Dades generals

Curs acadèmic:
2017
Descripció:
The aim of this course is to introduce all the steps needed to develop a CADx medical system, i.e. a system that help physicians to deliver a diagnosis. The topics cover both the traditional scheme including independent image segmentation, characterisation, and classification approaches as well as the recent groundbreaking deep learning technology.
Crèdits:
5

Grups

Grup A

Durada:
Semestral, 1r semestre
Professorat:
KAISAR KUSHIBAR  / XAVIER LLADO BARDERA  / ARNAU OLIVER I MALAGELADA
Idioma de les classes:
Anglès (100%)

Altres Competències

  • To have a good knowledge of the field of Computer Aided Diagnosis (CADx).
  • To have an overview of general image characterization.
  • Applying pattern recognition techniques to the field of medical imaging.
  • To learn what characteristics and what classifiers are more useful to the different medical images.
  • To be able to evaluate a previously developed algorithm and asses is usability for medical images and daily clinical usage. Estimate the crucial factors for it to be successful.
  • To learn what algorithm(s) could fit better for a particular application.

Continguts

1. Introduction to diagnosis and CADx

2. Image characterization: morphological, texture, and shape descriptors

3. Deformable template matching and active shape models

4. Free-form segmentation and active contours

5. Interest point detectors and descriptors

6. Object and image characterization

7. CADx evaluation and applications

Activitats

Tipus d’activitat Hores amb professor Hores sense professor Total
Aprenentatge basat en problemes (PBL) 16 56 72
Classes expositives 14 14 28
Lectura / comentari de textos 6 22 28
Total 36 92 128

Bibliografia

    Avaluació i qualificació

    Activitats d'avaluació:

    Descripció de l'activitat Avaluació de l'activitat %
    Lecture Activity 50% document + 50% presentation and interaction 30
    Lab session 1: Active Shape Models 70% strategy and results + 30% document 20
    Lab session 2: Image characterisation and diagnosis 70% strategy and results + 30% document 20
    Lab Project: CADx development on Mammography 70% strategy and results + 30% document 40

    Qualificació

    The evaluation is based on three different activities: 40% from Labs P1&P2 + 40 % from final Lab Project + 30% by evaluating the lectures given by the students.

    Criteris específics de la nota «No Presentat»:
    NP will be considered when the students do not submit any of the evaluation activities (P1, P2, Final project, or Lecture activity)

    Observacions

    Mentoring will be held in the office 015 of building P-IV.