3 ECTS credits
90 u studietijd

Aanbieding 1 met studiegidsnummer 4021676ENR voor alle studenten in het 2e semester met een verdiepend master niveau.

Semester
2e semester
Inschrijving onder examencontract
Niet mogelijk
Beoordelingsvoet
Beoordeling (0 tot 20)
2e zittijd mogelijk
Ja
Onderwijstaal
Engels
Faculteit
Faculteit Ingenieurswetenschappen
Verantwoordelijke vakgroep
Elektronica en Informatica
Onderwijsteam
Jef Vandemeulebroucke (titularis)
Onderdelen en contacturen
18 contacturen Hoorcollege
18 contacturen Werkvormen en Praktische Oef.
Inhoud

Goal of the course

The goal of this course is to provide an overview of the field of clinical decision support. The role of decision support in clinical care is presented. Different types of approaches are explained, based on the type of data on which they operate, and the types of techniques which they employ.

A brief introduction to health information systems is given. Knowledge-based systems for decision support are introduced, including rule-based systems, fuzzy logic systems and Bayesian belief networks. Next, data-driven approaches are discussed, and several popular methods from machine learning are introduced. Specific attention is given to computer-aided diagnosis, for providing decision support for unstructured data such as biomedical signals and images. Methods for improving learning from data are desribed. Finally, considerations for system design, validation, certification and ethics are discussed.

Contents

  • Introduction to health information systems
  • Knowledge-based decision support systems: rule-based systems, fuzzy logic systems and Bayesian belief networks
  • Data-driven decision support systems: perceptron, support vector machine, decision trees, nearest neighbor, neural networks
  • Learning aides: data balancing, normalization, feature selection, ensemble methods, bagging and boosting
  • Computer-aided diagnosis: feature extraction and classification, convolutional neural networks
  • System design, validation, certification and ethical considerations

Practical sessions and exercises:

The lectures are supported by 4 practical sessions, covering selected topics from the lectures:

  • Introduction to data loading, processing and visualization
  • Knowledge-based decision support systems
  • Data-driven decision support systems
  • Computer-aided diagnosis

 

Studiemateriaal
Digitaal cursusmateriaal (Vereist) : Slides presented during lectures, Jef Vandemeulebroucke
Digitaal cursusmateriaal (Vereist) : Exercises from practical sessions, Jef Vandemeulebroucke
Digitaal cursusmateriaal (Vereist) : Scientific articles provided
Bijkomende info

Lectures will cover the theoretical part of the course. Practical sessions will consist of exercises in which the concepts seen during the lectures are applied. Practical sessions will be guided by assistants. Reports on the practical sessions can be finalized afterwards.

Leerresultaten

Learning outcomes

After completing this course, the student will be able to:

  • List the current information systems used in health care, differentiate the type of information that is stored, and identify the current limitations
  • Illustrate the principle of knowledge-based decision support systems, explain the techniques treated during the lectures, and apply such decision support given the description of real medical problems
  • Illustrate the principle of data-driven decision support systems, explain the techniques treated during the lectures, and apply such decision support given real data sets
  • Summarize the main problems which can occur when learning from data and carry out common operations that can aid the learning process.
  • Illustrate the principle of computer-aided diagnosis, explain the techniques treated during the lectures, and apply such algorithms on real signals and images
  • Compare the different types of systems and identify the advantages and disadvantages
  • Outline the current possibilities and opportunities for clinical decision support systems, and the technological and legal challenges.

 

Beoordelingsinformatie

De beoordeling bestaat uit volgende opdrachtcategorieën:
Examen Mondeling bepaalt 65% van het eindcijfer

Examen Praktijk bepaalt 35% van het eindcijfer

Binnen de categorie Examen Mondeling dient men volgende opdrachten af te werken:

  • Oral exam on theory met een wegingsfactor 1 en aldus 65% van het totale eindcijfer.

    Toelichting: Oral exam with written preparation (closed book). After the questions are given, students will be given time to prepare the answers on paper, before explaining these during oral examination.

Binnen de categorie Examen Praktijk dient men volgende opdrachten af te werken:

  • Reports practical sessions met een wegingsfactor 1 en aldus 35% van het totale eindcijfer.

    Toelichting: Reports describing the results on the assignments given during the practical sessions will be evaluated. Evaluation will be based on the correctness of the performed assignments, the answers to the questions and the completeness of the reports.

Aanvullende info mbt evaluatie

Students must participate to the oral exam and complete the reports on the practical sessions. Students must pass both parts (oral exam and practical sessions) in order to pass the course. An exemption for either part can be obtained for the second session, if a passing grade was obtained for that part in the first session.

 

Toegestane onvoldoende
Kijk in het aanvullend OER van je faculteit na of een toegestane onvoldoende mogelijk is voor dit opleidingsonderdeel.

Academische context

Deze aanbieding maakt deel uit van de volgende studieplannen:
Master in de ingenieurswetenschappen: toegepaste computerwetenschappen: Standaard traject
Master of Applied Sciences and Engineering: Applied Computer Science: Standaard traject (enkel aangeboden in het Engels)