TeachingWs15MLMI

Chair for Computer Aided Medical Procedures & Augmented Reality
Lehrstuhl für Informatikanwendungen in der Medizin & Augmented Reality

THIS WEBPAGE IS DEPRECATED - please visit our new website

Practical Course:
Machine Learning in Medical Imaging

Nassir Navab, Diana Mateus, Loic Peter, Fausto Milletarì, Sailesh Conjeti, Shadi Albarqouni, Pascal Fallavollita

Type: Master Practical Course Module IN2106
SWS: 6
ECTS: 10 Credits
Location: 03.13.008, Praktikumsraum
Time: 10.00 - 13.00
Course Language: English

Announcements

  • To register for the course, please use the TUM-Matching-System
  • The preliminary meeting will take place on the 17th of July at 16:00 in MI 03.13.010 and will have an approximative duration of 1 hour
  • During the preliminary meeting we will try to define the schedule of the course in order to fit the needs of as many students as possible: try to have a rough idea about your schedule for next semester before the preliminary meeting
  • You are required to form groups using the doodle available at the address link before Monday 30th November 2015 at 7.00 CET..

About the Course

The aim of the course is to provide the students with notions about various machine learning techniques. The course is subdivided in three lecture blocks. We will talk about data processing and analysis, unsupervised methods and supervised methods. In the introductory meeting we will provide you with a broader overview of what the topics of the course will be.

The students will be required to submit a final project and a 8 page long report (llncs2 format LaTeX) at the end of the semester. The students should also give a final presentation about their work.

This year we propose a new format for the MLMI course. We will have shorter lectures oriented towards implementation more than towards theory. Exercises will be assigned on a weekly basis and they are designed to be completed within the time-block allocated for the course. Optionally the students can complete their work, individually, at home. The exercises must be submitted to our systems using GIT. We recommend the students to commit very often during development of their solutions such that their work can be properly evaluated.

Evaluation

  • The exercises will be automatically evaluated using our platform. The impact of the exercises on the final grade is 50%.
  • The final projects will be evaluated by the tutors using their own criterion. The impact of the final project on the final grade is 50%.

Registration

  • Interested students should attend the introductory meeting to enlist in the course.
  • Students in the waiting list of summer term would have priority for registration.
  • Students can only register through TUMOnline themselves if the maximum number of participants hasn't been reached by end of August.
  • Maximum number of participants: 15.

Requirements

  • Strong knowledge of MATLAB
  • Basic knowledge of C++ is advantageous (in case you decide to use c++ for your final project): knowledge of main concepts of object oriented programming, basic software engineering, image processing libraries and linear algebra libraries.
  • Basic knowledge of Python is advantageous (in case you decide to use python for your final project): numpy, scipy, scikit-learn.
  • Strong background in image processing using MATLAB
  • Basic knowledge of naive machine learning methods and terminology: knowledge of the meaning of the words classification, regression, data-set, cross-validation ...

Project Assignment

Group #1 Subash Thapa
Stefan Marinov
Project #1
Group #2 Manuel Bergler
Avinash Kumar
Project #6
Group #3 Richard Schweiger Project #3
Group #4 Sergey Zakharov
Mahdi Saleh
Project #4
Group #5 Eva Breznik
Muriel Keribin
Project #5
Group #6 Tobias Wolff
Jan Wymenga
Project #2

Schedule

Date Topic By Slides Homework
17.07.2015 Introductory meeting Fausto Milletari None None
14.10.2015 Guided exercise example Fausto Milletari None GIT
21.10.2015 Image Representation and Feature Extraction Loïc Peter Slides GIT
28.10.2015 Principal Component Analysis (PCA) and Linear Discriminant Analysis (LDA) Fausto Milletari Slides GIT
Data
04.11.2015 Clustering Shadi Albarqouni Slides GIT
11.11.2015 Introduction to Sparse Methods Shadi Albarqouni Slides  
18.11.2015 Linear Classifiers and Support Vector Machines (SVMs) Diana Mateus Slides Exercise sheet
Data
link to libsvm
ExampleSolutionRegression
Example Solution Linear SVM
Example Solution rbf SVM
25.11.2015 Boosting Loic Peter Slides
Projects
Doodle to define group participants
Exercise sheet
Code
02.12.2015 Random Forests Sailesh Conjeti Slides Exercise sheet
Code
16.12.2015 Deep Learning Wadim Kehl Slides  
13.01.2015 Attestation
10:00-11:20 (Students from Groups 1, 2 and 4)
11:30-12:30 (Students from Groups 3, 5 and 6)
Students    
24.02.2015 Final Presentations Students 10 minutes each plus 5 minutes for questions.  

Reading Material

As a complement for the projects you might want to check the material on Evaluation Measures Evaluation Measures




TeachingForm
Title: Machine Learning In Medical Imaging
Professor: Nassir Navab
Tutors: Loic Peter, Diana Mateus, Fausto Milletari, Sailesh Conjeti, Shadi Albarqouni; Pascal Fallavollita
Type: Praktikum
Information: 6 SWS, 10 ECTS Credits (IN2106 IN4142)
Term: 2015WiSe
Abstract:  


Edit | Attach | Refresh | Diffs | More | Revision r1.42 - 17 Feb 2016 - 14:41 - ShadiAlbarqouni

Lehrstuhl für Computer Aided Medical Procedures & Augmented Reality    rss.gif