Multi-Instance Multi-Stage Deep Learning for Medical Image Recognition
Mastrodicasa, Simon
Promotor(s) : Phillips, Christophe
Date of defense : 25-Jun-2018/26-Jun-2018 • Permalink : http://hdl.handle.net/2268.2/4673
Details
Title : | Multi-Instance Multi-Stage Deep Learning for Medical Image Recognition |
Translated title : | [fr] Deep Learning Multi-Cas Multi-Etapes pour la Reconnaissance en Imagerie Medicale |
Author : | Mastrodicasa, Simon |
Date of defense : | 25-Jun-2018/26-Jun-2018 |
Advisor(s) : | Phillips, Christophe |
Committee's member(s) : | Zhang, Gary
Geurts, Pierre Louppe, Gilles |
Language : | English |
Number of pages : | 65 |
Keywords : | [en] Machine learning [en] MRI [en] Deep Learning [en] Multi-Instance |
Discipline(s) : | Engineering, computing & technology > Computer science |
Research unit : | University College of London |
Target public : | Researchers |
Complementary URL : | https: //github.com/Mastrodicasa/Master_Thesis.git |
Institution(s) : | Université de Liège, Liège, Belgique |
Degree: | Master en ingénieur civil biomédical, à finalité spécialisée |
Faculty: | Master thesis of the Faculté des Sciences appliquées |
Abstract
[en] Given a 2D transversal slice, identify which body section it belongs to, which is a
classification problem.
The twist is, instead of having a database with segmented body parts (example:
in this image, the heart is placed there, which is the discriminative feature of the
"cardiac" body section), image level label (example:this image belongs to the "car-
diac" body section) are used to train the algorithm, reducing greatly the annotation
time done by a specialist.
File(s)
Document(s)
Multi_Stage_Medical_Imaging_Recognition_Mastrodicasa_Simon.pdf
Description: -
Size: 1.32 MB
Format: Adobe PDF
Description: -
Size: 1.32 MB
Format: Adobe PDF
Cite this master thesis
All documents available on MatheO are protected by copyright and subject to the usual rules for fair use.
The University of Liège does not guarantee the scientific quality of these students' works or the accuracy of all the information they contain.
The University of Liège does not guarantee the scientific quality of these students' works or the accuracy of all the information they contain.