{"id":16,"date":"2016-01-19T11:25:54","date_gmt":"2016-01-19T10:25:54","guid":{"rendered":"http:\/\/pagesperso.litislab.fr\/suruan\/?page_id=16"},"modified":"2024-06-24T08:22:32","modified_gmt":"2024-06-24T07:22:32","slug":"recherche","status":"publish","type":"page","link":"https:\/\/pagesperso.litislab.fr\/suruan\/recherche\/","title":{"rendered":"Research Areas"},"content":{"rendered":"

(<\/em><\/strong>Th\u00e8mes de recherche d\u00e9velopp\u00e9s) <\/strong><\/h2>\n

a) Image segmentation and Pattern Recognition <\/strong>(La segmentation d’images et la reconnaissance de formes )<\/h3>\n

Segmentation :\u00a0 Machine learning & Deep learning<\/span>, Markov random fields,\u00a0 Active contours (level sets), \u00a0statistical shape models, non-rigid registration, Graph based methods.<\/p>\n

Pattern Recognition: A priori knowledge modeling, Statistical shape models, Machine learning & Deep learning,\u00a0 Feature selection.<\/span><\/p>\n

(Segmentation : apprentissage automatique & apprentissage profond,\u00a0 champs al\u00e9atoires de Markov, contours actifs (level sets), mod\u00e8les de forme statistiques, recalage non rigide, m\u00e9thodes bas\u00e9es sur les graphes.)<\/p>\n

(Reconnaissance de formes :\u00a0Mod\u00e9lisation des connaissances a priori. Mod\u00e8les de formes statistiques \u00e0 partir d’un\u00a0 ensemble d’apprentissage sur les formes 3D. Classification par apprentissage automatique & apprentissage profond. S\u00e9lection des caract\u00e9ristiques.)<\/p>\n

b) Information\u00a0fusion<\/strong> (La fusion des informations)<\/em><\/h3>\n

Fusion based on \u00a0deep learning,\u00a0 fuzzy theory, Bayes’ <\/span>statistical theory<\/span>, <\/em>and<\/span> the theory of belief functions.<\/span><\/p>\n

(Apprentissage profond, <\/span>Fusion floue des informations mod\u00e9lis\u00e9es par des ensembles flous, Fusion bas\u00e9e sur la th\u00e9orie statistique de Bayes<\/span><\/span>, Fusion bas\u00e9e sur la th\u00e9orie des fonctions de croyance.)<\/p>\n

c) Applications <\/strong>(Applications)<\/h3>\n

– Medical imaging: brain MRI,\u00a0Prostate MRI, MR spectroscopy (MRS), thoracic PET-CT<\/span><\/p>\n

– Segmentation and classification of \u00a0medical images for diagnostic\u00a0and\u00a0therapy<\/span><\/p>\n

– Longitudinal follow-up and selection of predictive characteristics for cancer therapy<\/span><\/p>\n

– Radiomics: patient outcome prediction<\/span><\/p>\n

(Imageries\u00a0m\u00e9dicales: IRM c\u00e9r\u00e9brale, IRM prostatique, Spectroscopie par RM (SRM) ,\u00a0PET-TDM\u00a0thoracique; Segmentation et classification des images m\u00e9dicales; Suivi longitudinal et s\u00e9lection des caract\u00e9ristiques pr\u00e9dictives pour le\u00a0traitement\u00a0du cancer (radiomiques)).<\/p>\n