Soutenance de thèse de Jean-Rassaire FOUEFACK : " Towards a framework for multi class statistical modelling of shape, intensity, and kinematics in medical images "

Jeudi 18.03.2021
Horaires :
De 16:00 à 18:00

Adresse :

En visio-conférence totale (dispositions exceptionnelles durant la crise sanitaire liée à la Covid19)
le lien public Youtube est le suivant : https://youtu.be/gXWYxY-xpnk

Jean-Rassaire FOUEFACK doctorant au département ITI, et appartenant au laboratoire LATIM, présentera ses travaux de thèse intitulés :

" Towards a framework for multi class statistical modelling of shape, intensity, and kinematics in medical images "

Avis de soutenance

En savoir +

Vous trouverez, ci-dessous, un résumé de sa thèse :
This thesis focuses on the development of statistical learning-based models for human joints from medical images. The main contribution is a new mathematical formulation to model musculoskeletal systems by developing a unified computational latent space that embeds shape, kinematics and intensity features from given observations. This new space provides a continuous model capable of generating instances leveraging learnt feature correlations. A fitting method to apply models developed using this framework to unseen data is also proposed. The complete modeling and prediction framework is validated using bespoke synthetic data, showing that the framework faithfully encapsulates any prescribed morpho-functional relationships between objects, as well as their internal structural information. Finally, the framework is applied to the analysis of shoulder and hip joints from CT. The clinical interest is that the feature correlations learned by the model improve premorbid shape prediction and joint motion estimation accuracy, from two- and three-dimensional medical.


Mots-clés : Shape, pose and intensity latent space, shoulder and hip human joint, medical image analysis, pattern recognition

Publié le 15.03.2021
 
 
 
 
 
1
 
2
 
3
 
4
 
5
 
6
 
7
 
8
 
9
 
10
 
11
 
12
 
13
 
14
 
15
 
16
 
17
 
18
 
19
 
20
 
21
 
22
 
23
 
24
 
25
 
26
 
27
 
28
 
29
 
30
 
31