Applied biomedical signal processing

EE-512

This file is part of the content downloaded from Applied biomedical signal processing.
Course summary

General info

Course: Applied Biomedical Signal Processing

Teachers: Dr. Mathieu Lemay (responsible), Dr. João Jorge (main support), Dr. Philippe Renevey, Dr. Martin Proença, Dr. Adrian Luca, Dr. Guillaume Bonnier, Dr. Karen Adam, Dr. Ramin Soltani, Dr. Clémentine Aguet

Course description
The goal of this course is two-fold:

  1. to introduce physiological bases, signal acquisition modalities and state-of-the-art signal processing techniques, and,
  2. to propose concrete examples of applications for vital sign monitoring and diagnostic purposes.


The main signal processing topics presented will be:

  • Basics of continuous and discrete time Fourier transform
  • Linear filter design
  • Stochastic signals and filtering
  • Power spectral density
  • Basics of time-frequency analysis
  • Time-frequency representations
  • Auto-regressive, Moving average and ARMA signal modeling
  • Instantaneous frequency
  • Adaptive filter frequency tracking
  • Singular value decomposition
  • Principal component analysis
  • Linear/non-linear regression
  • Classification and feature selection
  • Perceptron, MLP & activation functions
  • Gradient descent and back-propagation
  • Convolutional and recurrent neural networks

As the course scheduling may evolve over time, the detailed agenda and materials are provided on a short-term basis.

This course includes weekly theoretical and practical sessions. All sessions are currently scheduled to take place in room CM 1 4 (new room).

The practical sessions require a laptop. During these sessions, experimental biomedical signals will be investigated to address a series of questions / exercises, in groups of 4 to 5 students. The students are encouraged to ask questions and discuss, together and with the teacher. The answers are to be compiled in group reports, to be handed-in within a week.

The date, time and location of the final exam need to be confirmed. All printed/written documents will be allowed (open book); devices such as laptops, tablets, and phones are prohibited.

Thursday (15h15-19h00)

15h15-19h00: Lecture on Module 01 - Introduction. ROOM MEB331
Presentation of the course, general context and module structure. Introduction to biomedical signal processing with examples and short practical exercises in Python.

Please note:
  • No report for this practical session needs to be submitted.
  • The course will not be recorded.
  • Laptop necessary for practical exercises.

Thursday (15h15-19h00)

15h15-19h00: Lecture on Module 02 - Basics I. ROOM MEB331

Please note:
  • The course will not be recorded.
  • Laptop necessary for practical exercises.

Thursday (15h15-19h00)

15h15-19h00: Lecture on Module 03 - Basics II. ROOM MEB331

Please note:
  • The course will not be recorded.
  • Laptop necessary for practical exercises.

Thursday (15h15-19h00)

15h15-19h00: Lecture on Module 04 - Time-Frequency Analysis. ROOM MEB331

Please note:
  • The course will not be recorded.
  • Laptop necessary for practical exercises.

Thursday (15h15-19h00)

15h15-19h00: Lecture on Module 05 - Linear Models I. NEW ROOM CM 1 4

Please note:
  • The course will not be recorded.
  • Laptop necessary for practical exercises.

Thursday (15h15-19h00)

15h15-19h00: Lecture on Module 06 - Linear Models II. CM 1 4

Please note:
  • The course will not be recorded.
  • Laptop necessary for practical exercises.

Holidays

Thursday (15h15-19h00)

15h15-19h00: Lecture on Module 07 - Instantaneous frequency estimation. CM 1 4

Please note:
  • The course will not be recorded.
  • Laptop necessary for practical exercises.

Thursday (15h15-19h00)

15h15-19h00: Module 08 - Midterm (mock) exam. CM 1 4

Please note:
  • The purpose of this test is to provide an idea of what can be expected for the final exam.
  • Unlike the final exam, the midterm only covers the modules already taught until now.
  • The test is optional and will not be evaluated.

Thursday (15h15-19h00)

15h15-19h00: Lecture on Module 09 - Singular Value Decomposition. CM 1 4

Please note:
  • The course will not be recorded.
  • Laptop necessary for practical exercises.

Thursday (15h15-19h00)

15h15-19h00: Lecture on Module 10 - Principal Component Analysis. CM 1 4

Please note:
  • The course will not be recorded.
  • Laptop necessary for practical exercises.

Thursday (15h15-19h00)

15h15-19h00: Lecture on Module 11 - Regression and Classification. CM 1 4

Please note:
  • The course will not be recorded.
  • Laptop necessary for practical exercises.

Thursday (15h15-19h00)

15h15-19h00: Lecture on Module 12 - Introduction to Neural Networks - CM 1 4


Please note:
  • The course will not be recorded.
  • Laptop necessary for practical exercises.

15h15-19h00: Lecture on Module 13 - Neural network regularization and applications - CM 1 4


Please note:
  • The course will not be recorded.
  • Laptop necessary for practical exercises.

15h15-19h00: Module 14 - Open Q&A session - CM 1 4


Please note:
  • This session is reserved for discussing questions regarding the course topics and the exam.
  • As announced in the forum, we ask the students to please submit their questions in advance (by e-mail).
  • No new concepts will be introduced in this module.