Dades generals

Curs acadèmic:
2017
Descripció:
The aim of this course is to introduce all the steps needed to develop a CADe medical system. The main focus will be on the analysis of different techniques to perform medical image segmentation tasks.
Crèdits:
6

Grups

Grup A

Durada:
Semestral, 1r semestre
Professorat:
JOSE ALFONSO BERNAL MOYANO  / XAVIER LLADO BARDERA  / ROBERT MARTI MARLY
Idioma de les classes:
Anglès (100%)

Altres Competències

  • To have a good knowledge of the field of Computer Aided Detection (CADe).
  • To analyse the state of the art segmentation algorithms used in medical image analysis, from the perspective of the computer vision engineer.
  • To be able to evaluate a segmentation algorithm and asses is usability for 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 Medical Image Segmentation and Applications

2. Image preprocessing

3. Clustering segmentation techniques

4. Region-based segmentation techniques

5. Atlas based segmentation EM/Bayesian +atlas / Markov Random Fields

6. Segmentation via detection + Patches + classification

7. Deep learning for image segmentation (CNNs)

8. Evaluation of segmentation algorithms for medical applications

Activitats

Tipus d’activitat Hores amb professor Hores sense professor Total
Anàlisi / estudi de casos 8 25 33
Aprenentatge basat en problemes (PBL) 10 50 60
Classes expositives 22 25 47
Seminaris 4 6 10
Total 44 106 150

Bibliografia

  • González, Rafael C. (2004). Digital image processing using Matlab. Upper Saddle River : Prentice Hall, cop. 2004.
  • Forsyth, David A., Ponce, Jean (2003). Computer vision : a modern approach. Upper Saddle River: Prentice Hall.
  • Robert B. Fisher (2007). CVonline: The Evolving, Distributed, Non-Proprietary. On-Line Compendium of Comp.

Avaluació i qualificació

Activitats d'avaluació:

Descripció de l'activitat Avaluació de l'activitat %
Theoretical lectures 20
Lab sessions 40
Segmentation project 40

Qualificació

The evaluation is based on three different activities: 40% from small lab sessios P1&P2 + 40% from the final project + 20% from the exam.

Evaluation Criteria:
From Labs: 70% strategy and results + 30% document

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 exam)

Observacions

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