Analysis of biological data

Teacher(s)

Jodogne Sébastien;

Language

French

Prerequisites

This teaching unit assumes that the student has skills about the Java programming language (as for instance targeted in courses LSINC1402 and LEPL1402), about linear algebra (as for instance targeted in courses LSINC1112 and LINFO1112), as well as about Web technologies (as for instance targeted in courses LSINC1002 et LINFO1002).

Main themes

This teaching unit proposes an introduction to the spatial and temporal analysis of neurophysiological signals, particularly electroencephalograms (EEG), as well as to the analysis of medical images. It is focused on the development of algorithms that are applicable to such data, as well as on the deployment of these algorithms as Web applications.

Learning outcomes

At the end of this learning unit, the student is able to :

In consideration of the reference table AA of the program "Bachelor degree in Computer Science", this course contributes to the development, to the acquisition and to the evaluation of the following experiences of learning: AA1.I3, AA1.I6, AA1.G2, AA1.G3 AA 2.4 AA 4.4, AA4.6 AA5.3 S1.I6, S1.G3,  S2.4 S4.4 S5.5  More specifically, at the end of the course, the student will be able to: understand the fundamental methods for the preprocessing and filtering of signals and images; apply techniques for the extraction of information from time series of electroencephalograms, as well as from medical images; implement algorithms for the processing of 1D and 2D signals in a compiled language (Java); create Web applications that rely on scientific computations executed on a remote server.

Content

• Biological data:
  • Time series for neurophysiological data, notably electroencephalograms (EEG).
  • Introduction to the acquisition of medical images (radiographs and CT-scans).
• Introduction to the analysis of 1D and 2D signals:
  • Time-domain and frequency-domain analysis, and feature extraction.
  • Fast Fourier Transform (FFT).
  • Independent component analysis.
  • Principal component analysis.
  • Image processing (gray-level mappings, convolution, non-linear filters and morphology).
  • Image segmentation.
• Development of scientific applications in client/server mode:
  • Interoperability standards for EEG and medical imaging (European Data Format, DICOM...).
  • Data rendering using the HTML5 canvas.
  • Design of REST APIs using the Java programming language.

Teaching methods

• Lectures in auditorium.
• Practical exercises to be completed individually on the INGInious platform (in the Java programming language).
• A teaching assistant will be available to answer questions about the exercises during the practical sessions scheduled in the timetable.

Evaluation methods

In the first examination session, the exam will be held in person, without access to course materials, and will consist of open-ended questions. The assessment covers all the material presented during the lectures and practical sessions. The final grade is calculated as a weighted average of the exam (90%) and the continuous assessment from the practical sessions (10%). In the event of re-enrollment, the practical exercises must be completed again in their entirety.
In the second examination session, the exam will be oral. Continuous assessment no longer counts in the second session: the exam accounts for 100% of the grade.
Continuous assessment consists of the practical exercises, which are weighted equally. They result in a single overall grade, communicated after the final session. Failure to comply with the methodological guidelines communicated by the instructor, particularly regarding the use of online resources, plagiarism, or collaboration between students on an exercise, will result in an overall mark of 0 for the continuous assessment.
In particular, the use of generative AI tools and any form of collaboration is strictly prohibited for the exercises. The distribution or exchange of (fragments of) code between students is not permitted by any means whatsoever (GitHub, Facebook, Discord, etc.), even after the deadline for submitting the exercises.

Online resources

Moodle UCLouvain -> https://moodle.uclouvain.be/course/view.php?id=5834

Teaching materials

• Les transparents présentés lors des exposés théoriques, de même que les notes relatives aux séances de cours et quelques références bibliographiques, sont disponibles sur Moodle.
• The slides presented during the theoretical lectures, as well as the course notes and some bibliographical references, are available on Moodle.

Faculty or entity

> SINC