Medical Image Processing with Ai, Emd, and Entropy - Angers, France - UNIVERSITE D'ANGERS

UNIVERSITE D'ANGERS

Entreprise vérifiée

Angers, France

il y a 2 semaines

Posté par:

Sophie Dupont

beBee Recruiter

Description

Medical image processing with AI, EMD, and entropy measures:

Réf
ABG-122976
Sujet de Thèse 17/04/2024
Contrat doctoral
UNIVERSITE D'ANGERS
Lieu de travail
Angers
Pays de la Loire
France
Intitulé du sujet
Medical image processing with AI, EMD, and entropy measures
Champs scientifiques
Sciences de l'ingénieur
Mots clés entropy, empirical mode decomposition, machine learning, deep learning
Description du sujet:

Entropy-based measures developed in the 1990s to quantify the irregularity of time series have evolved to become a powerful tool for analysing medical-based data.

Notable advancements in this research include the development of sample entropy (Richman and Moorman, 2000), permutation entropy (Bandt and Pompe, 2002), fuzzy entropy (Chen et al., 2007), distribution entropy (Li et al., 2015), dispersion entropy (Rostaghi et al., Some of these measures have also been extended to the bidimensional case or even tridimensional case to analyze the texture of images or volumes (Humeau-Heurtier, 2019, All these measures have their advantages (e.g., low computational time) and drawbacks (e.g., lead to undefined values for short data).

Among the existing entropy measures, permutation entropy has the advantage of being computationally fast and simple. It is based on ordinal patterns (Amigo et al.

, 2023):

each sequence of _n_ successive values of a given data is mapped into a permutation pattern that represents the rank order of the samples within the sequence.

Then, the Shannon entropy is estimated using the relative frequency (probability of occurrence) of each permutation pattern (Bandt and Pompe, Ordinal patterns have an implicit connection with many signal processing methods such as the empirical mode decomposition (EMD) (Huang et al., 1998).

The PhD thesis work will be conducted in Angers (France), in SETU and ATU (Ireland). The PhD student will spend his first and third years in Angers and his second year in Ireland.

Nature du financement:

Contrat doctoral

Précisions sur le financement:

Présentation établissement et labo d'accueil:

UNIVERSITE D'ANGERS

The work will be performed in LARIS (University of Angers, France), Atlantic Technological University (Donegal, Ireland), and South East Technological University (Carlow, Ireland).

Site web: