CD Modèles d'apprentissage automatique utilisant l'imagerie TEP aux acides aminés pour guider les décisions de routine clinique en neuro-oncologie // CD Guiding clinical routine practice decisions using amino acid PET imaging for neuro-oncology with artif
ABG-129298
ADUM-62998 |
Thesis topic | |
2025-03-08 | Public funding alone (i.e. government, region, European, international organization research grant) |
Université de Lorraine
VANDOEUVRE - France
CD Modèles d'apprentissage automatique utilisant l'imagerie TEP aux acides aminés pour guider les décisions de routine clinique en neuro-oncologie // CD Guiding clinical routine practice decisions using amino acid PET imaging for neuro-oncology with artif
Imagerie, Neuro-oncologie, Tomographie par émission de positions, Apprentissage auto-supervisé, Intelligence artificielle explicable
Imaging, Neuro-oncology, Positrong emission tomography, Self-supervised learning, Explainable artificial intelligence
Imaging, Neuro-oncology, Positrong emission tomography, Self-supervised learning, Explainable artificial intelligence
Topic description
À l'ère de la médecine personnalisée, il est essentiel de disposer d'outils non invasifs permettant de mieux caractériser les gliomes afin d'optimiser la prise de décision clinique. L'imagerie médicale joue un rôle clé dans ce domaine, notamment l'imagerie TEP aux acides aminés, recommandée en complément de l'IRM par les groupes d'experts internationaux. L'objectif de ce projet est de développer des modèles d'apprentissage automatique capables de capturer les relations complexes entre les voxels des images TEP aux acides aminés dans le gliome. Une attention particulière sera portée sur l'explicabilité afin de favoriser leur adoption en routine clinique. Le premier volet du projet consiste à construire un modèle d'apprentissage profond auto-supervisé permettant d'apprendre une représentation robuste des gliomes à partir d'images IRM et TEP multimodales. Cette représentation sera ensuite exploitée dans deux applications cliniques majeures : (i) la planification de la biopsie en identifiant une sous-région agressive d'un gliome afin d'améliorer la précision du prélèvement et (ii) le diagnostic différentiel entre radionécrose et progression tumorale avec une évaluation par les médecins en condition clinique. En résumé, ce projet vise à exploiter les avancées de l'IA et de l'imagerie médicale pour améliorer la caractérisation des gliomes et leur prise en charge clinique.
------------------------------------------------------------------------------------------------------------------------------------------------------------------------
------------------------------------------------------------------------------------------------------------------------------------------------------------------------
In the era of personalized medicine, it is essential to develop non-invasive tools to better characterize gliomas and optimize clinical decision-making. Medical imaging plays a key role in this field, particularly amino acid PET imaging, which is recommended alongside MRI by international expert groups. The objective of this project is to develop machine learning models capable of capturing the complex relationships between voxels in amino acid PET images of gliomas. A strong emphasis will be placed on explainability to facilitate their adoption in clinical practice. The first part of the project involves building a self-supervised deep learning model to learn a robust representation of gliomas from multimodal MRI and PET images. This representation will then be used for two major clinical applications: (i) biopsy planning by identifying an aggressive subregion of a glioma to improve sampling precision and (ii) differential diagnosis between radionecrosis and tumor progression, with evaluation by physicians under clinical conditions. In summary, this project aims to leverage advances in AI and medical imaging to improve glioma characterization and clinical management.
------------------------------------------------------------------------------------------------------------------------------------------------------------------------
------------------------------------------------------------------------------------------------------------------------------------------------------------------------
Début de la thèse : 01/10/2025
------------------------------------------------------------------------------------------------------------------------------------------------------------------------
------------------------------------------------------------------------------------------------------------------------------------------------------------------------
In the era of personalized medicine, it is essential to develop non-invasive tools to better characterize gliomas and optimize clinical decision-making. Medical imaging plays a key role in this field, particularly amino acid PET imaging, which is recommended alongside MRI by international expert groups. The objective of this project is to develop machine learning models capable of capturing the complex relationships between voxels in amino acid PET images of gliomas. A strong emphasis will be placed on explainability to facilitate their adoption in clinical practice. The first part of the project involves building a self-supervised deep learning model to learn a robust representation of gliomas from multimodal MRI and PET images. This representation will then be used for two major clinical applications: (i) biopsy planning by identifying an aggressive subregion of a glioma to improve sampling precision and (ii) differential diagnosis between radionecrosis and tumor progression, with evaluation by physicians under clinical conditions. In summary, this project aims to leverage advances in AI and medical imaging to improve glioma characterization and clinical management.
------------------------------------------------------------------------------------------------------------------------------------------------------------------------
------------------------------------------------------------------------------------------------------------------------------------------------------------------------
Début de la thèse : 01/10/2025
Funding category
Public funding alone (i.e. government, region, European, international organization research grant)
Funding further details
Concours pour un contrat doctoral
Presentation of host institution and host laboratory
Université de Lorraine
Institution awarding doctoral degree
Université de Lorraine
Graduate school
266 BioSE - Biologie Santé Environnement
Candidate's profile
Nous recherchons un(e) chercheur(se) passionné(e) par le traitement d'images et l'apprentissage automatique pour rejoindre notreéquipe. Vous travaillerez sur des projets de pointe pour développer des solutions innovantes à des problèmes complexes.
Compétences requises :
Compétences techniques :
• Connaissances avancées en traitement d'images (acquisition, filtrage, segmentation, reconnaissance d'objets, etc.)
• Connaissances avancées en apprentissage automatique (classification, régression, clustering, deep learning, etc.)
• Maitrise de langages de programmation (Python, ....)
• Expérience avec des librairies de ML (scikit-learn, pytorch, tensorflow, etc.)
• Capacité à développer et à déployer des modèles d'apprentissage automatique
• Aptitude à travailler avec des environnements de calcul haute performance (GPU, etc.)
Compétences comportementales :
• Indépendant et capable de travailler de manière autonome
• Capacité à travailler en équipe et à collaborer avec des experts d'autres domaines
• Rigoureux et soucieux du détail
• Excellent esprit d'analyse et de résolution de problèmes
• Capacité à communiquer clairement et efficacement
Formations / diplômes :
• Master en informatique, mathématiques, ingénierie ou domaine connexe
We are looking for a researcher with a passion for image processing and machine learning to join our team. You will be working oncutting-edge projects to develop innovative solutions to complex problems. Skills required: Technical skills : - Advanced knowledge of image processing (acquisition, filtering, segmentation, object recognition, etc.) - Advanced knowledge of machine learning (classification, regression, clustering, deep learning, etc.) - Proficiency in programming languages (Python, C++, etc.) - Experience with ML libraries (scikit-learn, pytorch, tensorflow, ...) - Ability to develop and deploy machine learning models - Ability to work with high-performance computing environments (GPU, etc.) Behavioural skills : - Independent and able to work autonomously - Ability to work in a team and collaborate with experts from other fields - Thorough and detail-oriented - Excellent analytical and problem-solving skills - Ability to communicate clearly and effectively Education / qualifications : - Master's degree in computer science, mathematics, engineering or related field
We are looking for a researcher with a passion for image processing and machine learning to join our team. You will be working oncutting-edge projects to develop innovative solutions to complex problems. Skills required: Technical skills : - Advanced knowledge of image processing (acquisition, filtering, segmentation, object recognition, etc.) - Advanced knowledge of machine learning (classification, regression, clustering, deep learning, etc.) - Proficiency in programming languages (Python, C++, etc.) - Experience with ML libraries (scikit-learn, pytorch, tensorflow, ...) - Ability to develop and deploy machine learning models - Ability to work with high-performance computing environments (GPU, etc.) Behavioural skills : - Independent and able to work autonomously - Ability to work in a team and collaborate with experts from other fields - Thorough and detail-oriented - Excellent analytical and problem-solving skills - Ability to communicate clearly and effectively Education / qualifications : - Master's degree in computer science, mathematics, engineering or related field
2025-06-01
Apply
Close
Vous avez déjà un compte ?
Nouvel utilisateur ?
More information about ABG?
Get ABG’s monthly newsletters including news, job offers, grants & fellowships and a selection of relevant events…
Discover our members
PhDOOC
Tecknowmetrix
TotalEnergies
Généthon
ANRT
Ifremer
CESI
CASDEN
ONERA - The French Aerospace Lab
ADEME
SUEZ
ASNR - Autorité de sûreté nucléaire et de radioprotection - Siège
Laboratoire National de Métrologie et d'Essais - LNE
MabDesign
MabDesign
Groupe AFNOR - Association française de normalisation
Aérocentre, Pôle d'excellence régional
Institut Sup'biotech de Paris
Nokia Bell Labs France
-
JobPermanentRef. ABG129192Association Bernard Gregory (ABG)Paris (3ème) - Ile-de-France - France
Business Developer (F/H)
Open to all scientific expertisesAny -
JobPermanentRef. ABG128675Mini Green PowerHyères - Provence-Alpes-Côte d'Azur - France
Ingénieur (e) / Chercheur (se) R&D – Innovation en énergie décarbonée
Process engineeringJunior -
JobPermanentRef. ABG129119Aguaro- Pays de la Loire - France
Docteur·e Data Environnement
Ecology, environmentJunior