topics of the course, and c) individual mini-tests at the end of some classes. The latter (practical classes) include the laboratory realization of experimental work using Matlab, the Biopac platform for the acquisition and analysis of physiological signals, and a DSP-based platform for the realization of signal processing algorithms in real-time.

Software

Matlab 6 R12.1

Evaluation Type

Distributed evaluation with final exam

Assessment Components

Description	Type	Time (hours)	Weight (%)	End date
Attendance (estimated)	Participação presencial	52,00
Final exam	Exame	3,00
Preparation for mini-test / exam	Exame	42,00
	Total:	-	0,00

Amount of time allocated to each course unit

Description	Type	Time (hours)	End date
Regular Study	Estudo autónomo	65
	Total:	65,00

Eligibility for exams

In order to be admitted to the final exams, students should comply with the General Evaluation Rules of FEUP concerning the maximum number of missed classes, should perform at least two mini-tests and at least three graded lab projects during the semester. The continuous assessment combines the grades of mini-tests and graded lab projects. The grades of the (individual) mini-tests will be weighted at 40% and those of the graded lab projects (groups of two students) will be weighted at 60%.

Calculation formula of final grade

Final Exam consists of a closed-book written examination whose duration is 2 hours. Students will be provided with a formula sheet.
The final grade (FG) will be computed using the following formula which combines the grades of two components: continuous assessment (CA) and final exam (FE):

FG= 0.3 CA + 0.7 FE

The scale for both components is [0, 20].
The attendence and participation in class may influence a residual ajustment in the final grade.

Special assessment (TE, DA, ...)

According to Number 6 of Article 5 of the General Evaluation Rules of FEUP, students who did not obtain Continuous Assessment during the semester, will have to perform an extra practical exam including the utilization of Matlab.